DYING TOO YOUNG Addressing THE POOR HEALTH STATUS of Russia's economically active adult population--its human capital--challenges sustainable economic growth Premature and social development. President Vladimir Putin, in his annual address DYING TOO YOUNG to the Federal Assembly of the Russian Federation in May 2004 and Mortality more recently in the State-of-the-Nation Address on April 25, 2005, criticized the failure of health care reform to produce significant results, and as evidenced by Russia's lagging behind many countries in key health Ill indicators. Life expectancy in Russia at 66 years, he noted, is 12 years Health Addressing Premature Mortality and less than it is in the United States, 8 years less than in Poland, and 5 years less than in China--a situation President Putin attributed to Due Ill Health Due to Non-Communicable Diseases "the high death rate in the working-age population." to Non-Communicable and Injuries in the Russian Federation Dying Too Young aims to heighten understanding of the nature and characteristics of non communicable diseases and injuries as the leading killers in the Russian Federation, its associated risk factors, and their social and economic implications. The study outlines specific options and Diseases offers recommendations for addressing this problem, and projects the health and economic gains that could result from a comprehensive and program of action. Improving adult health would contribute to improved Injuries health status of the population, quality of life, labor productivity and sustainable economic growth in the country. in the Russian Federation Europe and Central Asia Human Development Department The World Bank DYING TOO YOUNG Addressing Premature Mortality and Ill Health Due to Non-Communicable Diseases and Injuries in the Russian Federation Europe and Central Asia Region Human Development Department Russia Country Management Unit The World Bank TABLE OF CONTENTS Preface ............................................................................................................................................ ix Acknowledgments .......................................................................................................................... xi Abbreviations .................................................................................................................................. xii EXECUTIVE SUMMARY .................................................................................................................. xiii The Demographic Decline and the Health Deficit .......................................................................... xiii Determinants of the Demographic Decline and the Health Deficit ................................................ xv What Are The Consequences Of The Demographic Decline And The Health Deficit?.................. xviii What Strategies and Interventions Prevent Non-Communicable Diseases and Injuries? .................. xx Is Russia Doing Enough to Confront Non-Communicable Diseases and Injuries? ........................ xxi What Additional Actions Can Russia Take? .................................................................................... xxi What Health Improvements Could Result From an Intensified Program of Action? ...................... xxii What Are the Potential Economic Benefits of Mortality Reduction In Russia?................................ xxii Conclusion ...................................................................................................................................... xxiii Chapter 1. The Demographic Decline and the Health Deficit................................................ 1 A Framework for Understanding the Demographic and Health Conditions in Russia .................. 1 The Shrinking of the Russian Population ........................................................................................ 2 Declining Fertility Rates .................................................................................................................. 4 High Mortality Rates ........................................................................................................................ 7 Mortality among Population Groups................................................................................................ 7 Life Expectancy and Economic Development ................................................................................ 10 Impact of Life Expectancy on Human Development ...................................................................... 13 Population Migration........................................................................................................................ 13 The Aging of the Population............................................................................................................ 14 Total Dependency Ratios ................................................................................................................ 15 What Are the Consequences of the Demographic Decline and the Health Deficit?...................... 15 Chapter 2. The Burden of Ill Health and Disability .................................................................. 19 Healthy Life Expectancy.................................................................................................................. 19 Comparison of Disability Levels...................................................................................................... 20 Policy Implications .......................................................................................................................... 21 Chapter 3. Determinants of the Demographic Decline and the Health Deficit ................ 25 Deaths and DALYs Lost Attributable to Leading Causes .............................................................. 25 The Leading Individual Causes of Death........................................................................................ 27 Policy Implications .......................................................................................................................... 32 Chapter 4. What Are the Major Risk Factors for NCDs and Injuries? ................................ 35 Major Individual Health Risk Factors .............................................................................................. 35 Interaction among Risk Factors ...................................................................................................... 44 Policy Implications .......................................................................................................................... 45 iii Chapter 5. The Impact of NCDs and Injuries on Health Care Costs.................................... 49 Study Method .................................................................................................................................. 49 Results ............................................................................................................................................ 50 Policy Implications .......................................................................................................................... 51 Chapter 6. How Do NCDs and Injuries Affect the Russian Economy? .............................. 55 Conceptual Framework .................................................................................................................. 55 What Effect has Adult Ill Health had on the Economic Outcome of the Population? .................... 56 Policy Implications .......................................................................................................................... 61 Chapter 7. What Strategies and Interventions Prevent NCDs and Injuries? .................... 65 International Evidence on Effective Interventions to Control Individual Risk-Factors Associated with the Onset of NCDs and Injuries............................................................................ 67 The Government's Role in Disease Prevention .............................................................................. 70 Role of Private Employers .............................................................................................................. 70 Chapter 8. How Is Russia Responding to NCDs and Injuries? ............................................ 73 Recent Organizational Reforms ...................................................................................................... 73 The Legal Framework for Action Against NCDs and Injuries ........................................................ 74 Institutional Weaknesses in Preventing and Controlling NCDs ...................................................... 76 What Are the Main Problems? ........................................................................................................ 76 Some Good Practices in Russia...................................................................................................... 77 Steps to Advance NCD Prevention and Control Efforts.................................................................. 80 Policy Implications .......................................................................................................................... 80 Chapter 9. What Additional Actions Can Russia Take?.......................................................... 83 The Rationale for a Comprehensive Program ................................................................................ 83 Federal Level Policies and Strategies ............................................................................................ 83 Priority Subprograms in Regions .................................................................................................... 84 Improved Road Safety and Emergency Services .......................................................................... 86 Chapter 10. What Health Improvements Could Result from an Intensified Program of Action? .................................................................................... 91 Mortality Reduction.......................................................................................................................... 91 The Impact on Achieving the Millennium Development Goals (MDGs) ........................................ 92 Chapter 11. What Are the Potential Economic Benefits of Mortality Reduction in Russia? .............................................................................................. 95 Benefits of Reducing NCDs and Injury Mortality: A Simple Static Calculation .............................. 95 Static Welfare Effects ...................................................................................................................... 97 Dynamic Effects: The Impact of Adult Health on Economic Growth .............................................. 99 Additional Estimates of the Macroeconomic Consequences of NCDs and Injuries in Russia ...... 100 Conclusion ...................................................................................................................................... 100 Epilogue .......................................................................................................................................... 103 Appendix A. .................................................................................................................................... 105 Appendix B. Data Sources, Methodologies, and Detailed Results for Chapters 6 and 11 .......... 106 Bibliography .................................................................................................................................... 131 Map of Russian Federation ............................................................................................................ 147 iv LIST OF TABLES Table 1.1 Life Expectancy and Adult Mortality in Selected Countries, 2000-2001 .................... 9 Table 1.2 Life Expectancy at Ages 60, 65, and 80 in Russia and Selected Countries, by Gender, 2000-2005 and 2025-2030........................................................................ 12 Table 1.3 Regional Variation in Mortality Rates for the Economically Active Population, 2000 .......................................................................................................... 12 Table 1.4 Regional Variation in Average Life Expectancy at Birth in Russia, 1999-2001 .......... 12 Table 1.5 Human Development Index for Selected European Countries, 2002 ........................ 13 Table 2.1 Healthy Life Expectancy (HALE) at Birth in Russia and Four European Countries, 2002 .......................................................................................... 19 Table 2.2 Life Expectancy and Healthy Life Expectancy at Ages 20, 40, and 65 in Russia, Eastern Europe, and Western Europe, 2000 .............................................. 20 Table 3.1 Deaths and DALYs Lost Attributable to the 10 Leading Diseases and Injuries, 2002 ........................................................................................................ 25 Table 3.2 Cause-Specific Adult Mortality Rates, Age 15-64 ...................................................... 26 Table 3.3 Regional Variation in Mortality Rate by Cause of Death, 2000.................................... 26 Table 3.4 Estimated Mortality from NCDs and Population Reduction, 2000-2030...................... 27 Table 3.5 Mortality Rates Due to Motor Vehicle Traffic Injuries in Selected Countries, 1998-2002 .................................................................................. 31 Table 4.1 Deaths and DALYs Lost Attributable to 10 Leading Risk Factors, 2002 .................... 35 Table 4.2 Percentage of 15-Year-Olds and Adults Who Smoke in Selected Countries, 2003 .... 38 Table 4.3 The Link between Modern Lifestyles and Low Physical Activity ................................ 41 Table 4.4 Prevalence of Diabetes Cases in G-8 Countries, 2000 .............................................. 42 Table 4.5 Occupational Accidents and Work-Related Disease, Russian Federation, 2001 ...... 45 Table 5.1 Estimated Government and Nongovernment Health Care Expenditures (and Percentage of GDP) to Treat Diseases, by Diagnostic Group, 2003.................. 51 Table 5.2 Estimated Government and Nongovernment Health Care Expenditures to Treat Diseases of the Circulatory System, 2003...................................................... 52 Table 6.1 Estimated Cost of Absenteeism Due to Illness in the Russian Federation ................ 57 Table 8.1 Active Federal Targeted Programs in the Russian Federation .................................... 75 Table 10.1 Estimated Improvements in Life Expectancy if Mortality Rates Are Reduced by 20 and 40 Percent, Russian Males, 2000 .............................................. 92 Table 10.2 Estimated Improvements in Life Expectancy if Mortality Rates Are Reduced by 20 and 40 Percent, Russian Females, 2000 .......................................... 92 Table 11.1 Economic Benefit Estimation for Scenario 2................................................................ 96 Table 11.2 Economic benefit estimation for Scenario 3 ................................................................ 99 Table 11.3 Welfare benefits of Scenario 3 and 2 .......................................................................... 98 v LIST OF FIGURES Figure 1.1 Estimated Population Growth Rates, Russia, 1980-85 to 2020-25 ............................ 3 Figure 1.2 Estimates of Total Russian Population, 2000-2025 .................................................... 4 Figure 1.3 Total Fertility Rates in Selected Developed Countries, 1950-2025............................ 4 Figure 1.4 Standardized Mortality Rate from All Causes for People Ages 0-64, 1986-2001, Selected Countries.................................................................................. 6 Figure 1.5 Infant Deaths per 1000 Live Births ............................................................................ 6 Figure 1.6 Trends in Under-5 Mortality in Selected Countries, 1990-2002 ................................ 7 Figure 1.7 Trends in Maternal Mortality in Selected Countries, 1991-99 .................................... 8 Figure 1.8 Life Expectancy at Birth, Russia and the European Union, 1970-2000 .................... 8 Figure 1.9 Mortality among the Working-Age Population in Russia Drives Fluctuations in Life Expectancy.................................................................................. 9 Figure 1.10 Male Adult Mortality and Gross National Income per Capita in Selected Countries, 2000 ...................................................................................... 10 Figure 1.11 Is Economic Growth Sustainable in Russia without Good Health?............................ 10 Figure 1.12 Life Expectancy at Birth, Males and Females, 1970-2002 ........................................ 11 Figure 1.13 Net Migration, 1985-2002 .......................................................................................... 13 Figure 1.14 The Population Structure, 2005 and 2020.................................................................. 14 Figure 1.15 Aging Index: Number of Persons Aged 60 Years and Over Compared to That of 100 People Aged 0-14 Years, 2005-25...................................................... 15 Figure 2.1 Russia and Sweden Survival without Disability at Different Ages.............................. 20 Figure 2.2 Russia and Sweden: Total Survival and Survival without Disability to Age 70-74 in Cohorts without Disability at Age 45-49 .......................................... 21 Figure 3.1 Cardiovascular Disease Mortality Rates in Russia as a Percentage of Swedish Rates........................................................................................................ 28 Figure 3.2 Trends in Standardized Death Rates from Diseases of the Circulatory System in Russia and EU-15 Countries .................................................. 28 Figure 3.3 Trends in Standardized Death Rates from Cerebrovascular Diseases in Russia and EU-15 Countries, All Ages .................................................................. 29 Figure 3.4 Trends in Standardized Death Rates for All Malignant Neoplasms (Cancer) Combined for Ages 0-64, in Russia and EU-15 Countries, 1980-2002 .................... 29 Figure 3.5 Trends in Standardized Death Rates for Cervical Cancer for Women Aged 0-64 in Russia and EU-15 Countries .................................................. 30 Figure 3.6 Trends in Standardized Death Rates for Trachea, Bronchus, and Lung Cancer in Russia and EU-15 Countries, Males, All Ages, 1980-2002 ............ 30 Figure 3.7 Trends in Standardized Death Rates for Suicide and Self-Inflicted Injuries in Russia and EU-15 Countries, Ages 0-64, 1980-2002 ............................................ 31 Figure 3.8 Injury Mortality Rates in Russia as a Percentage of Swedish Rates ........................ 33 Figure 4.1 Per Capita Alcohol Consumption in Russia, US, and UK, 1999................................ 36 vi Figure 4.2 Smoking Prevalence among Adults in Selected G-8 Countries ................................ 37 Figure 4.3 Russians' Self-Reported Health Status, by Consumption Quintile, 2003 .................. 43 Figure 4.4 Russians' Self-Reported Use of Hard Alcohol, by Consumption Quintile, 2003 ...... 43 Figure 4.5 Self-Reported Reasons for Not Seeking Health Care, by Consumption Quintile, 2003 ...................................................................................... 43 Figure 4.6 Self-Reported Reasons for Not Following Prescribed Treatment, by Consumption Quintile, 2003 ...................................................................................... 44 Figure 5.1 Health Care Expenditures to Treat the Four Most Costly Disease Categories, 2003 .......................................................................................... 50 Figure 6.1 From Health to Wealth (and Back) ............................................................................ 55 Figure 6.2 Annual Average Days of Absence Due to Illness, per Employee, Russia and EU-15 Countries ...................................................................................... 56 Figure 6.3 Probability to remain in the workforce with and without chronic illness in Russia, by age - based on Cox regression model ................................................ 59 Figure 6.4 Average Predicted Probability to Retire in the Subsequent Period for Hypothetical Male at Varying Income Levels in Russia (Based on Panel Logit Model).................................................................................... 60 Figure 7.1 Distribution of Risk Factors in a Population .............................................................. 65 Figure 10.1 Decline in Age-Adjusted Mortality from CHD in N. Karelia: 1969-71 to 1993-95 .................................................................................................... 91 Figure 10.2 Estimated Gains in Healthy Life Expectancy with Elimination of the 20 Leading Risk Factors by Subregion...................................................................... 93 Figure 11.1 Three Scenarios for Russian Adult Mortality Rates due to Noncommunicable Diseases and Injuries, Age 15-64, 2002-25 .............................. 95 Figure 11.2 GDP per Capita Forecasts in the Three Scenarios.................................................... 99 LIST OF BOXES Box 1.1 The Factors Behind the Epidemiological Transition ...................................................... 2 Box 1.2 The Global Context for Demographic Trends ................................................................ 3 Box 1.3 The Accuracy and Reliability of Demographic Projections............................................ 5 Box 7.1 An International Good Practice: North Karelia .............................................................. 66 Box 8.1 Chuvash Republic: Innovative Approaches to Healthy Living ...................................... 79 vii PREFACE I am deeply convinced that the success of our policy in all spheres of life is closely linked to the solution of our most acute demographic problems. We cannot reconcile ourselves to the fact that the life expectancy of Russian women is nearly 10 years and of men nearly 16 years shorter than in Western Europe. Many of the current mortality factors can be remedied, and without particular expense. In Russia near- ly 100 people a day die in road accidents. The reasons are well known. And we should implement a whole range of measures to overcome this dreadful situation. I would like to dwell on another subject which is difficult for our society--the con- sequences of alcoholism and drug addiction. Every year in Russia, about 40,000 people die from alcohol poisoning alone, caused first of all by alcohol substitutes. Mainly they are young men, breadwinners. However, this problem cannot be resolved through prohibition. Our work must result in the young generation recog- nizing the need for a healthy lifestyle and physical exercise. Each young person must realize that a healthy lifestyle means success, his or her personal success. PRESIDENT VLADIMIR PUTIN State-of-the-Nation Address Federal Assembly of the Russian Federation, The Kremlin, April 25, 2005 R ussia's Medium-Term Program for Social and diovascular disease (hypertension, high cholesterol, dia- Economic Development 2005-2008 emphasizes betes, obesity, and smoking); (b) alcohol abuse; (c) traf- integrating its economy into the world economy, fic accidents; and (d) HIV/AIDS. Reducing prime age raising competitiveness, and developing human capital. mortality and controlling risk factors cannot be achieved The poor health status of Russia's economically active through the acute medical care system alone--it will adult population--its human capital--challenges sus- require a comprehensive approach with multisectoral tainable growth. President Vladimir Putin, in his annual policies and programs. Russia's Ministry of Health and address to the federal assembly in May 2004 and more Social Development (MOHSD) must take the lead, but recently in the State-of-the-Nation Address on April 25, success will be achieved only with the active collabora- 2005, criticized the failure of health care reform to pro- tion of the Ministries of Economy, Finance, Education, duce significant results, as evidenced by Russia's lagging Transportation, and other ministries, and the active par- behind many countries in key health indicators. Life ticipation of civil society, including private enterprises, expectancy in Russia, he noted, is 12 years less than in nongovernmental organizations, communities, families, the United States, 8 years less than in Poland, and 5 and individuals. years less than in China--a situation President Putin attributed to "the high death rate in the working-age Over the last decade, the country has shown its willing- population." ness to change the health system to achieve better per- formance and outcomes, undertaking a review of its The factors that put human capital development at risk health financing and the relationships among citizens, in Russia are well documented: (a) risk factors for car- service providers, and financiers and initiating health ix care reforms in many regions. The country is aligning ments were also prepared on the demographic profile of with international guidelines in some of its approaches the country, the relative importance of NCDs and to infectious disease control, specifically for tuberculosis injuries on health care costs, and the economic conse- and HIV/AIDS. There are signs of increasing commit- quences of NCDs and injuries. ment to control non-communicable diseases (NCDs) and injuries, the leading causes of death, illness and dis- The primary audience is external: policy makers, senior ability in the country. These developments provide a analysts, program managers and their advisers in the basis for cautious optimism and a rationale for decisive Russian Government, private sector officials, and inter- action in support of a health system that can achieve national organizations and nongovernmental organiza- better outcomes. tions working on health issues in Russia. The report is intended to be a contribution to the debate on appro- This report assesses the factors associated with the onset priate choices and instruments for health development of NCDs and injuries, as well as the demographic, in Russia. The issues, options, and recommendations are financial and economic consequences of these condi- intended for discussion and do not constitute a position tions, summarizes relevant evidence and emerging les- paper by the World Bank. A secondary audience is inter- sons from international experience, proposes a compre- nal, particularly managers and staff of the World Bank hensive program for addressing this problem, and proj- addressing health, human development, and poverty ects the health and economic gains that could result reduction as well as managers and staff of sectors that from such a program. It conforms to the World Bank's affect health outcomes, such as the economic and finan- Country Assistance Strategy in the Russian Federation cial, education, and transport sectors. for 2003-2005, which recommends a focus on economic competitiveness and on mitigating social risks in It should be clear, however, that the understanding of Russia's regions the Russian situation cannot be based entirely on cur- rent information from Russia and research results from The report draws heavily on data from the World western countries. This means that the understanding of Health Organization, the United Nations, and the the Russian situation outlined in this report is indeed World Bank, as well as from scientific literature, the provisional and that further research is essential to be Russia Longitudinal Monitoring Survey (RLMS), and able to get more definitive estimates of its causal factors. the 2003 National Survey of Household Welfare and But in a world where there is a need to act without per- Program Participation (NOBUS). Background assess- fect data it should be seen as a start. x ACKNOWLEDGMENTS This report was prepared by Patricio V. Marquez, Lead The report benefited from comments and advice from Health Specialist, Europe and Central Asia Region World Bank managers and technical specialists: (ECA), The World Bank, with contributions from: Kristalina I. Georgieva, Charles Griffin, Armin Fidler, Enis Baris, Edmundo Murrugarra, Christoph Kurowski, Marc Suhrcke and Lorenzo Rocco, World Health Cem Mete, John Litwack, Maria Gracheva, Joy de Organization (WHO) European Office for Investment Beyer, Ben Eijbergen, Gillette Conner, Dorothee for Health and Development, Venice; Martin McKee Eckertz, Maris Jesse, Merrell Tuck, and Oscar Echeverri, and Dave Leon, London School of Hygiene and as well as from Mikko Vienonen, Country Director, Tropical Medicine; Jill Farrington and Anatoliy Nosikov, WHO Office, Moscow, Robert Beaglehole, WHO WHO Office for Europe, Copenhagen; Tiina Geneva, Brook Horowitz, The Prince of Wales Laatikainen, National Public Health Institute of International Business Leaders Forum, London, Carlos Finland; Kirill Danishevski, Moscow Medical Academy; Santos-Burgoa, Mexican Health Secretariat, and Derek and from members of the World Bank Russia Health Yach, Yale University School of Medicine. Team: Anne Margreth Bakilana, Tatyana Loginova, Willy De Geyndt, Edward Frid, John C. Beth Goodrich edited the report, and Anahit Poghosyan Langenbrunner, Ginny Hsieh and Alberto Gónima. was responsible for its production. Tressa Alfred also assisted in the preparation of the manuscript. The authorities of the Russian Ministry of Health and Social Development provided overall guidance and The report's findings and recommendations were pre- advice under the leadership of Deputy Minister V. sented at a workshop organized by the MOHSD in Starodubov and Department Director, Dr.R. Khalfin. Moscow on June 14-15, 2005. A summary of the Professor R. G. Oganov of the Russian State Research report was disseminated at the European Health Forum Center of Preventive Medicine and his team provided held in Gastein, Austria, on October 5-8, 2005, and advice, information, and assessments conducted by posted at the World Bank ECA Region and European Russian researchers. Observatory web sites: http://web.worldbank.org/WBSITE/EXTERNAL/COU Additional contributions were made by the team of the NTRIES/ECAEXT/0,,contentMDK:20661159~pagePK North Karelia Project and the National Public Health :146736~piPK:146830~theSitePK:258599,00.html Institute of Finland, under the leadership of Professor Pekka Puska, who advised on the lessons learned from www.euro.who.int/observatory/ctryinfo/CtryInfoRes?lan the Finnish experience during a visit to Joensuu and guage=English&Country=RUS Helsinki in April 2005. xi ABBREVIATIONS AIDS Acquired immunodeficiency syndrome MMA Moscow Medical Academy ALOS Average length of stay MOH Russian Ministry of Health (until 2004) ASDR Age-specific death rate MOHSD Russian Ministry of Health and Social BAC Blood alcohol concentration Development (after 2004) BMI Body mass index NCDs Non-communicable diseases CAS Country Assistance Strategy NGO Non-governmental organization CDT Carbohydrate-deficient transferring NOBUS National Survey of Household Welfare and Program Participation CEE Central and Eastern Europe OECD Organization for Economic Co-operation CHD Coronary heart disease and Development CINDI Countrywide Integrated PPP Purchasing power parity Non-communicable Disease Intervention Program RLMS Russian Longitudinal Monitoring Survey CIS Commonwealth of Independent States SanEpid Russian State Committee on Sanitary and Epidemiological Inspection CVD Cardiovascular disease SRI Science and Research Institute DALY Disability-adjusted life year TB Tuberculosis DOTS Direct observed treatment, short course, for Tuberculosis TMA Tver Medical Academy EMEs Established Market Economies UGMADO Urals Medical Academy of Additional Education EU European Union U.K. United Kingdom G-8 Group of Eight Industrialized Countries (Canada, France, Germany, Italy, Japan, UN United Nations Russia, United Kingdom, United States) UNICEF United Nations Children's Fund GDP Gross domestic product UNAIDS Joint United Nations Program on AIDS GNI Gross national income UNDP United Nations Development Program HIV Human immunodeficiency virus U.S. United States IEC Information, education, and USAID U.S. Agency for International communication Development IFC International Finance Corporation VSL Value of a statistical life IMF International Monetary Fund WHO World Health Organization LTC Long-term care xii Executive Summary T he poor health status of Russia's economically unhealthy lifestyles and environments not only halted active adult population--its human capital--is progress in improving health status, but shortened life imperiling sustainable economic and social devel- expectancy, particularly among working-age males. This opment. What factors contribute to the excessive mor- situation defines a new pattern of the epidemiological tality, ill health, and disability in Russia, particularly transition that deviates from that experienced by a num- among working-age adults? What are the demographic, ber of western countries where age-specific NCDs rates social, and economic consequences of this phenome- declined and life expectancy grew. non? Most importantly, what can be done to reduce these burdens? The shrinking of the population Russia's population was 149 million in 1992; it declined This report shows that non-communicable diseases by 6 million as of 2003 to an estimated 143 million. If (NCDs) and injuries are the leading causes of death, ill- current trends persist, the Russian population is expect- ness, and disability in Russia, and assesses the factors ed to decline by over 30 percent over the next 50 years, associated with the onset of these conditions, the demo- as all measures of demographic processes show that graphic, financial, and economic consequences of Russia will undergo further dramatic changes in its pop- NCDs and injuries; summarizes relevant evidence and ulation dynamics. The average annual population emerging lessons from international experience, propos- growth during 1990-2003 was -0.3 percent, and contin- es a comprehensive program for addressing this prob- ued high mortality and declines in fertility are expected lem; and projects the health and economic gains that to lead to further negative population growth. Estimates could result from such a program. suggest that Russia's population today would be 17 mil- lion higher if its age-specific mortality rates had fol- THE DEMOGRAPHIC DECLINE lowed the patterns experienced by the European Union- AND THE HEALTH DEFICIT 15 countries (EU-15) since the mid-1960s. How do Russia's demographic decline and Low life expectancy due to high adult ill health burden compare internationally? mortality Russia is suffering a prolonged and debilitating demo- Analysts have described the country's mortality profile as graphic decline fueled by low fertility and high mortali- the most puzzling aspect of the Russian transition. Poor ty. Some observers assess the decline as "devastation" data quality has been ruled out as an explanatory vari- and note that it is unprecedented among industrialized able because demographic data are considered reliable: nations. The decrease in fertility and increase in mortali- Russia's registration of vital events is nearly complete ty since the 1960s were exacerbated in the 1990s by the and its coding of broad categories of death is reasonably turbulent political and economic transition and an accurate. ensuing decline in the population's well-being. In recent years, prosperity has increased due to rapid economic Russia is one of few countries where life expectancy is growth, but demographic and health trends remain wor- falling. However, its situation is distinguished from risome--even alarming. other countries, such as several in sub-Saharan Africa, where a generalized HIV/AIDS epidemic is driving loss- Russia's current situation clearly exemplifies a reversal of es in life expectancy. Both the current low level of life the epidemiological transition as the economic stress of expectancy and the recent declines were driven largely the past 15 years and a prolonged period of highly by increasing mortality among those of working age, xiii with a singular rise in mortality at young adult ages and Ill health and disability: Russian adults also experi- with the greatest contribution from cardiovascular dis- ence lower healthy life expectancy, which can be eases and injuries. expressed through the healthy life expectancy indicator, or HALE, a summary measure that captures the full Life expectancy in the former Soviet Union had nearly health experience of a population to include mortality, reached that of the United States by the early 1960s, but morbidity, and adjustments for severity of illnesses. In death rates increased significantly, particularly at adult other words, it expresses years of life spent in full health. ages, from 1965 to 1984. By 1980, the difference in life The healthy life expectancy of males and females is expectancy was nearly 8 years. After 1984 Russia experi- much lower in Russia (less than 60 years) than in the enced wide swings in life expectancy, while the United Kingdom, Germany, France, and Italy (more European Union experienced steady increases. Russia than 70 years). A healthy middle-aged cohort in Russia mortality declined during the Gorbachev anti-alcohol would have less than a third the chance of one in campaign (1985-87), but its effects dwindled after its Sweden to survive into old age without disability. cessation. Furthermore, as the country entered a severe crisis associated with the sudden transition to a market Gender disparities: Russian women live approximately economy, a dramatic deterioration ensued from 1990 to 14 years longer than Russian men. The longevity gap is 1994. Life expectancy improved between 1995 and significantly wider than in other G-8 countries, where it 1998, but the gains eroded following the 1998 financial ranges between 5 and 7 years. The large difference by crisis, and male life expectancy started falling steadily. sex in Russia suggests that specific behavioral factors are Russia's total life expectancy at birth lags behind Japan's implicated, rather than factors related to the external by as much as 16 years and the European Union average environment or adequacy of health care, which affect by 14 years. men and women to somewhat similar degrees. Two major factors behind this big gender gap are smoking The scale of Russia's demographic challenge is made and alcohol consumption, as these behaviors are very apparent through examination of the World Bank's different between men and women--even those living World Development Indicators. Although male life in the same households. Although Russian women out- expectancy at birth is only about 2 years less than in live Russian men, they are generally in worse health Brazil or Poland, the probability that a 15-year-old than women in Eastern and Western Europe. Russian boy will die before he reaches 60 is over 40 per- cent, about 16 percentage points higher than in Brazil Regional variations: Mortality rates and life expectan- and double that of Turkey. Moreover, data from the cy in Russia vary greatly by region, in part because of World Health Organization (WHO) show that it is regional differences in socioeconomic and health levels. working-age mortality that has led reductions in life The mortality rate of the economically active male pop- expectancy in Russia, while mortality among youth has ulation from region to region ranges from 3.8 deaths per dropped. 100,000 people to 17.8. Data for 2001 show that peo- ple in regions such as the Republics of Ingushetia and Life expectancy and economic development Dagestan and in Moscow have the longest life expectan- cy and live 18 years longer than those in low-income A major determinant of a population's health is its coun- regions such as Republic of Tyva, Koryak Autonomous try's level of economic development and may in part Okrug, and Komi-Perm Autonomous Okrug. explain some of the differences in mortality rates. However, the World Development Indicators show that even when income differences are taken into account, The Russian population is also aging Russian male adult mortality rates still substantially Russia's population structure is characterized by a exceed those of countries with similar per capita shrinking proportion of youth and an expanding pro- incomes. Although the Russian economy has been grow- portion of people 60 and above. Contributing to this ing strongly as a result of high oil prices for almost five phenomenon are continued very low levels of fertility years, life expectancy at birth has continue to decline. and past higher fertility levels that produced compara- xiv tively larger cohorts. Two decades ago, youth aged 0-14 Cardiovascular disease: Russia's cardiovascular disease years constituted about a quarter of Russia's population, (CVD) death rate per 100,000 population in 2002 was and those aged 60 years and above made up 14 percent 994--one of the world's highest. The comparable rates of the total. Now, those aged 0-14 have dropped to 18 were 317 in the United States, 363 in Portugal, and 225 percent. With Russia's expected total fertility rates of in Brazil. CVD accounts for almost 52 percent of deaths between 1.1 and 1.3 children per woman of reproduc- in Russia, compared to 38 percent in the United States, tive age for the years 2005-25, projections suggest that 42 percent in Portugal, and 32 percent in Brazil. Russia's those aged 0-14 will remain at about 13 percent, and high mortality among working-age men is mainly attrib- population growth rates will remain negative, averaging utable to CVD. Such statistics led a 2004 World Bank between -0.6 and -0.8. As a result, the proportion of report to conclude that Russia could gain 6.7 years in persons aged 60 and over will become more than a quar- life expectancy by matching the European Union's CVD ter of the total. Unlike other G-8 countries, the rapid mortality rates. aging of the Russian population and growing dependen- cy ratio are occurring at a lower level of GDP per capita. Cancers: Cancer mortality rates in Russia are signifi- cantly above the EU-15 average. In addition, cancer in Russia is characterized by its extremely high lethality, DETERMINANTS OF THE with a high proportion of deaths within a year of first DEMOGRAPHIC DECLINE AND diagnosis of the disease (e.g., 56 percent for lung cancer THE HEALTH DEFICIT and 55 percent for stomach cancer). Men in Russia die from cancer twice as often as women, but the cancer Non-communicable diseases and injuries are the leading incidence rate among women is higher. causes of death and ill health in Russia Traffic injuries: Traffic injuries include crashes involv- Non-communicable diseases (NCDs) and injuries are ing motor vehicles, pedestrians, or cyclists. At 20.6 the ten leading causes of death in Russia, accounting for deaths per 100,000 population, Russia's traffic mortality 68 percent of deaths. In 2003, cardiovascular diseases, rate is higher than that of other former Soviet states and cancer, and injuries accounted for 78 percent of deaths nearly double that of the other G-8 countries at 11. and 15.2 million lost years of potential life among the This rate is even more remarkable considering that there working-age population (10.3 million among men and are fewer automobiles per capita in Russia than in 4.9 million among women). Western Europe. In 2004, more than 34,000 people in Russia died in road accidents, mostly working-age NCDs and injuries also cause much of the morbidity males. and disability among the Russian total population. Like HALE, "disability-adjusted life years" (DALYs) is a Suicide: Russia's suicide rate is much higher than that measure that captures mortality, morbidity, and the of other European Union countries. The Russian suicide severity of disability. The 10 leading causes of death and rate peaked in the early 1990s, particularly among mid- ill-health also account for about 50 percent of DALYs dle-aged men. In 1994 the suicide rate for Russian men lost among the Russian total population. These causes aged 50-54 was over six times that in the United States. account for 40 percent of DALYs lost for the entire It fell slightly between 1995 and 2002. Among the European region. Central and Eastern Europe (CEE) and the Commonwealth of Independent States (CIS), Russian The mortality rates from NCDs and injuries in Russia males aged 15-19 had the second highest suicide rate at are three and five times, respectively, those in the 38.2 per 100,000 relevant population, surpassed only by European Union. For 100,000 population, Russia had Lithuania at 38.4. 605 deaths from NCDs in 2002, while the EU had 206 in 2001 (or latest data available). The same population Alcohol poisoning: A primary external cause of mor- size experienced 281 deaths in Russia from injuries tality is alcohol poisoning: death occurs when, for exam- while the EU experienced 58. ple, a healthy adult male of average weight consumes a xv half liter or more of a strong beverage (40 percent alco- cent of women, and 30 percent of teenagers were hol) without food in less than an hour. This is a particu- drinkers. Russians in rural areas drink more alcohol of larly acute and rapidly growing problem in rural areas, poorer quality than people in cities, as demonstrated by where it causes 128 deaths per 100,000 adult men. This a long-standing tendency of higher morbidity due to rate is twice that of the country as a whole. alcohol poisoning among rural populations. Violence: Russia's homicide rate increased rapidly dur- In an ongoing case-control study in Izhevsk (a city in ing the 1990s and is now among the highest recorded the Urals), an interim analysis shows that, of 1,400 anywhere. In 1999, close to 30,000 persons died from deaths from all causes among all male residents aged 25- homicide in Russia. Furthermore, the average age of 54 years in 2003-2004, 18 percent were certified as homicide victims is much lower than for most other alcohol related: mental disorders due to alcohol; alco- causes of death, so it has a very negative impact on years holic cardiomyopathy; alcoholic cirrhosis of the liver; lost. Alcohol abuse is closely related; the proportion of and acute alcohol poisoning. However, this excludes the crimes committed by men and women while intoxicated substantial proportion of deaths that are attributable to in 2002 was about 30 percent and 12 percent, respec- alcohol but arise from causes that do not explicitly tively. Victims of homicide are also usually intoxicated imply alcohol: injuries, violence, and a wide range of ill- at the time of the crime. After 1998, with increased nesses. This risk factor may explain the mortality crisis alcohol consumption among young cohorts, violent in Russia more than any other as suggested by the deaths began to rise. Increased levels of violence also results of this study. adversely affect social and economic development. Tobacco consumption: Tobacco is the only consumer The preventable risk factors for NCDs product that eventually kills half of its regular users if and injuries they follow its manufacturers' recommendation. Though risk factors may not cause disease, their pres- Associated with higher rates of CVD, many cancers, and ence increases the probability that one will develop. The chronic lung diseases, cigarette smoking is the single mortality attributable to high blood pressure, high cho- most preventable cause of disease and death in Russia. lesterol, and tobacco is estimated to contribute to more Russia has one of the world's highest smoking rates than 75 percent of Russia's deaths. The same risk factors among men, and more Russian men smoke than accounted for more than 46 percent of DALYs lost in European men--the ratio is 2 to 1.19. In 1998, 6 out of 2002. However, in terms of ill health burden, alcohol 10 male adults in Russia were smokers--more than abuse tops the list in Russia, accounting for 16.5 percent twice the comparable rate in the United States and the of DALYs lost. United Kingdom. RLMS data indicate that while smok- ing prevalence among men decreased from about 65 Alcohol abuse: In addition to alcohol poisoning, the percent in 2002 to 61 percent in 2004, the prevalence actual death determinant discussed above, alcohol abuse of smoking among women increased from 7.3 percent (heavy or binge drinking) is a major risk factor and pub- in 1992 to 15 percent in 2004. Lower smoking preva- lic health problem in Russia. Adult per capita alcohol lence among older men reflects trends over time and consumption in 1999 was 10.7 liters per adult in Russia higher death rates among long-term smokers as they versus 8.6 liters in the United States and 9.7 in the age. United Kingdom. Although these levels are not dissimi- lar, the key difference is that 75 percent of the alcohol Drug Use: Illegal drug use has increased rapidly in consumed in Russia is spirits, whereas in the United Russia over the past 10 years. In early 2005, drug users Kingdom and the United States, 56 and 60 percent, were estimated to number 500,000, including more respectively, is beer. Recent data from the Russian than 340,000 drug addicts in the lists maintained by Longitudinal Monitoring Survey (RLMS) indicate that public institutions. The number of persons registered in in 2002, alcohol consumption increased 14.5, 2.4, and medical and prevention institutions with a diagnosis of 1.1 liters per year among men, women, and teenagers, drug abuse increased by a factor of 2.1. However, it is respectively. In 2004, about 70 percent of men, 47 per- estimated that the actual number of drug abusers in the xvi country exceeds those officially registered by five to the severely obese by an estimated 5 to 20 years. The eight times. Injecting drug users are estimated to have latest (2002) Russian Behavioral Risk Factor 20-times higher risk of death than the general popula- Surveillance System results indicate that the prevalence tion, making it conceivable that a portion of Russian of overweight adults aged 25-64 in various regions teenage mortality is connected to substance abuse. ranges from 47 to 54 percent for men and from 42 to Injecting drug use is also the driving force behind the 60 percent for women. RLMS data from 2000 found 33 HIV/AIDS epidemic in Russia. Although HIV/AIDS is percent of men and 30 percent of women overweight an infectious disease that shares certain characteristics and that 12 percent of men and 28.5 percent of women with many NCDs, such as a long latency period, and were obese. However, self-reported data are often under- affects predominantly young adults, it is not considered estimated, especially the female obesity rates. According in this assessment. to data from Pitkäranta, Karelia Republic, in the 1990s and 2000, about 35 percent of women had BMI (meas- Nutrition- and activity-related factors ured) of 30 or more. Cholesterol, obesity, and hypertension should not be High blood pressure: High blood pressure, or hyper- seen as separate, individual risk factors but as a matter of tension, is a leading cause of death in Russia and the diet. High saturated fat (animal fat, hydrogenated veg- third-highest cause of morbidity as reflected by DALYs. etable fats) intake, high salt intake, low vegetable and Individuals with uncontrolled hypertension have a three fruit intake, and low intake of good (vegetable and fish) to four times greater risk of developing coronary heart oils are the determinants of high blood cholesterol, high disease and a sevenfold greater risk of having a coronary body weight, and high blood pressure among Russians. event than those with normal blood pressure. About 34- The WHO estimates that about a third of all CVD is 46 percent of males and 32-46 percent of females in due to poor diets and that better diets could lower the Russia suffer from hypertension, although these self- number of cancer cases by about 30-40 percent. It has reports may be underestimates. More than 40 percent of been suggested that the decline in fruit and vegetable males and 25 percent of females in Pitkäranta were not consumption in Russia can explain 28 percent of the aware of their hypertension, suggesting the all-Russia increase in CVD mortality. Sedentary lifestyle aggravates self-reports are be low. the problem, as moderate, regular exercise improves both physical and mental well-being and reduces the Diabetes: Complications from diabetes include blind- risks of CVD, colon cancer, diabetes, and hypertension. ness, renal insufficiency, and cardiovascular and neuro- A 2002 survey reports that 73 to 81 percent of adult logical problems. While the incidence of diabetes in men and 73 to 86 percent of adult women reported Russia is on par with the world average of 2.5 percent, having low levels of physical activity in Russia. this disease may be underdiagnosed there, with a majori- ty of illnesses being registered under other, attendant High cholesterol: About 60 percent of Russian adults diseases. WHO considers Russia among the top 10 have higher than recommended cholesterol levels, and countries in number of diabetics. about 20 percent are at high risk and in need of medical attention. A study in St. Petersburg showed a significant Other risk factors decline in HDL (so-called "good") cholesterol to dan- gerously low levels across all men in the 20 to 69 age Psychosocial stress: Psychosocial problems in Russia group, as well as for women. were increased by the drastic changes in the economic transition and reduction in the social safety net in the Obesity: Adults who are overweight or obese are at risk last 15 years. Stress, manifesting as depression and anxi- of premature death and disability. Those with a body ety, can lead to incapacity, suicide, and other violence- mass index (BMI, a common measure expressing the related injuries, as well as the development of CVD and ratio of weight to height, calculated as wt/[ht2]) of 25 to increases in CVD mortality. 29.9 are considered overweight, while those with a BMI of 30 or more are considered obese. Obesity has a sub- Socioeconomic disadvantage: Low socioeconomic stantial effect on longevity, reducing the length of life of status has been shown to affect health outcomes in the xvii West. In Russia, two studies and the 2003 NOBUS sur- WHAT ARE THE CONSEQUENCES OF vey found that people of lower socioeconomic status THE DEMOGRAPHIC DECLINE AND have higher mortality and are more likely to report bad health than people of higher status. Furthermore, people THE HEALTH DEFICIT? with the lowest socioeconomic status are more likely to The unprecedented mortality upsurge Russia experi- report frequent consumption of hard alcohol and have enced in the 1990s is the product of a a long-term dete- greater problems accessing health care systems and fol- rioration of the health of the population associated with lowing treatment regiments than richer people. a prolonged period of highly unhealthy lifestyles and adverse environments, aggravated by a difficult socioeco- Road safety-related risk factors: A major risk factor is nomic and political adjustment process. Russia's total the lack of strict road safety measures similar to those in population is expected to decline in the years ahead and Europe and the United States, where improved road will be accompanied by the aging of society as fertility quality, better driver training, traffic rules, vehicle safety rates remain under the replacement level. This has sever- inspection, and road safety campaigns are common. al important implications: Other risk factors are drivers' and pedestrians' wide- spread failure to observe safety rules, speeding, and driv- Fewer workers: If trends persist, the size of the Russian ers' widespread failure to wear seat belts. The rates of labor pool will continue to shrink--significantly--in alcohol abuse make alcohol impairment an important the decades ahead. Furthermore, the population may risk factor in crashes. Inadequate post-crash emergency decrease further as a result of the changing sex ratio that medical care is also a major risk factor in some Russian will probably accelerate fertility decline, resulting in an regions. extreme sub-replacement fertility level. This presents a serious threat to Russia's development as the simultane- Work-related risk factors: Occupational risk factors ous decline in the working-age population and increase include exposure to hazardous substances and lack of in the elderly population could have an adverse impact safety measures. The International Labor Organization on the economy. According to IMF projections, a estimates that about 6,000 fatal accidents, 118,000 dis- decrease in working-age population will reduce labor ease cases, and 131,000 Russian deaths in 2001 were productivity and incentives for investment in human work related. and physical capital, which will in turn reduce per capita GDP growth. Government budgets will tighten as tax Interactions among risk factors revenues fall from shrinkage in the working-age popula- The relative risk of developing NCDs and suffering tion and the increasing needs and demands of an aging injuries increases sharply when various risk factors are population. As the elderly population increases, the combined. Multiple factors increase the risk of CVD overall rate of saving and investment will decline as pen- mortality by five to seven times. sions, health care, and long-term residential care absorb more resources. While there clearly are many open questions on the cur- rent health crisis in Russian, and there is much to be However, a healthy population of 65-75 year-olds may studied, the policy implications are clear: without a be a sizable untapped workforce, and, providing meas- doubt tobacco, alcohol abuse, and some aspects of ures are taken to reverse the mortality trends and assum- national diet and physical inactivity contribute signifi- ing lives lengthen overall, raising the retirement age cantly to premature mortality, ill health, and disability would improve the medium-term solvency of pension due to NCDs and injuries in Russia. These are also fac- systems. This change may require developing new rela- tors where effective and measurable interventions are tionships with older workers, including mechanisms to available that generate benefits in years rather than attract and retain them, especially knowledge workers. decades, as shown in developed countries such as Finland and the United States. The destabilization of families: The gender gap in life expectancy has led to instability in marriage and an xviii extremely high proportion of widows (the percentage of sis identified the four most expensive groups of diseases widows in Russia aged 30-44 is about four times that in and injuries as circulatory system diseases, respiratory the United States). diseases, external causes, and digestive system diseases. These four conditions account for more than 50 percent Growing regional disparities: Uneven fertility, mortali- of Russia's total health expenditures and pose tremen- ty, population growth, and life expectancy in different dous challenges to the health care system. In 2003, the regions and among social and ethnic groups could exacer- most expensive group comprised circulatory system dis- bate existing disparities. Difficulties in providing equitable eases--hypertension, ischemic heart disease, and cere- access to resources for the poor may give rise to social and bro-vascular disease--and cost the public health system political challenges, especially in a country as vast as US$2.7 billion (83 billion rubles), or 20.8 percent of Russia. This is an important policy consideration for any total expenditures. government, particularly when resources are scarce. What effect has adult ill health had on the eco- National security risks: The demographic and health nomic outcome of the population? The overarching crisis in Russia will present many challenges to national answer from an assessment prepared for this study is security: (a) the number of conscription-aged men will clear: poor adult health negatively affects economic well- plunge rapidly in the decades ahead; (b) a growing per- being at the individual and household level. If effective centage of the military budget will have to be allocated action were taken in Russia, improved health would play for the provision of medical, nutritional, and substance an important role in sustaining high economic growth abuse programs for draftees and soldiers deemed med- rates at the macro-level. ically unfit for duty; (c) long-term economic growth will erode without large cohorts of healthy and skilled young The main findings of this assessment are: and middle-aged adults; and (d) if its vast territory is The cost of absenteeism due to ill health: On aver- depopulated, instability could grow and the country age, 10 days per employee per year are lost due to illness could become increasingly difficult to govern. in Russia, while in the EU-15 the average is 7.9 days. Sickness absence incurs a direct cost from benefits paid Impact of NCDs and injuries on health to absent employees and an indirect cost of lost produc- care costs and the economy tivity. The overall cost varies between 0.55 percent and The large contribution of NCDs and injuries to sickness 1.37 percent of GDP (annual absenteeism rates can be and mortality in Russia raises two other serious econom- converted into a monetary value either by using the ic issues. First, as many NCDs are chronic conditions average wage rate, resulting in the lower value, or the requiring expensive medical treatment, to what extent is GDP per capita, resulting in the higher value). This is a the Russian health system burdened with the increasing significant impact, given that the indicator fails to cap- cost of treating NCDs and injuries? Second, to what ture the many other ways ill health impacts the labor extent does Russian society suffer economic conse- market. In particular, it does not capture the effects of quences from premature mortality, ill health, and dis- reduced productivity and mortality. ability among its working-age adults? The impact on the labor supply: Ill health also High medical treatment costs: The 2003 health care impacts labor supply because jobholders with chronic cost estimates from two regions in Russia were analyzed diseases or alcoholism are more likely than healthy indi- and extrapolated to the national level for this study viduals to either retire early or lose their jobs and draw using US$13 billion, the widely accepted estimate of on state pensions. While a hypothetical Russian male Russia's total health care expenditure, as a denominator, aged 55 with median income and other average charac- to determine the effect of NCDs and injuries on total teristics would be expected to retire at age 59, chronic health expenditures. The regions were the Chuvash illness would lower his expected retirement age by 2 Republic (an agricultural region) and Kemerovo Oblast years. Similar results are obtained for females. Also, an (an industrial region). The most expensive medical con- individual who suffers from chronic illness has a signifi- ditions proved to be NCDs or injury related. The analy- cantly higher probability of retiring in the subsequent xix year than the same individual free of chronic illness. appeared just 2-7 years after the elimination of risk expo- This all means that chronic illness is a highly significant sure, even among people in older age groups. predictor of subsequent retirement in Russia. The lower the income of an individual in Russia, the more chronic The population-based strategy aims to change disease- illness affects the decision to retire. This implies that related lifestyle choices, environmental factors, and their less-affluent people carry a double burden of ill health: social and economic determinants in an entire popula- first, they are more likely to suffer from chronic illness, tion (e.g., information and communication programs and second, once ill, they are more likely to suffer worse addressing the risks of smoking and the value of smok- economic consequences-less income-than rich people, ing avoidance, excise and other taxes to reduce smoking, tending to perpetuate socioeconomic disadvantage. and restrictions on smoking in public places and on tobacco advertising). The main argument for this strate- Job loss: Alcohol abuse, arguably an important factor gy is that it targets a high proportion of NCD morbidi- in explaining the high adult mortality in Russia, signifi- ty and mortality: 5 percent of a population have very cantly increases the probability of job loss. low exposure to risk factors and 25 percent have very high exposure. By targeting those with average levels of The impact on the family: The death of a household exposure, 70 percent of the population can learn to member affects other household members' welfare and avoid risk. Furthermore, interventions for this large behavior in various ways. Alcohol consumption was group are far less expensive than the intensive interven- found to increase by about 10 grams per day as a conse- tions needed by the high-risk group. quence of the death of an unemployed household mem- ber and by about 35 grams if the deceased had been In the "high-risk strategy," individuals at high risk of employed. Also, the probability of suffering depression developing selected diseases are identified, and actions are increased by 53 percent when controlling for other rele- planned to reduce their disease burden through provider- vant factors. Chronic illness has also negatively affected based interventions (e.g., clinical interventions to treat and household incomes, particularly during 1998-2002, counsel individuals about risk factors for CVD--smoking, when chronic illness contributed an estimated annual excessive alcohol consumption, hypertension, hyperlipi- loss of 5.6 percent of per capita income. demia, diabetes and obesity--management of patients at high risk according to established clinical practice guide- lines; and therapy for individuals once overt CVD has WHAT STRATEGIES AND occurred). High-risk strategies pose a major task for health INTERVENTIONS PREVENT services, and their per-person costs can be high. NON-COMMUNICABLE DISEASES The government's role in disease AND INJURIES? prevention Some developed countries have implemented a range of From an economic perspective, government intervention effective strategies to prevent deaths and illnesses from is justified as a means to achieve a net improvement in NCDs and injuries. The research on effective policy social welfare, in terms of increased efficiency, and for measures and NCD risk factor interventions is growing reasons of equity through redistribution. Alternative steadily, enabling informed policy making. While the government interventions, such as regulations, taxation, level of impact varies by strategy, local circumstances, and public provision, need to be assessed to determine and resource availability, the overwhelming evidence sug- their costs and benefits, as sometimes governments lack gests that an integrated strategy is most effective. Such the capacity to correct market failures. With regard to strategy incorporates all appropriate actions to reduce the health, individuals cannot clearly perceive the costs of burden of disease, including both population-based and their actions to themselves or others because informa- high-risk prevention strategies. The North Karelia Project tion is often incomplete or imperfect, justifying govern- in Finland exemplifies an integrated strategy. Its results ment intervention. Some argue that in the areas of show that the effect of prevention efforts may occur in behavior and lifestyle, a wider government role is justi- years rather than decades: measurable improvements fied when health benefits outweigh the curtailment or xx modification of individual choices. In the case of Recently, the Russian government set up institutional NCDs, a strong case could be make for government structures to respond to NCDs and injuries, but it still intervention as a result of a widespread policy failure needs greater institutional capacity to improve health that has often ignored these conditions relative to com- promotion and disease prevention programs. Many municable diseases. health experts, including Russian government represen- tatives, generally agree that the Russian health system As the etiology of most NCDs is complex, risk factors and the Ministry of Health and Social Development should be managed from multiple directions. The gov- (MOHSD) are not yet equipped to tackle NCDs and ernment's role in disease prevention is multifaceted and injuries effectively. includes setting priorities, formulating policies, enacting and enforcing regulations, rallying constituencies for dis- ease prevention, establishing infrastructure and institu- WHAT ADDITIONAL ACTIONS CAN tions, and educating the public. Many disease-preven- RUSSIA TAKE? tion efforts also require larger health system reform, for example, financial incentives to strengthen public health MOHSD is preparing a federally targeted program to prevention and community-based intersectoral interven- prevent and control premature death, ill health, and dis- tions to benefit the majority of the population. ability from NCDs and injuries. A well-defined and structured national program of population-based and Role of private employers clinical interventions to confront NCDs and injuries would help improve social welfare and contribute to sus- Private sector involvement is particularly important. The tainable economic growth by (a) bettering the health of poor health of employees quickly affects a company's the economically active population; (b) reducing labor bottom-line and has a longer-term impact on profits. supply and productivity losses from preventable deaths, Businesses have a vested interest in supporting activities illnesses, and disabilities from NCDs and injuries; and (c) to improve employee health and can have a strong influ- minimizing regional disparities by reducing social risks for ence on their employees' behavior and make them aware NCDs and injuries in the most vulnerable regions. A of health risks in ways unavailable to the government. nationwide mortality reduction program should include The involvement of major Russian companies, multina- three focus areas: (a) federal-level policies and strategies, tional corporations, and other stakeholders with experi- (b) priority sub-programs in regions, and (c) measures to ence in employee- and community-directed health pro- improve road safety and emergency services. grams will be critical in reducing NCDs and injuries. A. Federal Level Policies and Strategies IS RUSSIA DOING ENOUGH TO National subprograms must be defined in light of the CONFRONT NON-COMMUNICABLE magnitude of each health threat, the political commit- DISEASES AND INJURIES? ment required to solve or manage it, and feasibility and cost-effectiveness. The support mechanisms required for It is assumed that over the last three decades the Soviet success--legislative frameworks, institutional capacity, health care system lagged behind that of western coun- and federal oversight also need to be considered and tries in terms of health care outcomes due to its inability developed. to take full advantage of new medical knowledge and technological advances, including new treatment regimes and drugs. Ongoing health reforms were initiat- B. Priority Subprograms in Regions ed in Russia during in the 1990s to address this situa- Support should be provided for the implementation of tion, mainly focusing on financing and delivery of serv- national priority programs in the regions but allowing ices. Although these reforms changed the health sector's for regional differences in health status and choosing legal and organizational structures and contributed region-specific, appropriate, effective interventions. important legislation for addressing NCDs, the public Activities should include primary prevention (before health system largely retained its infectious disease focus. problems arise) through population-based interventions xxi targeting alcohol, tobacco, and diet/physical activity and gram. Regions should be responsible for implementing secondary prevention to control disease in its early stages national laws, standards, and guidelines; implementing and prevent progression through clinical interventions emergency medical services; and developing an inter- targeting individuals with high blood pressure, elevated institutional emergency medical network, including the cholesterol, and diabetes. ambulance network and emergency communication system. The subprograms should build on the Countrywide Integrated Non-communicable Disease Intervention (CINDI) Program experience and promote healthy WHAT HEALTH IMPROVEMENTS behaviors by awarding "health promotion initiative COULD RESULT FROM AN INTENSI- grants" to stimulate innovative health promotion and FIED PROGRAM OF ACTION? disease prevention initiatives at the regional and munici- pal levels and to build capacity at these levels for imple- To estimate the effect reductions in cardiovascular, menting integrated approaches. Technical assistance digestive, and external causes of diseases could have on should be provided to assess the capacity and readiness life expectancy in Russia, an assessment was conducted of various partners and organizations to undertake using the Multiple Decrement Life Table approach. health promotion and prevention activities, to establish efficient organizational and managerial structures for If mortality from preventable or treatable components health promotion, and to conduct process and outcome of circulatory and digestive diseases and external causes evaluation studies to measure federal program success. were reduced in working-age adults in Russia, important improvements would result in life expectancy for both C. Improved Road Safety and men and women. For example, life expectancy at birth Emergency Services for men could be improved by as much as 5 years by The federal government should (a) ensure that an reducing CVD by 20 percent. The impact of reducing enforceable legislative framework is in place by review- external causes of mortality (road accidents, intentional ing and revising existing laws and reviewing and adjust- self-harm, and assaults) suggests another useful avenue ing regulations that implement them; (b) formulate a for raising life expectancy, especially for those in middle national road safety strategy and subprogram, identify- age. The impact of reducing these diseases among men ing risk factors as a basis for planning and improving is more evident in adulthood. effective prevention of injuries through a combination of education, regulation, enforcement, engineering, and WHO estimates also show that major improvements in technology; (c) establish mechanisms to forge links with healthy life expectancy can be expected by reducing public sector entities, industry, and NGOs; (d) identify major risk factors. The region that includes Russia, dangerous road corridors for early action; (e) set stan- EUR-C, can expect to gain more than 10 years of dards and disseminate clinical protocols specifying pro- HALE by reducing the 20 leading risk factors. This is a cedures on clinical management of patients during very important finding relative to economic develop- emergency medical services on the road, while in trans- ment and cost-benefit analyses because increasing life port, and in trauma centers and hospitals to reduce pre- expectancy but not healthy life expectancy would signif- admission death rates and disability from accidents; (f ) icantly burden the health system. create a national road accident database and establish/upgrade a management information system on road traffic accidents; (g) establish, implement, and WHAT ARE THE POTENTIAL ECONOMIC maintain a robust monitoring and evaluation framework BENEFITS OF MORTALITY REDUCTION with indicators and national and regional targets; and IN RUSSIA? (h) provide technical assistance to regions. Policies to reduce adult mortality would have a signifi- In turn, each region should develop its own road safety cant effect on the economy that may be seen by com- strategy and action program following the federal pro- paring three scenarios. The status quo scenario assumes xxii that 2002 levels of adult mortality from NCDs and estimates, the Russian Federation will lose in 2005 injuries will remain constant until 2025. An optimistic US$11.1 billion of national income as a result of the scenario assumes that new policies would bring Russian impact of deaths from heart disease, stroke and diabetes mortality rates down to the current level among EU-15 on labor supplies and savings. This figure is estimated countries: an annual rate of reduction of 4.6 percent for to increase to US$66.4 billion by 2015. The magnitude NCDs and 6.6 percent for injuries. An intermediate sce- of the estimated loss in Russia is better appreciated nario assumes that new policies would achieve half the when it is compared with the significant lower estimated improvement of the optimistic scenario. losses in the United Kingdom: US$1.6 billion in 2005 and US$6.4 billion by 2015. As these losses accumulate The main conclusion of an assessment prepared for this over time because each year more people die, the esti- study is that the benefits would be substantial for the mated accumulated loss in Russia during the 2005-15 Russian economy as a whole, irrespective of how they period amounts to US$303.2 billion as compared to are evaluated. This occurs despite the fact that the only US$32.8 billion in the United Kingdom. assessment concentrates only on the effect of mortality reductions, without considering the affect of morbidity When these losses are translated into percentage reduc- reductions. The main findings are: tion in GDP, WHO estimates that in 2005 1 percent of the Russian GDP was reduced, and by · The static economic benefits (i.e., valuing a year of 2015 the percentage reduction in GDP would be over 5 life by one GDP per capita) of gradually bringing percent of GDP, far higher than the estimated reduction the adult NCD--and injury--mortality rates down of 1 percent of GDP in other countries such as Brazil, to the EU-15 rates by 2025 are estimated to be China and the United Kingdom. In large measure the between 3.6 percent and 4.8 percent of the 2002 estimated large losses in Russia are due to higher rates of Russian GDP. CVDs (the leading killer of the Russian working-age · When a broader concept than GDP per capita is population as discussed in chapter 3) than in other considered (measured by adding the value of changes countries. in annual mortality rates using a "value of a statisti- cal life" to changes in annual GDP per capita), the "welfare" benefits from achieving EU-15 rates by CONCLUSION 2025 are estimated to be as high as 28.9 percent of the 2002 Russian GDP. Reducing NCDs and injury-related mortality rates among Russian working-age adults will have a major · The dynamic benefits of improving adult health, i.e., macroeconomic and poverty reduction impact, regard- the effect on economic growth rates, are massive and less of how this is measured. Based on the results of the growing over time: while in 2005 the difference in assessment conducted for this study, as well as the recent the per capita GDP between the status quo scenario WHO estimates described above, the expected econom- and the most optimistic scenario is only US$105- ic benefits are of a magnitude that easily outweighs the 324, by 2025 this difference would have grown to costs of health promotion and disease prevention pro- US$2,856-9,243. grams. Given the significant positive effect on economic growth from investing in health (Barro 1997, Suhrcke at al. 2005b), governmental intervention is urgently need- Additional Estimates of the ed in Russia to develop health-enhancing policies and Macroeconomic Consequences of NCDs programs to address the alarmingly high mortality rates and Injuries in Russia among the working-age population. These efforts should In addition to the economic assessment prepared for this be seen as key investments to help improve the general study and discussed in the previous sections of this welfare of the population and secure sustainable eco- chapter, recent estimates undertaken by WHO (2005) nomic growth in the future. also show a very dire picture. According to the WHO xxiii 1 Chapter I. The Demographic Decline and the Health Deficit R ussia is suffering a prolonged and debilitating rates of chronic, noncommunicable diseases (NCDs) demographic crisis fueled by low fertility and with "double burden of disease" to (c) increasing life high mortality. Some observers note that this expectancy with a primarily high burden of NCD and demographic "devastation" is unprecedented among finally to (d) reduction of age-specific NCD rates. The industrialized nations. The decrease in fertility and last stage is illustrated by a number of Western coun- increase in mortality since the 1960s were exacerbated tries. The increase in NCDs is classically explained by in the 1990s by the turbulent political and economic two developments: aging of the population and changes transition and a decline in the population's well-being in lifestyles (especially in dietary and drinking habits, that occurred soon after. In recent years prosperity has reduced physical activity, and smoking). Analysis of past increased due to rapid economic growth, but current changes does not support a single explanation for demographic and health trends remain unsatisfactory-- improvements in a population's health status. Neither even alarming. This chapter examines recent and pro- higher income (and the related better nutrition and jected fertility and mortality trends in Russia; assesses increased resistance against infections) nor increased their impact on the size, growth, and structure of the benefit from specific professional interventions alone population; and presents the challenges that will likely can satisfactorily explain historic declines in mortality. be posed to the country's general welfare during the Rather, complex combinations of societal changes fed by next few decades. new knowledge shape the health status of a population. A Framework for Understanding In line with the open nature of the health transition the Demographic and Health school of thought, updates to the classical model have been proposed periodically. These updates, while Conditions in Russia describing improvements in health conditions due to Conceived by Omran in the mid-1960s (Omran 1971), socioeconomic changes and medical advances, acknowl- the epidemiologic transition theory focuses on different edge uneven development resulting from several persist- societies' changing health, mortality, survival, and fertility ing and new health problems related to rapid unplanned over time and place, linking these effects to their socioe- urbanization, emerging and re-emerging infectious dis- conomic, environmental, lifestyle, demographic, health eases, and increasing obesity linked to changes in nutri- care, and technological determinants and correlates (Box tion and exercise. As Omran noted in revisiting the epi- 1.1). This concept provides a useful framework for ana- demiologic transition theory nearly 30 years after its lyzing past patterns in the health of a population publication, stagnation or reversal of the transition is (McKeown 1985; Gribble and Preston 1993). It offers possible during economic, political, environmental, or empirically tested explanations of the mechanisms operat- morbidity crises (Omran 1999). Indeed, the experience ing on the transformations in a population's health and is in Central and Eastern Europe since the early 1970s, open to adaptations as new evidence emerges. It does not when life expectancy began to plateau or even decrease, purport to make predictions, but offers a coherent theory also confirms the omnipresent risk of counter-transi- of past changes and patterns, and such theory can illumi- tions (Mesle et al. 1996). nate thinking about the future. The current situation in Russia clearly exemplifies a According to classical theory, countries transition with reversal of the epidemiological transition as a prolonged economic development from (a) primitive and poor period of highly unhealthy lifestyles and environments societies with mainly infectious diseases to (b) increasing coupled by the economic stress of the past 15 years not 1 Box 1.1 The Factors Behind the Epidemiological Transition During an epidemiological transition, long-term shifts occur in mortality, disease, and survival patterns. Several factors contribute to such transition, the most important of which include the following (Feachem et al. 1992): Demographic and health factors: Changes in fertility and mortality, along with changing disease and health patterns, are the main agents of the epidemiologic transition. This is inexorable as populations become older and live longer, and the numbers of adults and elderly increase. Largely as a function of changes in the age structure of a population, the absolute number of sick and dying adults and elderly has increased, imply- ing a shift of the burden of death and disease from the younger to the older groups. Changes in risk factors: Though risk factors may not cause disease, their presence increases the probabili- ty that disease will eventually develop. Risk factors involved in the epidemiological transition include biologi- cal and environmental factors, as well as social, cultural, and behavioral (e.g., changing lifestyles) factors. Changes in the exposure to and the magnitude of risk factors, as well as risk-averting interventions, have altered age-specific morbidity and mortality rates. For example, improvements in nutrition, personal hygiene, and housing in Western countries during the 18th and 19th centuries contributed to the decline in overall mortality from infectious diseases and to the steady rise in life expectancy. Changes in health care practices: Health care improvements have caused changes in absolute and rela- tive rates of ill health and death. Changes in access to, the use of, and the effectiveness of health services may influence an epidemiologic transition by affecting disability and mortality. For example, after World War II, the widespread use of antibiotics and the large-scale use of vaccines helped improve health dramatically across Europe. Mortality, through its declining or rising trends, is the most fundamental force in the epidemiological transition. Fertility is a major co-variable, as it sets the pace of population growth when mortality reaches low levels (with modifications by net migration). only halted progress in improving health status, but figure is expected to decline by over 30 percent during shortened life expectancy, particularly among working- the next 50 years, as all measures of demographic age males. This situation defines a new pattern of the processes show that Russia will continue to undergo dra- epidemiological transition that deviates from the experi- matic changes in its population dynamics in the coming ence in a number of western countries that is character- decades. The average annual population growth during ized by a reduction of age-specific NCDs rates and 1990-2003 was -0.3 percent, and continued high mor- increasing life expectancy (Omran 1999, Murray and tality and declines in fertility are expected to lead to fur- Bobadilla 1997; Kingkade and Arriagada 1997). ther negative population growth (Figure 1.1). Overall, it is estimated that the population of Russia The Shrinking of the would be 17 million higher than at present if age-specif- Russian Population ic mortality rates in Russia had followed the patterns experienced by European Union-15 countries (EU-15) Russia's population was 149 million in 1992 but since the mid-1960s (Andreev 2005). This figure is declined by 6 million as of 2003 to 143 million. If cur- comparable to the country's total lives lost in World War rent low fertility and high mortality trends persist, this II (Andreev 2005). 2 Figure 1.1 Estimated Population Growth Rates, Russia, 1980-85 to 2020-25 0.8 0.6 0.4 0.2 0 Percent -0.2 -0.4 -0.6 -0.8 -1 1980- 1985- 1990- 1995- 2000- 2005- 2010- 2015- 2020- 1985 1990 1995 2000 2005 2010 2015 2020 2025 Source: UN Population Database. Box 1.2 The Global Context for Demographic Trends Global demographic trends since the 1950s--notably declining fertility rates and increasing life expectancy, espe- cially in developed countries--are expected to have major ramifications in the next several decades. Although fer- tility rates remain high in some global regions (Africa and the Middle East), in others (East Asia and Central and Eastern Europe), they have fallen below the replacement rate (2.1 births per woman of reproductive age). Simultaneous with declining fertility rates, global life expectancy has risen, especially in developed countries. As a result of these trends, the rate of growth in the world's population is expected to decrease in the coming years. Worldwide annual population growth was 1.25 percent in 2003 but is projected to drop to 0.25 percent by 2050. In some Central and Eastern European (CEE) countries, populations are expected to decline by over 30 percent in the next 50 years. Demographic trends also are expected to produce changes in the population structure by age groups. Currently, approximately 600 million people worldwide are aged 60 years and over. The global population will continue to age in the coming years, and the median age will increase from 27 in 2003 to 37 in 2050. The number of people 60 and over will double to 1.2 billion by 2025 and reach 2 billion by 2050. The vast majority of people 60 and over will reside in the developing world. The aging of the CEE population is expected to accelerate beginning in 2015, increasing economic and social demands on affected countries. The working-age population as a percentage of the entire population will start to decline dramatically in some regions before 2050, first in CEE and especially in Russia. 3 These trends, which are Figure 1.2.Estimates of Total Russian Population, 2000-2025 (millions) occurring in the context of the global demographic 150,000 trends described in Box 1.2 (on page 3), will contribute 145,000 Total Population to continued population 140,000 shrinkage in the coming years. 135,000 130,000 The United Nations Population Division projects 125,000 that Russia will lose approxi- 120,000 mately 18 million people between 2000 and 2025 115,000 (Figure 1.2). Box 1.3 2000 2005 2010 2015 2020 2025 explains the assumptions Source: UN Population Database. underlying these projections. Declining Fertility Rates 1960s, Russia's total fertility rate stood at approximately Russia is among many Western countries with fertility 2.6. Its fertility rate dropped to about 2.2 in the late rates below the replacement level of 2.1 children per 1980s, fell below replacement in the early 1990s, and in woman of reproductive age (Figure 1.3). In the early 2000-2005, it was 1.1. Projections suggest that Russia's Figure 1.3 Total Fertility Rates in Selected Developed Countries, 1950-2025 4.2 4.0 France 3.8 Germany 3.6 Italy 3.4 3.2 Russia 3.0 UK oman 2.8 US w 2.6 per 2.4 ths 2.2 Bir 2.0 1.8 1.6 1.4 1.2 1.0 1950- 1955- 1960- 1965- 1970- 1975- 1980- 1985- 1990- 1995- 2000- 2005- 2010- 2015- 2020- 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 Source: UN Population Database. 4 Box 1.3 The Accuracy and Reliability of Demographic Projections "Demographic projections" are conditional statements about the future, given a set of assumptions about the key population flow variables: fertility, mortality, and migration. If the computation is performed accurately, demo- graphic projections provide a picture of the future based on assumptions. "Demographic forecasts," on the other hand, are analysts' statements on what the situation will most likely be in the future. Since no one can know with certainty what will determine birth rates in the next 10 years or of the emergence of an epidemic that would sub- stantially raise mortality, forecasts have a high probability of being inaccurate. Projections, by their very nature, can never be inaccurate, since they are simply conditional statements based on some assumptions. The United Nations (UN) Population Division produces population projections. In the latest set of projections (U.N. Population Database), the future population of each country is projected from an estimated population for July 1, 2005, a figure that in Russia's case is based on the most recent population data available (from a census or population register, updated to 2005) as well as available data on fertility, mortality, and international migration. To project population until 2050, the UN Population Division applies assumptions regarding future trends in fer- tility, mortality, and migration. As the future is uncertain, the UN Population Division produces several projection variants, meaning that different combinations of assumptions are applied and different sets of projections are produced. The most commonly used of these variants is the medium variant, which assumes medium fertility and normal mortality and migration. Fertility might be assumed to follow a path derived from models of fertility decline that are established on the basis of experiences for countries that saw falling fertility during 1950­2005. In Russia's case, fertility has been below 1.85 children per woman of reproductive age for some time. In this case, the assumption is that over the first 5 or 10 years of the projection period, fertility will follow the recently observed trends in Russia. After that period, fer- tility is assumed to increase linearly at a rate of 0.07 children per woman per year over each five-year period. Projections under the normal mortality assumption are done on the basis of demographic models of change of life expectancy. These models produce smaller gains the higher the life expectancy already reached. The selection of a model for each country is based on recent trends in life expectancy by sex. These projections are done carefully: the 2004 revision, for example, incorporates for the first time a longer survival for persons receiving treatment with highly active antiretroviral therapy for HIV/AIDS. At the global level, projections of population size, for example, have tended to be quite accurate, with the error size at less than 3 percent. At the country level, errors can be larger, especially in looking at very long time peri- ods. Also, errors can be larger for less-developed than developed countries and for larger countries. It is usually the case that fertility is overestimated, whereas mortality improvement has usually been underestimated (though not for Africa or the Commonwealth of Independent States [CIS]). The most important thing to remember for this report is that, by their nature demographic projections do not provide a forecast. total fertility rate will remain below replacement beyond As will be discussed in the next section, while many 2025. The crude birth rate (live births per 1,000 popu- developed countries are experiencing low birth rates, lation) decreased dramatically from 14.7 in 1989 to 8.7 Russia's death rate is unusually high, and it is the combi- in 2000 but increased slightly to 9.7 in 2002. nation of these two rates that cause concern. 5 Figure 1.4 Standardized Mortality Rate from All Causes for People Ages 0-64, 1986-2001, Selected Countries Deaths per 100,000 population 900 800 700 600 France 500 Germany Italy 400 Russia UK 300 200 100 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Source: WHO European Health for All Database. Figure 1.5 Infant Deaths per 1000 Live Births 23 22 21 20 19 Russian Federation 18 17 16 15 14 1980 1985 1990 1995 2000 2005 2010 Source: WHO European Health for All Database. 6 High Mortality Rates Maternal Mortality According to data from the World Health Organization's The estimated maternal mortality ratio (the number of (WHO) European Health for All Database, the mortali- maternal deaths during a given period per 100,000 live ty rate from all causes of death in Russia is dramatically births during the same period) in Russia was 31.3 in higher than that in any other G-8 country (Figure 1.4). 2003, approximately six times the average ratio for EU Russia's standardized death rate in 1986 was about 1.6 countries of 4.9 (Figure 1.7, on page 8, shows the trends times higher than for the rest of the group. This differ- in the 1990s). According to Russian experts, 70 percent ence continued to grow, reaching a peak of about three of maternal deaths are avoidable if measures are adopted times higher in Russia than in the other G-8 countries in to address leading causes of death, such as post-partum the 1993-95 period. The large difference between hemorrhage and sepsis after delivery (Sakevich 2002). Russia's standardized mortality rate and those of the other G-8 countries was holding as of 2002. Data suggest that women in Russia are more at risk to undergo an abortion and to die from it than women elsewhere in Europe and Central Asia. Abortions Mortality among Population Groups accounted for 16 percent of Russian maternal deaths in 2003. Unlike other countries, where deaths due to the Infant and Child Mortality consequences of abortion performed outside medical Although the infant mortality rate (an important institutions after 12 weeks of gestation involve predomi- indicator of quality of life in general and quality of nantly young or unmarried women, half of the women medical assistance delivered to children under one year who died in Russia were aged 30 to 40 years and had of age) in Russia has been declining (Figure 1.5), it previous pregnancies (Zhirova et al. 2004). The leading remains higher than in any other G-8 country. Russia's cause of death was post-abortion infections (in 80 per- 2003 infant mortality rate, 12.4 deaths per 1,000 births, cent of the cases). Apparently, abortion is used as a pri- is still three times those of France, Germany, and Italy. mary method of birth control, for reasons that might Also, despite a slight decline since the mid-1990s, mor- include lack of awareness and limited access to effective tality rates for children under age five are also signifi- methods of contraception (perhaps due to high costs or cantly higher in Russia than in the other G-8 countries lack of availability) or doctors' preference for abortions (Figure 1.6). over family planning methods (perhaps due to doctors' Figure 1.6 Trends in Under-5 Mortality in Selected Countries, 1990-2002 25 1990 20 1995 2000 2002 15 10 5 0 Canada France Germany Italy Japan Russia UK US Source: World Bank World Development Indicators 2004. 7 Figure 1.7 Trends in Maternal Mortality in Selected Countries, 1991-99 60 Russia ths 50 UK Bir US Live 40 France 30 Germany 100,000 Italy 20 per Japan 10 Deaths 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: World Bank World Development Indicators. financial gains from abortions). Further understanding data quality has been ruled out as an explanatory vari- of the reasons for these high abortion rates would help able because demographic data are considered reliable: in identifying and removing obstacles to the use of Russia's registration of vital events is nearly complete modern contraceptive methods that would contribute to and its coding of broad categories of death is reasonably reducing maternal mortality, improving women's repro- accurate (Leon et al. 1997, Shkolnikov et al. 1997, ductive health, and reducing secondary infertility, all of 2001). which in the long run may have a positive influence in overcoming the current demographic deficit. Russia is one of the few countries in the world where life expectancy is falling. However, the situation in Low life expectancy due to Russia has specific features that distinguish it from other high adult mortality countries, such as several in sub-Saharan Africa, where a generalized HIV/AIDS epidemic is driving losses in life Analysts have described the country's mortality profile as expectancy. Both the current low level of life expectancy the most puzzling aspect of the Russian transition. Poor and the recent declines were driven largely by increasing mortality among those of working age, Figure 1.8 Life Expectancy at Birth, Russia and the with a singular rise in mortality at young European Union, 1970-2000 adult ages, with the greatest contribution from cardiovascular diseases and injuries 80 (Mesle 2002). 78 76 After lagging far behind in the 1940s, the 74 EU average average life expectancy at birth for both 72 Russian Federation sexes in Russia almost caught up with the 70 most advanced Western countries in the 68 1960s (Vallin and Mesle 2001). By the early 1960s life expectancy in the former 66 Soviet Union had nearly reached that of 64 the United States, but death rates increased 62 significantly, particularly among adults, 1970 1975 1980 1985 1990 1995 2000 thereafter. By 1980, the difference in life expectancy was nearly 8 years. As shown in Source: WHO European Health for All Database. 8 Figure 1.8, the post-1984 period Figure 1.9 Mortality among the Working-Age Population in was marked by wide swings in life Russia Drives Fluctuations in Life Expectancy expectancy. A reduction in mortali- ty occurred during the Gorbachev 150 anti-alcohol campaign (1985-87), 25-64 years 75-84 years but as the effects of the campaign 125 0-14 years dwindled, a dramatic deterioration ensued (1990-94) as the country tality 100] entered a severe crisis associated Mor = 100 with the sudden transition to a mar- ket economy (Shkolnikov et al. [1980 1997, Shkolnikov and Nemtsov. Relative 75 1997, Shkolnikov et al. 2001, Mesle 2002). Life expectancy improved by three years between 1995 and 1998, 50 but the gains eroded following the 1980 1985 1990 1995 2000 1998 financial crisis, and male life M+F combined expectancy started falling steadily. Source: WHO European Health for All Database. Currently, total life expectancy at birth in Russia lags behind that of Japan by as much as almost 36,000 deaths in this age group annually. The 16 years and the European Union average by 14 years. scale of this phenomenon in Russia is apparent in Table By 2002, life expectancy in Russia had fallen below 66 1.1. It shows that, although male life expectancy at birth years, well below the 1965 peak. Russia is the first coun- is about 2 years less than in Brazil or Poland, the proba- try in the history of modern nations to experience such bility that a 15-year-old Russian boy will die before he a significant peacetime loss in life expectancy. reaches 60 years is over 40 percent, about 16 percentage Changes in mortality rates among different age groups Table 1.1 Life Expectancy and Adult Mortality in Selected Countries, 2000-2001 over the 1980-2000 period show unusually steep fluctua- Life expectancy Probability of dying Probability of dying tions in life expectancy. The at birth (2001) between ages 15 and 60 between ages 15 and 60 trends shown in Figure 1.9 Country (2000-2001, % males) (2000-2001, % females) underline that it is mortality Russian Federation 66 42.4 15.3 among the working-age Japan 81 9.8 4.4 group that has played a key France 79 13.7 5.7 role in reducing life expectan- United States 78 14.1 8.2 cy in Russia, while the con- tribution of younger age Germany 78 12.6 6.0 groups to overall mortality United Kingdom 77 10.9 6.6 has dropped. Denmark 77 12.9 8.1 Mexico 73 18.0 10.1 Mortality rates among the Poland 70 22.8 8.8 youth and young adults are Turkey 70 21.8 12.0 particularly high in Russia. Brazil 68 25.9 13.6 UNICEF has reported Kyrgyz Republic 66 33.5 29.9 (2004) that Russians aged 15-24 have the highest mor- Source: World Bank World Development Indicators. tality rate in Europe, with 9 Figure 1.10 Male Adult Mortality and Gross National Income per Capita in Selected Countries, 2000 7.0 log) 6.5 (in RUSSIA 6.0 BLR UKR 2000 MDA EST KGZ TKM KAZ LVA TJK HUN GEO AZE ROM LTU tality 5.5 UZB ALB BGR SVK ARM POL mor SVN MKD 5.0 HRV CZE adult 4.5 Male 4.0 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 Source: World Bank World Development Indicators. Note: "PPP" is purchasing power parity. points higher than the rate in Brazil, double that of explain some of the differences in mortality rates observed Turkey, and quadruple that of the United Kingdom. in Table 1.1. However, as Figure 1.10, shows, even when income differences are taken into account, Russian male adult mortality rates are still substantially higher than Life Expectancy and those of countries with similar per capita incomes. Economic Development A major determinant of a population's health is its coun- And, as a result of high mortality rates among the adult try's level of economic development, which may in part population, life expectancy at birth in Russia has Figure 1.11 Is Economic Growth Sustainable in Russia without Good Health? GNI pc Atlas method Life expectancy current US$ at birth 3000 68 Note: 2500 67 When calculating GNI in U.S. dollars 2000 from GNI reported 66 in national curren- cies, the World Bank 1500 follows the Atlas conversion method, 65 1000 using a three-year average of exchange rates to smooth the 500 64 effects of transitory 1998 1999 2000 2001 2002 2003 exchange rate fluctu- ations. Source: World Bank World Development Indicators 2005/WHO/ERO HFA Database 2005. 10 continued to decrease in recent years when Figure 1.12 Life Expectancy at Birth, Males and Females, 1970-2002 the economy has been growing strongly on the back of high oil prices (Figure 1.11). 75 Gender Differences in Life Expectancy 70 (Years) Gender differences in life expectancy occur Female throughout the world. Women generally Birth Male enjoy longer lives than men for a number at 65 of biological and socioeconomic reasons. In 2002, women in the United Kingdom lived 5.1 years longer than men; women in Expectancy 60 France lived 7.5 years longer; and women Life in Japan lived nearly 7 years longer. This gender gap in Russia was 13 years in 2002, almost twice the largest difference else- 55 1970 1975 1980 1985 1990 1995 2000 where. Life expectancy among Russian men peaked in 1964 at 65 years--not much different from other G-8 countries. Source: WHO European Health for All Database. In the late 1960s and throughout the 1970s, longevity stagnated or dropped for Russian men, but Russian women lived 5 to 10 years longer. By 1990, life expectancy had increased to 74 years for Russian women. During the and alcohol consumption, as these behaviors are very transition years in the 1990s, the escalation of mortality different between men and women--even those living rates affected both Russian men and women, but the in the same households. size of the effect was greater for men: they lost 5 years of life expectancy between 1990 and 1994 while Russian Life expectancy in Russia for males and females at 60, women lost 2, keeping the gender gap in life expectancy 65, and 80 years of age is also quite low compared with at about 13 years. Recovery of life expectancy post-1995 that at the same ages in other G-8 countries (Table 1.2 has been slow for both men and women. In 2003, male on page 12). Furthermore, projections for the next 25 life expectancy was only 58 years and that for females years indicate that although Russia will experience was 72 years. The gender gap in life expectancy is improvements in life expectancy in these three age expected to persist in the coming years. As a result of groups, these figures will remain below those expected the gender gap in mortality, the 2002 all-Russian popu- for Russia's G-8 partners. lation census showed that the sex ratio decreased after 1989 (the number of men per 1,000 women fell from Regional Disparities in Mortality and 877 in 1989 to 872 in 2002) (Andreev 2005). Life Expectancy Figure 1.12 shows trends in life expectancy for Russian Mortality rates and life expectancy in Russia vary greatly males and females separately. The large difference by sex by region, in part because of regional differences in suggests that specific behavioral factors are implicated, socioeconomic and health levels. The mortality rate of rather than factors related to the external environment the economically active male population ranges from or adequacy of health care, which affect men and 3.8 deaths per 100,000 people in the region with the women to somewhat similar degrees. As discussed in lowest rate to 17.8 deaths in the region with the high- Chapter 3, two factors that may explain an important est--a nearly fivefold differential (Table 1.3 on page part of this exceptionally large gender gap are smoking 12). For women, the regional differential is even greater. 11 Table 1.2 Life Expectancy at Ages 60, 65, and 80 in Russia and Selected Countries, by Gender, 2000-2005 and 2025-2030 AGE 60 AGE 65 AGE 80 2000-05 2025-30 2000-05 2025-30 2000-05 2025-30 Country M F M F M F M F M F M F Russia 13.5 18.7 16.2 21.4 11.1 15 13.2 17.5 5.8 6.8 6.5 8.2 Ireland 18.3 22.4 20.1 24.6 14.6 18.3 16.3 20.3 6.3 8.2 7.3 9.6 Italy 19.5 24.3 21.2 26.0 15.8 20.0 17.3 21.6 7.0 8.9 7.8 10.2 Canada 20.3 24.5 21.9 26.1 16.5 20.3 17.9 21.8 7.5 9.6 8.3 10.7 France 20.0 25.5 22.1 27.3 16.3 21.2 18.1 22.9 7.2 9.8 8.3 11.2 Germany 18.9 23.7 21.6 26.2 15.2 19.5 17.7 21.8 6.7 8.8 8.1 10.5 Japan 21.4 27.0 23.9 31.1 17.4 22.6 19.7 26.5 7.8 10.8 9.3 14.1 United Kingdom 19.4 23.2 21.6 25.9 15.6 19.2 17.6 21.6 7.1 9.0 8.2 10.6 United States 19.5 23.9 21.7 26.1 15.9 19.9 17.8 21.9 7.3 9.8 8.3 11.0 Source: UN Population Database. Average life expectancy at birth varies by region as well. Table 1.3 Regional Variation in Mortality Rates for the Data for 2001 show that people in regions such as the Economically Active Population, 2000 Republic of Igushetia and Dagestan and in Moscow Deaths per 100,000 population have the longest life expectancy and live 18 years longer than those in regions with the shortest, such as Republic Mortality Rates of Tyva, Koryak Autonomous Okrug, and Komi-Perm Men Women Autonomous Okrug (Table 1.4). Russian Federation's mortality rate 11.5 3.0 Region with the highest mortality rate 17.8 7.2 Chapter 4 provides details of observations indicating Region with the lowest mortality rate that the mortality crisis in Russia seems to be concen- 3.8 1.3 trated in the least developed regions (with the exception Median mortality rate 11.4 2.9 of some republics in the Caucusus which have the coun- Source: Goskomstat 2000. try's longest life expectancy) and mainly affects an underclass of young and middle-aged men with limited Note: Men aged 16 to 59; women aged 16 to 54. educations and vocational skills, who are often unem- Table 1.4 Regional Variation in Average Life Expectancy at Birth in Russia, 1999-2001 AVERAGE LIFE EXPECTANCY 1999 2000 2001 Total Male Female Total Male Female Total Male Female Region with the longest 73.35 68.10 78.57 74.01 68.57 79.03 74.6 70.05 79.08 life expectancy Region with the shortest 56.00 50.72 62.10 56.14 50.41 62.98 56.48 51.06 62.78 life expectancy Source: Goskomstat 2003. 12 Table 1.5 Human Development Index for the low life expectancy of its population. As shown by Selected European Countries, 2002 Barro (1991), based on panel data of around 100 coun- tries for the 1960-90 period, the impact of overall life Country Index expectancy on economic growth is substantial. The Russia 0.795 results from the Barro study imply, other things being United Kingdom 0.936 equal, that a rise in life expectancy from 50 to 70 years Germany 0.925 (about 40%) would raise the economic growth rate by France 0.932 1.4 percentage point per year. Jamison et al. (2004), using data from 53 countries also found that improved Italy 0.920 health status as measured by the survival rate of males Source: UNDP Human Development Indicators. between 15 and 60 years of age accounted for one-tenth of growth over 1965-90. ployed or do manual work, live in urban areas, and come from incomplete families, often with migrant or Population Migration ethnic minority backgrounds. Statistics on migration patterns in Russia show that migration's role in its population dynamics has been Impact of Life Expectancy on important, but its role in offsetting the negative growth Human Development in population size has been of significance only for lim- ited periods in the past 25 years. Prior to 1992, migra- Russia lags behind European countries in terms of its human development index1 (Table 1.5), in part due to tion was relatively low, averaging less than 130,000 per- sons between 1985 and 1992. The transition years expe- rienced a very rapid increase in population migration 1 The Human Development Index, developed by the United such that net migration increased rapidly from about Nations Development Program (UNDP), is a composite statisti- 176,000 persons in 1992 to a peak of 800,000 persons cal indicator using life expectancy at birth, adult literacy, and real gross domestic product (GDP) per capita in purchasing power in 1994 and remained at levels slightly above those of parity (PPP) in U.S. dollars. pre-transitional years until about 1999 (Figure 1.13). Figure 1.13 Net Migration, 1985-2002 1200 1000 Net Migration in 000 800 Natural Increase 600 400 200 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 -200 -400 -600 -800 -1000 -1200 Source: Heleniak (2005), citing material from Goskomstat. 13 Figure 1.14 The Population Structure, 2005 and 2020 Russia 2005 Russia 2020 2.73 6.96 2.58 6.06 70-74 2.44 3.87 70-74 3.33 4.92 4.75 6.52 5.32 6.83 60-64 2.69 3.25 60-64 7.04 8.11 5.82 6.23 7.12 7.53 50-54 7.59 7.63 50-54 6.49 6.30 8.78 8.44 7.39 7.08 40-44 8.12 7.51 40-44 7.96 7.62 6.96 6.12 9.05 8.40 30-34 7.62 6.75 30-34 9.34 8.38 7.99 7.18 6.26 5.42 20-24 8.92 7.85 20-24 5.61 4.76 9.06 7.79 5.88 4.90 10-14 5.95 5.01 10-14 5.95 4.92 5.22 4.39 5.34 4.40 0-4 5.38 4.51 0-4 5.32 4.37 10 5 0 5 10 10 5 0 5 10 MALES FEMALES MALES FEMALES Percentage Percentage Source: UN Population Database. At the same time, the peak in the natural increase of As shown in Figure 1.14, Russia's population structure is population (excess of births over deaths), which had already characterized by a shrinking youth base and an occurred in the mid-1980s, experienced a very rapid expanding proportion of the population aged 60 and decline up to the mid-1990s when the trend seems to over. Two factors, (a) continued very low levels of fertili- have leveled off at a deficit of about 600,000 people a ty and (2) past higher fertility levels that produce com- year. Data show that prior to 1992, before the return paratively larger cohorts, combine to result in a dispro- migrations of Russians and Russian speakers from the portionately high number of older persons in compari- former Soviet Union swelled, net migration made only a son to younger ones. limited contribution to the natural population increase, which averaged about 800,000 a year for the years Two decades ago, youth aged 0-14 years constituted 1985-90. about a quarter of Russia's population, and those aged 60 years and above were only about 14 percent of the It is clear, then, that net migration made an important total. Now, those aged 0-14 have dropped to 18 per- difference in population dynamics in the mid-1990s, cent--a level nearly equal that of the population 60 and when the natural increase of population reached the over. Projections suggest that given Russia's expected bottom; migration was highest at this time. However, total fertility rates of between 1.1 and 1.3 children per the mid-1990s migration levels have reverted to pre- woman of reproductive age for the years 2005-25, per- transitional years; and without the positive natural sons aged 0-14 will remain at about 13 percent, popula- increase of population of the mid-1980s, the alleviation tion growth rates will remain negative averaging in population deficits has been considerably reduced. between -0.6 and -0.8. As a result, the proportion of persons aged 60 and over will increase to more than a quarter of the Russian population. The extreme and The Aging of the Population rapid aging of the Russian population is occurring at a A country's population structure by age groups is direct- low level of GDP per capita (Eberstadt 2005). ly influenced by fertility and mortality trends. A coun- try with high fertility rates will generally have a large The U.N. Population Division uses an aging index to proportion of its population in the lower age groups. measure how fast a population is aging by comparing, Falling fertility rates add progressively smaller propor- over time, the number of persons 60 and over to that of tions of lower age groups to a population. 100 young people aged 0-14 years. In 1990, the ratio in 14 Russia was approximately 0.7-that is, there were 70 peo- What are the Consequences of ple aged 60 and over for every 100 persons aged 0-14 the Demographic Decline and the years. In 2015, the ratio will rise to 1.55--that is, there Health Deficit? will be 155 people aged 60 and over for every 100 per- sons 0-14. By 2050 the ratio will rise to 2.0--there will The unprecedented mortality upsurge Russia experi- be 200 people aged 60 and over for every 100 aged 0-14 enced in the 1990s is the product of a a long-term dete- (Figure 1.15). rioration of the health of the population associated with a prolonged period of highly unhealthy lifestyles and adverse environments, aggravated by a difficult socio- Total Dependency Ratios economic and political adjustment process (Cornia Dependency ratios give a rough estimate of what the 2002, Mesle 2002). Russia's total population is expected potential social support needs or burden of dependency to decline in the years ahead and will be accompanied is in a particular country. The main assumption embod- by the aging of society as fertility rates remain under ied in this measure is that everyone aged over 65 and the replacement level. This has several important those under 15 are likely to be dependent on those in implications: the 15-64 age group, who are likely to be working adults. Total dependency ratios are calculated as a ratio Fewer workers. If trends persist, the size of the Russian of the sum of the population aged 0-14 plus that aged labor pool will continue to shrink significantly in the over 65 to the population aged 15-64; the ratios are pre- decades ahead. But population decrease may also result sented as number of dependants per 100 persons of from the changing sex ratio that will probably lead to a working age (15-64). new fertility decline in the future--an extreme sub- replacement fertility level (Andreev 2005). This presents Data for the two decades before 2000 show that a serious threat to Russia's development as the corre- dependency ratios in developed countries remained sponding decline in the share of working-age adults due more or less flat, ranging between 45 and 55 depend- to high mortality and the increasing proportion of eld- ents per 100 people aged 15-54. Russia's total depend- erly people could adversely impact the economy. ency ratio up until 1995 was in the middle of the range, i.e., between 48 and 50. The pro- jected trend for developed coun- tries after 2005 is for significant Figure 1.15 Aging Index: Number of Persons Aged 60 Years and Over increases in their total dependency Compared to That of 100 People Aged 0-14 Years, 2005-25 ratios caused mainly by the increase in the proportion of the 210 population in the 65-plus-age 200 bracket. Japan, one of the world's 190 oldest societies, is projected to see 180 its total dependency ratio rise by 170 about 50 percent in the next 20 years. Russia, on the other hand 160 will see its total dependency ratio 150 fall to below 40 in 2010 (from 50 140 in 1995), but then the ratio will 130 accelerate to exceed 50 in 2025, an increase of about 30 percent in 120 a 15-year period, at a lower per 110 2005 2010 2015 2020 2025 capita income level than in other G-8 countries. Source: UN Population Database. 15 According to IMF projections (IMF 2004), a decrease widows in Russia at ages 30-44 is about four times the in working-age population will reduce labor productivi- share in the United States). ty and incentives for investment in human and physical capital, which will in turn reduce per capita GDP Growing regional disparities. Uneven fertility, mortali- growth. Government budgets will be squeezed as tax ty, population growth, and life expectancy in different revenues fall because of a decline in the size of the work- regions and among social and ethnic groups could exacer- ing-age population and an increase in the needs and bate existing disparities. Difficulties in providing equitable demands of an aging population. As the elderly popula- access to resources for the poor may give rise to social and tion rises, the overall rate of saving and investment in a political challenges, especially in a country as vast as society will decline as more resources need to go to pen- Russia. This is an important policy consideration for any sions, health care, and long-term residential care. government, particularly when resources are scarce. However, a healthy population of 65-75 year-olds may National security risks. From a national security be a sizable untapped workforce, and, providing meas- point of view, the demographic and health crisis in ures are taken to reverse the mortality trends and assum- Russia will present many challenges (Twigg 2004): (a) ing lives lengthen overall, increasing the retirement age the number of men around conscription age will plunge becomes an option to ensure the medium-term solvency rapidly in the decades ahead; (b) a growing percentage of pension systems. This situation may also require the of the military budget will have to be allocated for the development of new work relationships with older peo- provision of medical, nutritional, and substance abuse ple, including developing mechanisms to attract and programs for draftees and soldiers that are deemed med- retain older workers, especially knowledge workers ically unfit for duty; (c) long-term economic growth will (Drucker 1999). depend on large cohorts of healthy and skilled young and middle-aged adults; and (d) if its vast territory is The destabilization of families. The gender gap in depopulated, instability could grow and the country life expectancy has led to instability in marriage and an could become increasingly difficult to govern. extremely high proportion of widows (the percentage of 16 2 Chapter 2. The Burden of Ill Health and Disability I ll health and disability measures have been devel- Iceland to 51.6 in Turkmenistan; for females it was oped to capture the full health status of a popula- from 75.9 in San Marino to 56.4 in Tajikistan (WHO tion and supplement mortality measures. This chap- European Health for All Database). Compared to other ter assesses the estimated years that Russians can expect developed countries, Russia has one of the lowest to live without disability or in less than full health, as healthy life expectancy rates. Table 2.1 shows that in compared with experiences in other countries. These Russia, HALE is 64 for females, about 10 years less measures add to the mortality information given in the than in France and Germany, but only 53 years for previous chapter to enable a closer examination of the males, 16 and 18 years less, respectively, than in the number of healthy, productive years people have and, United Kingdom and Italy. Russia's gender gap is one of subsequently, of prospects for economic growth. the widest in HALE worldwide and reflects the sharp increase in Russian adult male mortality in recent decades. Healthy Life Expectancy Morbidity data contain important information not evi- These data portend a difficult future, not only for aging denced by mortality data alone. The World Health Russians themselves, but also for an economy that Organization (WHO) uses healthy life expectancy needs a productive workforce and will likely rely on (HALE) as a summary measure of health to capture the older working adults to economically support those full health experience--not just mortality--of a popu- who cannot work. The differences for men, who lation (WHO 2002a). HALE is most easily understood comprise the larger portion of the workforce, are partic- as lifespan in full health, i.e., without disability. ularly skewed. In considering data on healthy life expectancy, one While a lower life expectancy is usually associated with finds larger differences between Russia and Western a lower HALE, large variations exist in HALE at differ- Europe than in considering data on life expectancy ent age groups. As shown in Table 2.2 on page 20, the alone. In 2002, HALE at birth in the European region difference in HALE is especially pronounced for (where overall health levels are among the world's high- healthy life expectancy at age 20. In addition, data on est) ranged from 73.4 years in San Marino to 54.4 in healthy life expectancy reveal that both Russian women Turkmenistan. The range for males was from 72.1 in and men bear an enormous burden of ill health, partic- ularly during their working-age years. Male healthy life expectancy is consistently lower Table 2.1 Healthy Life Expectancy (HALE) at Birth in in Russia than in either Eastern or Western Russia and Four European Countries, 2002 Europe at ages 20, 40, and 65. The largest difference for males occurs at age 20 Country Males Females Total between Russia and Western Europe, and Russia 53 64 59 the smallest difference at age 65 between United Kingdom 69 72 71 Russia and Eastern Europe. The pattern Germany 70 74 72 reflects higher mortality rates for Russian France 69 75 71 men at younger ages. The gap in life expectancy between males and females in Italy 71 75 73 Russia is largest at younger ages but con- Source: WHO European Health for All Database. verges to European averages later in life. 19 Table 2.2 Life Expectancy and Healthy Life Expectancy at Ages 20, 40, and 65 in Russia, Eastern Europe, and Western Europe, 2000 AT AGE 20 AT AGE 40 AT AGE 65 Sex Country/Region e(x) h(x) e(x) h(x) e(x) H(x) Male Russia 41.9 36.7 22.4 17.3 11.4 6.7 Eastern Europe 49.1 41.9 26.6 20.5 12.7 8.3 Western Europe 54.5 50.4 31.2 27.6 15.0 12.5 Female Russia 54.2 40.6 31.1 18.5 15.2 5.8 Eastern Europe 56.8 44.5 32.8 22.7 15.9 9.3 Western Europe 60.2 53.7 36.0 30.3 18.1 14.0 Female-male gap Russia 12.3 3.0 8.7 1.2 3.9 -0.9 Eastern Europe 7.6 2.6 6.2 2.2 3.3 1.1 Western Europe 5.7 3.3 4.8 2.7 3.1 1.5 Source: Adapted from Andreev, McKee, Shkolnikov 2003. Note: Health expectancy is calculated by Sullivan's method (Sullivan 1964; Robine et al. 1993). The expression "e(x)" means life expectancy, and "h(x)" means healthy life expectancy. The rapid decline in healthy life expectancy in Russia is Although the stress of the 1990s transition has affected mainly due to the high probability of death for men and the health status of the Russian population, particularly to poor health for women. Russian men aged 40, for among men, the exact psychological pathways have not example, have 10.3 years less of healthy life expectancy been established. While Russian women survive longer than Western European men. Russian women in this than men, they are generally in worse health than age group have 11.8 years less of healthy life expectancy women in Eastern and Western Europe. due to poor health than Western European women. Comparison of Disability Levels Figure 2.1 Russia and Sweden Survival without Disability at Different Ages The 2003 NOBUS (National Survey of 1.0 Household Welfare and Program Participation) survey estimates the num- ber of disabled people in Russia at more 0.8 than 9 million or about 7 percent of the Disability population (NOBUS 2003; Baskakov and 0.6 Yenenko 2005). In looking at the rates of Without disability in communities in Russia and 0.4 Sweden, Bobak et al. (2004) compared Alive RUSSIA SWEDEN official life table data, self-rated health tion Men Men status, and physical functioning surveys 0.2 Women Women opor measuring limitations in everyday activi- Pr ties (e.g., shopping, bathing, walking). 0 The results paint an alarming picture of 45-49 50-54 55-59 60-64 65-69 70-74 disability in middle-aged and older men Age Group and women in Russia. The combination of high mortality and high disability leads Source: Adapted from Bobak et al. 2004. 20 to a large difference between the two countries in the In another study, Plavinski, Plavinskaya, and Klimov estimated number of years without disability in people (2003) examined the association between social factors in their late 40s to mid-70s. They also show that the and the sharp increases in mortality and morbidity in problem of disability affects more people in their prime Russia in the 1990s. They found that men in lower working age, with likely impacts on economic outcomes socioeconomic groups were most affected by the increas- at the individual and aggregate levels. es in mortality and ill health. The most pronounced dif- ferences were among people with the lowest level of For both men and women, the 2004 Bobak study education. Furthermore, these patterns held for heart reports that health and physical functioning declined diseases and deaths from cancer. There was no recorded much faster in Russia than in Sweden. Figure 2.1 increase in mortality throughout the 1990s among men depicts the proportion of people ages 45-49 to ages with university degrees. 70-74 both alive and without disability relative to peo- ple ages 45-49. In 25 years, 65 percent of healthy mid- dle-aged Swedish men but only 17 percent of Russian Policy Implications men would be alive without disability; for women, the This chapter adds the burden of ill health and disability comparable figures are 65 percent of Swedish women in Russia to the mortality information in the previous versus 22 percent of Russian women. A healthy middle- one to offer a fuller picture of health trends and their aged cohort in Russia would have less than a third the implications for working-age adults and the future of chance that Swedes have to survive into old age without the economy. This chapter also clarifies that the chal- disability. lenge for Russia is not only to increase life expectancy by reducing mortality but also to help people stay Figure 2.2 sharply contrasts the total survival and sur- healthy into older age. Indeed, it is clear from the vival without disability to age 70-74 in cohorts without European data that with improved health behaviors not disability in Russia and Sweden. The most striking dif- only do people live longer, but their quality of life is ferences between these countries stem primarily from the high death rate for Russian men and high disability rate for Russian women. Figure 2.2 Russia and Sweden: Total Survival and Survival without Data from the Russia Longitudinal Disability to Age 70-74 in Cohorts without Disability Monitoring Survey, which has been con- at Age 45-49 ducted annually since 1992, also indicate that rating one's own health as less than 1.0 With disability Without disability good became more common in all age groups during the 1990s (Palosuo 2003). 0.8 However, according to data from the 2003 NOBUS survey, women in all age groups report a worse self-health assess- 0.6 ment than men (while the average self- health assessment among men was 3.16 0.4 points in all respondents on a 5-point scale, it was as low as 2.96 for women). 0.2 Given the lower life expectancy and HALE among men, these surprising results are explained by psychological and 0 behavioral factors: women are more con- Russian Swedish Russian Swedish Men Men Women Women cerned about their health than men, more often seek medical advice and care, and do Source: Adapted from Bobak et al. 2004. not neglect their diseases until they reach an advanced stage (Korkhova 2001). 21 also improved by the reduction of sickness and disability the high levels of mortality and ill health among these at the end of life. A response in Russia would also need working-age adults. The low levels of life expectancy to include regular monitoring of indicators such as and healthy life expectancy, as well as high disability lev- HALE by gender, age, region, and social groups for a els, are driven primarily by the high level of mortality better understanding of the population's health status. and ill health among working-age adults, particularly males. Chapters 3 and 4 examine the primary causes of It is widely accepted in the research community that the the high level of mortality among working-age adults-- exceptionally high levels of mortality and ill health in non-communicable diseases and injuries--and their Russia are not an artifact of poor data. Much work has associated risk factors. been done to shed light on what factors contribute to 22 3 Chapter 3. Determinants of the Demographic Decline and the Health Deficit T he rise of noncommunicable diseases (NCDs) This chapter reviews the NCDs and injuries that cause and injuries presents a major challenge to global most of the deaths and ill health in Russia on the bases development, as it threatens countries' economic of estimates of mortality data (death is still the most and social well-being. NCDs are a set of chronic dis- extensively and reliably recorded vital event in Russia) eases, including among others cardiovascular disease, and of disability-adjusted life years (DALYs). This chap- cancer, chronic respiratory diseases, mental disorders, ter also presents a detailed assessment of Russia's three and diabetes, that are characterized by a long latency leading killers: CVD, cancer, and injuries. period, prolonged clinical course, and debilitating mani- festations. As NCDs usually have long incubation peri- Deaths and DALYs Lost Attributable ods, with symptoms appearing after 5 to 30 years of to Leading Causes exposure to lifestyle and environmental risk factors, the future disease burden will be determined by the preva- The relative burden of NCDs and injuries as leading lence of key risk factors occurring today coupled with causes of morbidity and mortality is now larger than the aging of a population. However, as will be discussed that of infectious diseases. The distribution of the 10 later in this report, acute alcohol consumption plays a leading causes of death in Russia in 2002 is presented in major role in accelerating the onset of NCDs and con- Table 3.1. It shows that NCDs and injuries account for tributing to the increase in mortality due to external 68 percent of total deaths. In 2003, among the work- causes in Russia. Evidence from different countries indi- ing-age population, CVDs, cancer, and injuries account- cates that reversal of risk factors even late in life reduces ed for about 78 percent of deaths and 15.2 million lost NCD risk (especially for cardiovascular diseases [CVDs] years of potential life (10.3 million among working-age and diabetes) within a relatively short period (2-5 years). men and 4.9 million among women of the same age) Table 3.1 Deaths and DALYs Lost Attributable to the 10 Leading Diseases and Injuries, 2002 Rank Cause Total deaths Total % Rank Cause Total DALYs lost Total % 1 Ischemic heart disease 711,571 29.6% 1 Ischemic heart disease 5,472,308 13.9% 2 Cerebrovascular disease 533,675 22.2% 2 Cerebrovascular disease 3,930,367 10.0% 3 Poisonings 66,930 2.8% 3 Unipolar depressive disorders 1,574,695 4.0% 4 Self-inflicted injuries 59,015 2.5% 4 Violence 1,459,927 3.7% 5 Trachea, bronchus, lung cancers 58,899 2.4% 5 Self-inflicted injuries 1,297,152 3.3% 6 Violence 47,461 2.0% 6 Road traffic accidents 1,292,752 3.3% 7 Road traffic accidents 44,580 1.9% 7 Poisonings 1,272,366 3.2% 8 Stomach cancer 44,557 1.9% 8 Alcohol use disorders 1,258,936 3.2% 9 Colon and rectal cancers 38,141 1.6% 9 Hearing loss, adult onset 765,988 1.9% 10 Cirrhosis of the liver 37,426 1.6% 10 Tuberculosis 700,997 1.8% ALL CAUSES 2,405,721 100.0% ALL CAUSES 39,409,946 100.0% Source: WHO-EURO (2005). The European Health Report (http://www.euro.who.int/eprise/main/who/progs/ehro5/home). 25 Table 3.2 Cause-Specific Adult Mortality Like HALE, which was discussed in the previous chap- Rates, Age 15-64 ter and is based on life expectancy at birth plus an adjustment for time spent in poor health, disability- Deaths per 100,000 population adjusted life years, or DALY, is an indicator used to sup- Russia's death rates plement mortality data as it captures the full disease Russia EU-15 as % of EU-15 burden. DALYs are calculated by summing the years of NCDs 605 206 294% life lost due to premature death (defined as the differ- Injuries 281 58 484% ence between the actual age of death and the life expectancy at that age in a low-mortality population) Source: WHO Mortality Database. and the years of productive life lost because of a disabili- Notes: Russian rates refer to 2002; EU-15 rates to 2001 or ty (Murray 1993; Murray and Lopez 1993).2 latest available. The EU-15 average is population weighted. Table 3.1 shows that in 2002 the Russian population lost an estimated total of 39.4 million DALYs due to all causes. The NCDs and injuries account for close to 50 (Oganov and Maslennikova 2005). CVDs (heart attack, percent of DALYs lost from all causes. This table also myocardial infarction, acute coronary syndrome, con- gestive heart failure, stroke, kidney disease, and periph- 2 eral vascular disease) alone account for over half of all More specifically, DALYs lost to mortality are calculated by sum- ming the discounted value of years lost to premature death across deaths. Infectious, respiratory, and parasitic diseases all causes and age groups. DALYs lost due to disability are based account for less than 10 percent of all deaths. on the incidence and duration of various types of disability mul- tiplied by a severity weight that accounts for the severity of the As shown in Table 3.2, the mortality rates from NCDs disability. Total DALYs result from the sum of DALYs lost to mortality and disability, adjusted by a discount rate so that future and injuries in Russia are dramatically higher than those years of healthy life are valued at progressively lower rates and by in the European Union countries: about 3 and 5 times, age group weightings so that years of life lost at different ages are respectively. given different relative values. Table 3.3 Regional Variation in Mortality Rate by Cause of Death, 2000 Deaths per 100,000 population Infectious Diseases of Digestive Accidents, and parasitic Cardiovascular respiratory tract poisoning, diseases Tuberculosis Cancer Diseases system diseases and injuries Russian Federation 25.0 20.6 205.5 849.4 70.5 44.6 219.9 Maximum Region with the highest 92.2 80.3 282.2 1,338.4 134.0 564.2 444.7 mortality rate Region with the lowest 4.7 2.3 56.9 198.6 9.3 9.7 37.3 mortality rate Median 22.9 19.0 189.8 719.6 66.3 43.4 53.8 mortality rate Source: Goskomstat (2000). 26 shows that injuries cause considerably less burden overall Table 3.4 Estimated Mortality from NCDs and Population among the 10 leading causes; however, they are very rele- Reduction, 2000-2030 vant for some population groups, particularly the young. Estimated mortality Reduction in Most NCDs and injuries are largely preventable, as their Years from NCDs population size main risk factors are well known and their occurrence 2000-2005 2,194,000 -4,059,000 can be influenced by the implementation of tested public 2005-2010 2,260,000 -4,052,000 health interventions (Beaglehole and Yach 2003). 2010-2015 2,390,000 -4,072,000 Regional variations. Rates of mortality due to NCDs 2015-2020 2,515,000 -4,411,000 and injuries vary significantly by region in Russia as 2020-2025 2,518,000 -4,590,000 shown in Table 3.3. In the less-developed regions, with 2025-2030 2,630,000 -4,715,000 the highest mortality, the rates for NCDs and injuries are much greater than the infectious diseases and tuber- Source: Bakilana 2005. Calculations on the basis of WHO data and UN culosis rates, and the rate for CVDs is much greater Population Database. than those for any other diseases and injuries. Mortality Projections Due to NCDs and Injuries Diseases of the Circulatory System Table A1 in Appendix A provides the projected numbers Diseases of the circulatory system, a group of diseases of deaths caused by 12 major categories of diseases (plus that affect the heart, brain, and circulatory system, pri- conditions related to pregnancy) in Russia over the marly include cerebrovascular accidents or stroke and 2000-2025 period if year 2000 age-specific death rates coronary heart diseases (CHDs). The most important (ASDRs) had remained the same. Overall, three cate- modifiable risk factors, termed primary risk factors, that gories of diseases--CVDs, cancer, and injuries-are can independently produce clinical complications due to expected to cause between 80 and 90 percent of deaths cardiovascular arteriosclerotic diseases are dietary intake, by 2025, up from 68 percent in 2002. Table 3.4 sum- failure to treat hypertension, high cholesterol, and marizes estimates for total mortality from these cate- smoking. In Russia, alcohol consumption does not seem gories for six chosen time intervals to be between 2.2 to be associated with reduced risk of diseases of the cir- and 2.6 million deaths per year, resulting in a net reduc- culatory system, and its abuse is plausibly the main tion of the population of up to 4.7 million by 2030. explanation for their observed fluctuations since the Overall, due to the population's aging as discussed in mid-1980s (details are in chapter 4). Other risk factors chapter 1, the contribution of NCDs and injuries to like obesity, diabetes, and sedentary lifestyle contribute total mortality in Russia would continue to grown to circulatory disease morbidity and mortality in con- unless comprehensive prevention activities are undertak- junction with primary risk factors. en to reduce the prevalence of the key risk factors for these diseases. Russia's cardiovascular disease (CVD) death rate per 100,000 population in 2002 was 994--one of the world's highest. The comparable rates were 317 in the The Leading Individual United States, 363 in Portugal, and 225 in Brazil (WHO Global Infobase Online, 2005). CVD accounts Causes of Death for almost 52 percent of deaths in Russia, compared to Information on the existing and growing burden of 38 percent in the United States, 42 percent in Portugal, NCDs and injuries in Russia is still mainly based on and 32 percent in Brazil. Russia's high mortality among mortality data, which cannot adequately reflect disease working-age men is mainly attributable to CVD. Such outcomes in terms of the burden of ill health and disabili- statistics led a 2004 World Bank report to conclude ty on individuals, health systems, and society as a whole. that Russia could gain 6.7 years in life expectancy by The mortality caused by Russia's three leading killers-- working to match the European Union's CVD mortality CVDs, cancer, and injuries--is discussed in this section. rates (World Bank 2004a, p.29). 27 Figure 3.1 Cardiovascular Disease Mortality Rates in Russia as a Percentage of Swedish Rates 1400% 1200% 1000% 800% 600% 400% 200% 0% 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Source: WHO Mortality Database. are more often prevalent among people over 60, The social consequences of this high toll of avoidable working-age Russians are also affected. The Novosibirsk mortality are great. Despite a widely held view that Stroke and Myocardial Infaraction Registries in the age NCDs primarily impact people beyond retirement age, groups of 25 and 64 years showed that stroke and CHD a crucial aspect of the Russian situation is the young age rates are somewhat similar (about 4 and 3 cases per at which people are affected. Although stroke and CHD 1,000 population of these ages, respectively) (Suslina 2005). Moreover, it was estimated that dis- eases of the circulatory systems account for 30 percent of all deaths among working- age Russians. Figure 3.1 illustrates this Figure 3.2 Trends in Standardized Death Rates from Diseases of the point by displaying the ratio of mortality Circulatory System in Russia and EU-15 Countries in Russia from CVD in different age Per 100,000 groups to that in Sweden (Suslina 2005). While the death rate is two- to three-times 900 higher in older ages, it is a remarkable 12 800 times higher at 30-34 years. 700 Figure 3.2 shows Russia's increasing trend in death rates of diseases of the circulatory 600 Russian Federation system against the EU-15's decreasing EU members before trend. This figure clearly shows that con- 500 May 2004 trary to most other developed countries the rate in Russia is rising. 400 Figure 3.3 shows a similar increasing trend 300 in the standardized death rate from cere- 200 brovascular diseases in Russia and a 1970 1980 1990 2000 2001 decreasing trend in the EU-15 countries (albeit a less pronounced decrease than Source: WHO/EURO HFA Database. seen in figure 3.2). The death rate from Note: "CIS" is Commonwealth of Independent States. 28 cerebrovascular diseases in Russia Figure 3.3 Trends in Standardized Death Rates from Cerebrovascular Diseases in at 306 per 100,000 population Russia and EU-15 Countries, All Ages is about five times that of the Per 100,000 EU-15 countries (62 deaths per 100,000). In Russia, acute stroke 400 leads to 35 percent of deaths; the European Stroke Council believes it is possible to decrease 300 this indicator to 20 percent or lower (Suslina 2005). 200 Russian Federation EU members before Cancer May 2004 Cancer is a group of diseases 100 characterized by the uncon- trolled growth and spread of abnormal cells. Many cancer 0 deaths are preventable, either 1970 1980 1990 2000 2010 corresponding to tumors suscep- Source: WHO/EURO HFA Database. tible to primary prevention (such as those associated with smoking) or susceptible to screening and early treatment (like cervical cancer). Diet 3.5 on page 30), which show a lower and steadily (including alcohol use) is an extremely important factor decreasing trend. in terms of both cancer prevention and cancer risk. Lung Cancer. Russia's mortality rate per 100,000 pop- Cancer mortality rates in Russia are significantly above ulation for cancer of the trachea, bronchus, and lungs the average of the EU-15 countries (Figure 3.4). In ranged from 34.5 deaths in 1980 to 36.5 in 2002 with a addition, cancer in Russia is characterized by its extremely high lethality, with a high pro- Figure 3.4 Trends in Standardized Death Rates for All Malignant Neoplasms portion of deaths within a year (Cancer) Combined for Ages 0-64, in Russia and EU-15 Countries, of first diagnosis of the disease 1980-2002 (e.g., 56 percent for lung cancer Per 100,000 and 55 percent for stomach can- cer) (Goskomstat 2004). Men in 130 Russia die from cancer twice as often as women, but the cancer 120 incidence rate among women is higher. Cancer deaths reduce the 110 average life expectancy in Russia by about 100 2 years. Russian Federation 90 EU members before May 2004 Cervical Cancer. The cervical cancer mortality rate per 80 100,000 women aged 0-64 in Russia is far higher that the rate 70 1970 1980 1990 2000 2010 of the EU-15 countries (Figure Source: WHO/EURO HFA Database. 29 Figure 3.5 Trends in Standardized Death Rates for Cervical peak of over 45 deaths in the early 1990s. These Cancer for Women Aged 0-64 in Russia and mortality rates are quite similar to those reported for EU-15 Countries European countries, where the rates for these can- cers were 35.6 in 1980 and 35.7 in 2002. WHO Per 100,000 data show a small difference between lung cancer 9 death rates in Russia--25.9 per 100,000 people aged 0-64--and EU-15 at 21.8 and 20.3, respec- 8 Russian Federation EU members before tively. Figure 3.6 indicates a declining trend in stan- 7 May 2004 dardized death rates for lung cancer among males of all ages in Russia and in the EU-15 countries. 6 However, the incidence of lung cancer is related to the cumulative effect of smoking over people's life 5 spans, and age-period-cohort analyses show that the 4 downward trend in Russia's death rates from lung cancer is partly due to a cohort effect (i.e., the indi- 3 viduals contributing to the peaks reached the age of 65, when age-specific death rates are highest for 2 lung cancer). It is expected that the favorable-look- 1 ing trend will soon reverse (Shkolnikov et al. 1999). 1970 1980 1990 2000 Source: WHO/EURO HFA Database. Injuries, Poisoning, and Violence Injuries cause more than 5 million deaths annually worldwide. They are classified as either unintention- al or intentional: intentional ones are caused by Figure 3.6 Trends in Standardized Death Rates for Trachea, deliberate violence, and unintentional ones by Bronchus, and Lung Cancer in Russia and EU-15 events such as traffic accidents, falls, fires, asphyxia- Countries, Males, All Ages, 1980-2002 tion, drowning, accidents involving firearms, and Per 100,000 other external causes. Although the discussion that follows focuses on injuries that result in death, the 110 number of deaths is small in com- parison to the number of sur- Russian Federation vivors of violence and injuries, 100 EU members before many of whom spend weeks in May 2004 hospitals and are often perma- nently disabled. 90 The excessive death rate among Russian adult men is significantly influenced by the impact of external 80 causes. Death rates due to these causes are estimated at 463 per 100,000 among men or about five times 70 the rate for women (97 per100,000). In Russia, as discussed later in this chapter, younger age groups suffer disproportionately from injuries. 60 1970 1980 1990 2000 Traffic injuries: Traffic injuries include crashes Source: WHO/EURO HFA Database. involving motor vehicles, pedestrians, or cyclists. At 20.6 deaths per 100,000 population, Russia's mor- tality rates due to traffic injuries compare unfavor- 30 Table 3.5 Mortality Rates Due to Motor By a second and more meaningful measure, fatalities per Vehicle Traffic Injuries in Selected 10,000 vehicles, Russia's rate of 12.2 is exceptionally Countries, 1998-2002 high: 50 percent higher than the second highest (the Republic of Korea at 8.2) among the reporting countries. Per 100,000 Mortality rate The severity of road accidents is often measured by the due to motor vehicle proportion of fatalities to serious injuries. Here, Russia Country traffic injuries is similar to the other transition countries, with an Russian Federation 20.6 index of 0.12--that is, for every eight people injured in Commonwealth of Independent road accidents, one is killed. But Russia's index for the 15.1 States (12) average severity of accidents is 5 to 10 times higher than that in France 12.6 most developed countries (which may result partly from Italy 12.0 poor registration of road crashes not leading to deaths or injuries). European Union (25) average 11.0 Ireland 10.4 However the problem is measured, Russia clearly has a Germany 8.8 serious problem of road fatalities and injuries. The Finland 7.5 Transport Strategy for the Russian Federation, in which Netherlands 6.6 developing safety standards for transport operations is United Kingdom 5.7 seen as one of the principal spheres of government Sweden 5.5 responsibility, recognizes this problem. Source: WHO 2005. Suicide. Figure 3.7 shows the standardized death Note: Data are 1998-2002 or latest available. rates from suicide and self-inflicted injury for all ages ably with other G-8 countries at Figure 3.7 Trends in Standardized Death Rates for Suicide and Self-Inflicted 11 (Table 3.5). This nearly dou- Injuries in Russia and EU-15 Countries, Ages 0-64, 1980-2002 ble rate is even more remarkable Per 100,000 considering that there are fewer automobiles per capita in Russia 45 than in Western Europe. Russia's traffic mortality rate is also high- 40 er than that of other former 35 Soviet states. In 2004, more than 34,000 people in Russia died in 30 road accidents. Most of these 25 deaths occurred among males of Russian Federation prime working age. Driver fac- EU members before 20 May 2004 tors are the major cause of road 15 accidents--alcohol intoxication, lack of discipline, and aggressive 10 driving. In addition, nearly half of all motor vehicle-related 5 deaths in Russia victimize pedes- 0 trians, not drivers or passengers. 1970 1980 1990 2000 Source: WHO/EURO HFA Database. 31 per 100,000 people, comparing Russia with the EU while intoxicated in 2002 was about 30 percent and 12 countries for the period 1980-2002. Russia's suicide rate percent, respectively. In a majority of cases, victims of is much higher than in the EU countries. The Russian homicide are also often intoxicated at the time of the rate peaked in the mid-1990s, particularly among mid- crime (Chervyakov et al. 2002). Gorbachev's 1984-1988 dle-aged men. By 1994 the suicide rate for Russian men anti-alcoholism campaign produced a decline in mortal- aged 50-54 was over six times that in the United States ity due to violent deaths, particularly mortality from (139 and 22.5 deaths per 100,000 population, respec- homicide and road traffic accidents, but with the tively). It then fell slightly between 1995 and 2002. In increase in alcohol consumption among young cohorts, all the Central and Eastern Europe and CIS countries, violent deaths began to rise again after 1998 Russian males aged 15-19 had the second highest sui- (Shkolnikov and Nemtsov 1997; Mesle 2002). Increased cide rate at 38.2 per 100,000 relevant population, barely levels of violence not only add to the increasing mortali- surpassed by Lithuania at 38.4. ty burden, particularly among the young, but it adverse- ly affects social and economic development. Alcohol Poisoning An important external cause is accidental alcohol poi- Injuries among Youth soning: death occurs when, for example, a healthy adult The rate of violent deaths among young Russians is male of average weight consumes a large amount (half alarming, as all forms of violent deaths are rising dramat- liter) of vodka or a similarly strong beverage (40 percent ically among Russians aged 15-29. The proximal causes alcohol) without food in less than an hour. Alcohol poi- of violent deaths seem to be the use of alcohol and illegal soning is one of the most alarming phenomena related drugs. In 2001, over 41,000 people in Russia died of to alcohol abuse in Russia (Tremil 1997); this is a par- alcohol poisoning. This number likely includes some ticularly acute and rapidly growing problem in rural teenagers, as more and more teens are reporting having areas, where the death rate is 128 per 100,000 among been drunk: an increase of 18 percent between 1993 and adult men. Largely due to the poorer quality of alcohol, 2002 (UNICEF 2004). Figure 3.8 shows injury mortali- the death rate in rural areas is twice that of the country ty rates in Russia as a percentage of rates in Sweden and as a whole. clearly illustrates Russia's high burden of injury mortality among the young compared to Sweden's. The observed changes in death rates from suicides and alcoholism in Russia appear to be linked with the social Russia's young people are its future adults, and youths' and economic turmoil of recent years. For example, health patterns, unless redirected, will largely be carried between 1995 and 1998, a period of relative well-being, through life. Young people's health issues are largely deaths rates from these causes decreased in both urban rooted within the country's social conditions. UNICEF and rural areas, while they started to grow continuously cites poverty, unemployment, insecurity, and substance after the 1998 crisis. abuse as the social determinants of health (2004). Russian anthropologists and sociologists think that a Violence poor social adaptation to urbanization and migration The homicide rate in Russia increased rapidly during may have caused young people to rely on unhealthy the 1990s. It is now about 20 times higher than in substances, such as alcohol and smoking, to relieve Western Europe and among the highest recorded any- stress. At least one report indicates that the lack of a where in the world (Chervyakov et al. 2002). In 1999, social network, which existed in the past, coupled with for example, close to 30,000 persons died from homi- the pressures from growing expectations (bolstered by cide in Russia. As the average age of homicide victims is the media), is prompting young people to turn to crime, much lower than for most other causes of death, it has a violence, and suicide (Abdullaev 2004). very negative impact on the number of years lost. Also, alcoholism is closely related to crime, as it prompts risk Policy Implications taking and aggressive behavior, and extreme violence is seen as a way to resolve problems. Furthermore, the The vast majority of Russia's burden of disease and proportion of crimes committed by men and women mortality is caused by (a) NCDs, such as cardiovascular 32 diseases and cancer Figure 3.8 Injury Mortality Rates in Russia as a Percentage of Swedish Rates and (b) injuries due to motor vehicle 1400% accidents, suicide, and other external 1200% causes. Young adults 1000% in Russia are hit hardest. Nearly 7 of 800% 10 Russians today die from CVDs or 600% injuries. NCDs are generally not curable 400% once they develop, and most become 200% chronic conditions. However, many are 0% preventable, and <1 1-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 85+ 75-79 80-84 people with NCDs can continue to lead Source: WHO Mortality Database. normal, productive lives if the condition is diagnosed early and managed correctly. Furthermore, both because of the high costs of treatment and because well-recognized and inexpensive interventions to screen of decreased productivity from ill health. Given Russia's for and control NCDs exist and are widely applied in high regional disparities, substantial improvements in most developed countries. Effective interventions could the population's health could be achieved by targeting start soon to improve the health of Russian adults, with- its less-developed regions. out having to wait for the next generation. This is highly relevant, as Russian policy initiatives often emphasize This all means that Russia faces a major challenge asso- interventions for the young. More importantly, if NCDs ciated with the growing NCD and injury burden: insti- are not managed correctly, they lead to costly hospitaliza- tutional realignments are required because effective tions, the need for long-term care, and the loss of pro- approaches to deal with this challenge demand an inter- ductivity and income. Policy discussions that involve the disciplinary approach and interventions that span differ- efficiency of the health system and health services will of ent sectors. Developed-country experience suggests that necessity increasingly have to include the prevention and prevention policies and interventions are more afford- management of NCDs. Their treatment and long-term able and effective when they are oriented to the popula- management is already a major cost in middle-income tion at risk of illness and injury rather than solely at and developed countries. individuals who are already ill or disabled. As it will be discussed in Chapters 7 and 8, reaching the source of There is a misconception that NCDs and injuries affect the problem requires new capacities and skills in the the rich and not the poor, but their prevalence is actual- areas of regulation; information management; education ly higher in less-developed regions and among low and communication; mobilization and management of income groups, for two reasons. First, the poor are more community-based and nongovernmental organizations; likely to be exposed to the risk factors leading to NCDs and working across sectors, including the private sector. (alcohol abuse, smoking, physical inactivity, high Health services will also need to significantly improve to fat/high carbohydrate diet without fruits and vegetables, better manage chronic health problems as well as acute obesity, stress, and high blood pressure). Second, the ones. Treatment adherence and long-term follow-up will poor are less likely to have access to the medical care become major issues; lastly, patients must be involved in that alleviates the consequences of NCDs. Also, NCDs their own health care management. potentially impoverish those on the low income groups 33 4 Chapter 4. What Are the Major Risk Factors for NCDs and Injuries? R isk factors are defined as any attribute or charac- Alcohol Abuse teristic of an individual or exposure to an element Alcohol can benefit as well as harm individuals. In that increases the likelihood of developing a dis- many countries drinking alcoholic beverages is a pleas- ease or injury. Though risk factors may not cause dis- urable part of social life and is seen as cardio-protective ease, their presence increases the probability that it will (OECD 2003). Alcohol use in Russia is closely tied to develop. This chapter discusses the main preventable social norms, and heavy drinking constitutes a major risk factors for noncommunicable diseases (NCDs) and public health problem. injuries in Russia to lay a foundation for a discussion on reducing those risk factors. Official statistics indicate that during 1965-87, the prevalence of alcoholism rose in Russia to a level of 2,008 per 100,000 population. The morbidity from Major Individual Health Risk Factors alcoholic psychoses (a good measure of morbidity from The mortality attributable to 10 leading risk factors in alcoholism) also rose in the 1960s and 1970s, and began Russia in 2002 is shown in Table 4.1. The first three-- to decrease in the 1980s due in large part to high blood pressure, high cholesterol, and tobacco-are Gorbachev's 1985 anti-alcohol legislation that limited estimated to contribute to more than 75 percent of the alcohol sales and raised the legal purchasing age to 21. country's deaths. The table also shows the DALYs (dis- However, alcohol consumption decreased only 26 per- ability-adjusted life years, defined in the previous chap- cent despite a 63 percent decrease in alcohol sales: the ter) lost attributable to the same risk factors. In this consumption of unregistered alcohol, including moon- case, the top three risk factors--alcohol, high blood shine, nearly doubled, and increases were also seen in pressure, and tobacco-account for more than 46 percent substance abuse and related poisoning, especially among of DALYs lost in 2002. the youth. The law was repealed after three years, Table 4.1 Deaths and DALYs Lost Attributable to 10 Leading Risk Factors, 2002 Rank Risk factor Total Deaths % Rank Risk factor Total DALYs % 1 High blood pressure 35.5 1 Alcohol 16.5 2 High cholesterol 23.0 2 High blood pressure 16.3 3 Tobacco 17.1 3 Tobacco 13.4 4 Low fruit and vegetable intake 12.9 4 High cholesterol 12.3 5 High BMI 12.5 5 High BMI 8.5 6 Alcohol 11.9 6 Low fruit and vegetable intake 7.0 7 Physical inactivity 9.0 7 Physical inactivity 4.6 8 Urban outdoor air pollution 1.2 8 Illicit drugs 2.2 9 Lead 1.2 9 Lead 1.1 10 Illicit drugs 0.9 10 Unsafe sex 1.0 Source: WHO-EURO (2005). The European Health Report (http://www.euro.who.int/eprise/main/who/progs/ehro5/home). 35 Figure 4.1 Per Capita Alcohol Consumption in Russia, US, and UK, 1999 more alcohol of poorer quality than people in cities as demon- Liters per adult strated by a long-standing tenden- cy of higher morbidity due to alco- 12 Russia United States United Kingdom hol poisoning among rural popula- tions. 10 8 However, due to the lack of partic- ipation of alcoholics and heavy 6 drinkers in population surveys, as well as underreporting, data from 4 self-reported alcohol consumption in Russia among men and women 2 is considerably less than reported 0 in official statistics. Using biologi- Alcohol Spirits cal markers of alcohol drinking (e.g., carbohydrate-deficient trans- Source: WHO 2003a. ferring (CDT)) to determine the alcohol-associated risk, a study in Russia's Karelia Republic in 1997 although morbidity was minimal and alcohol consump- found CDT values both among men (about 37 percent) tion dropped while it was in effect. and women (about 18 percent), indicating heavy alco- hol consumption (Laatikainen et al. 2002a). The respec- With the liberalization of alcohol sales and relaxation of tive rates in Finland in the same year were 9.6 and 9.4 administrative controls, an increased morbidity trend percent. Evidence from epidemiological surveys reported from alcohol psychosis occurred. Surveys show a sub- by the Russian Ministry of Health (MOH) in the 1990s stantial increase in alcohol consumption between 1992 suggests that the prevalence of habitual drunkenness is and 1994 consistent with a sharp price reduction. about 20 percent of the population in several regions of Alcohol consumption rose by 25 to 30 percent as the the country. price dropped by 58 percent during this period. Researchers estimate the price elasticity of demand for Alcohol abuse contributes to a large proportion of alcohol to be -.36. A reduction in alcohol consumption deaths among working-age males. Some estimate that and a relative rise in the price of alcohol were observed between a quarter and a third of all adult male deaths between 1994 and 1996. may be directly related to alcohol abuse (McKee 2005; MOH 1994). Adult per capita alcohol consumption in 1999 was 10.7 liters per adult in Russia versus 8.6 liters in the United Alcohol may also affect mortality through violent death, States and 9.7 liters in the United Kingdom (Figure as discussed in chapter 3. The Russian MOH reported 4.1). Although these levels are not dissimilar, the key sample survey results in the 1990s showing that alcohol difference is that 75 percent of the alcohol consumed in was implicated in 50 percent of deaths from accidents, Russia is spirits, whereas in the United Kingdom and poisonings, and injuries. There is also a strong direct the United States, 56 and 60 percent, respectively, is correlation between mortality from cirrhosis of the liver beer. Recent data from the Russian Longitudinal and per capita consumption of alcohol: it is estimated Monitoring Survey (RLMS) indicate that in 2002, alco- that 30-50 percent of all cases of such cirrhosis are due hol consumption for all groups increased: 14.5, 2.4, and to alcohol abuse. In alcoholics, mortality from cancer of 1.1 liters per year among men, women, and teenagers, all sites is 25 percent higher than in the general popula- respectively. In 2004, about 70 percent of men, 47 per- tion. Binge drinking, common in Russia, leads to cent of women, and 30 percent of teenagers were increased cardiovascular disease (CVD) mortality, partic- drinkers (RLMS 2005). In rural areas Russians drink ularly sudden cardiac death (scientific evidence indicates 36 that alcohol abuse can injure the myocardium and cause Use of Tobacco a dangerous arrhythmia (Leon et al.2005). Indeed, one Tobacco is the only consumer product that eventually study found a significant increase in deaths from alcohol kills half of its regular users if they follow its manufac- poisoning, accidents, violence, and CVD on Saturdays, turers' recommendation (Beaglehole and Yach 2003). Sundays, and Mondays that were associated with binge The WHO estimates that 300,000 deaths worldwide are drinking on weekends (McKee et al. 1998). Among attributable to smoking every year (with 100,000 caused other behaviors that compound the alcohol problem in by cancer), more than the number of deaths caused by Russia are drinking alcoholic substances not intended traffic accidents, use of illegal drugs, and suicide com- for consumption, such as perfume and medicine, and bined. Peto et al. (1994) estimate that smoking account- days-long binging (McKee 2005). ed for 30 percent of all male deaths (42 percent in the age group 35-69) and 4 percent of all female deaths (6 Today a very high proportion of deaths at working ages percent in the age group 35-69). Gerasimenko and in Russia are attributable to causes of death that explic- Demine (2001) have shown that smoking shortens life itly involve alcohol. In an ongoing case-control study in expectancy by 6.7 years for men and 5.3 years for Izhevsk (a city in the Urals), an interim analysis shows women. that of 1,400 deaths from all causes occurring among all male residents aged 25-54 years in 2003-04, 18 percent Associated with higher rates of CVDs, many cancers, were certified by a medical expert as having one of four and chronic lung diseases, cigarette smoking is the sin- causes that explicitly specify alcohol: mental disorders gle most preventable cause of disease and death in due to alcohol; alcoholic cardiomyopathy; alcoholic cir- Russia. The prevalence of smoking among Russian rhosis of the liver; and acute alcohol poisoning (Leon et adults is shown in Figure 4.2. Russia has one of the al. 2005). This 18 percent, however, is a minimum in world's highest rates of smoking among men: more this age group, as it excludes the substantial proportion Russian men smoke than European men--the ratio is 2 of deaths attributable to alcohol but arising from causes to 1.19. In 1998, 6 out of 10 male adults in Russia were that do not explicitly imply alcohol, such as injuries and violence as well as a wide range of illnesses, including respiratory and cardiovascular disease. The Figure 4.2 Smoking Prevalence among Adults in Selected G-8 Countries results from this study suggest that this risk factor may explain 70 the mortality crisis in Russia % Men Who Smoke 60 more than any other. % Women Who Smoke 50 In terms of morbidity, the preva- lence of mental disorders in 40 heavy users is twice as high as in the general population, and mor- 30 bidity with temporal disability is 1.5 times higher among alcohol 20 abusers than among moderate drinkers. Other problems associ- 10 ated with alcohol abuse are alco- 0 holic gastritis in 95 percent of Russia Japan Germany France Italy UK USA Canada alcoholism cases, ulcers, and alcoholic cardiomyopathy, which Source: Shafey, Dolwick, and Guidon 2003. results in cardiac failure to vari- Note: Data are from 2000, except for Russia (1990s), Canada and the United Kingdom (2001), and ous degrees. Italy (2002). Data cover the population aged 20 and older except for Germany (18-59), France and USA (18+), Italy and Canada (15+), and the United Kingdom (16+). 37 Table 4.2 Percentage of 15-Year-Olds and than 340,000 drug addicts in the lists maintained by Adults Who Smoke in Selected public institutions (Cherkesov 2005). The number of Countries, 2003 persons registered in medical and prevention institutions with a diagnosis of drug abuse increased by a factor of 15-year-olds Adults 2.1. However, it is estimated that the actual number of Russia 15.8 36 drug abusers in the country exceeds those officially reg- Germany 27.5 35 istered by five to eight times (Cherkesov 2005). Ireland 15.8 31 Injecting drug users are estimated to have 20-times United Kingdom 17.7 27 higher risk of death than the general population, mak- France 19.8 27 ing it conceivable that a portion of Russian teenage Italy mortality is connected to substance abuse (UNICEF 16.1 25 2004). Injecting drug use is also the driving force Source: WHO 2003a. behind the HIV/AIDS epidemic in Russia (although HIV/AIDS is an infectious disease that shares certain smokers--more than twice the comparable rate in the characteristics with many NCDs, such as a long latency United States and the United Kingdom. Recent RLMS period, and affects predominantly young adults, it is not data indicate that while smoking prevalence among men considered in this assessment. For additional informa- decreased from about 65 percent in 2002 to 61 percent tion see World Bank 2003b). in 2004, the prevalence of smoking among women increased from 7.3 percent in 1992 to 15 percent in Nutrition- and Activity-Related Factors 2004--a more than 100 percent increase over a 12-year Dietary intake, cholesterol, obesity, and hypertension period (RLMS 2005). Lower smoking prevalence should not be seen as separate, individual risk factors. In among older men reflects trends over time and (much) fact, it is question of certain major problems in the diet: higher death rates among long-term smokers as they intake of high saturated fat (animal fat, hydrogenated grow older (McKee et al. 1998). vegetable fats), high salt intake, low vegetable and fruit intake, and low intake of good (vegetable and fish) oils. The percentage of 15-year-olds who smoke in Russia is These aspects of diet, coupled with a sedentary lifestyle reported to be less than the percentage in some are the determinants of high blood cholesterol, high European countries even though the percentage of body weight, and high blood pressure among Russians. adults who smoke is higher (Table 4.2). This rate has Thus, these factors should be tackled as a whole--as also increased in Russia since the 1990s, particularly in nutrition- and activity-related factors. girls (Gilmore and McKee 2004a). These smokers, espe- cially men, generally reported initiating their smoking Food intake. Dietary intake in Russia is characterized habit before 18 years of age and sometimes even in by a diet high in animal fat and salt, and low in fruits childhood (Gilmore et al. 2004). Since smoking is an and vegetables. RLMS data for 2004 indicate that fat addictive behavior, the prevalence of young smokers is intake in Russia is much higher than the recommended indicative of future adult patterns. level of 30 percent of total energy intake for all age groups (the percentage rates are 32.2, 33.9, and 31.3 While the average daily number of cigarettes smoked among children 0-17, adults 18-59, and elderly 60 and among men has hovered near 16 during the 1992-2004 over, respectively). Consumption of most foods has fall- period, among women and teenagers a pronounced en, with the exception of bread and potatoes (Hawkes increase is observed: from 8 to 11, and from 8 to 9.4, 2004). Among the poor, consumption of milk, diary respectively. products, fruit, vegetables, fish, and meat is less than the average intake, while their consumption of bakery prod- Drug Use ucts and potatoes is high. Protein intake, which was Over the last 10 years illegal drug use has increased rap- showing a slow decrease until 2000, has been increasing idly in Russia. At the beginning of 2005, drug users slowly since then. The decline in fruit and vegetable were estimated to number 500,000, including more consumption in Russia is estimated to explain 28 per- 38 cent of the increase in CVD mortality (Brainerd and sity has a substantial effect on longevity, reducing the Cutler 2004). Food accounts for over 70 percent of length of life of people who are severely obese by an household expenditures of the poor and 35 percent of estimated 5 to 20 years, while Olshansky et al. (2005) the richest 10 percent. found that the substantial rise in the prevalence of obe- sity and its life-shortening complications (i.e., diabetes) The importance of food in lowering the burden of dis- in the United States pose a threat to the steady rise in ease is often not fully appreciated. Along with smoking life expectancy observed in the modern era. One study and sedentary lifestyles, a diet high in fats and sugar and has attributed severe obesity to a 12-fold increase in low in vegetables and fruits is a major cause of CVDs, mortality among 25- to 35-year-olds as compared to obesity, and cancers. The WHO estimates that about a lean individuals (PAHO 2003). Being obese also ele- third of all CVDs are due to poor diets and that better vates an individual's risk for Type 2 diabetes, heart dis- diets could lower cancer cases by about 30-40 percent. ease, and certain cancers (breast, prostate, and colon). It Evidence from North Karelia, Finland, shows that in the similarly is associated with an elevated risk for nonfatal years 1972-92, mortality from blood pressure and but debilitating conditions affecting the skin, respiratory CVDs dramatically fell following a successful program systems, musculoskeletal systems, and fertility (WHO of good diet (see box 7.1, p. 61). 2004a). In addition, obesity is stigmatized in society, so overweight individuals may suffer psychological stress High cholesterol. Different epidemiological studies relating to low self-esteem (PAHO 2003). (including MOH 1997) in several regions in Russia indicate that about 60 percent of adults have blood cho- Russian men are much less likely to be obese than men lesterol levels above those recommended (200 mg/dL in the United States. In 2000, 12 percent of Russian and 5.2 mmol/L), and about 20 percent have levels that men aged 25-64 were obese, compared with 25 percent put them at high risk and in need of medical care (250 of U.S. men in 2001. Russian and U.S. women are clos- mg/dL, 6.2 mmol/L). Among men in the 20-54 age er at 28.5 percent and 24.7 percent, respectively. The group in selected cities, about every sixth one (16 per- latest (2002) Russian Behavioral Risk Factor cent) suffers from marked hypercholestrolemia. A study Surveillance System results indicate that the prevalence in St. Petersburg showed a significant decline in HDL of overweight among adults aged 25-64 in various (so-called good) cholesterol to dangerously low levels regions of Russia ranges from 47 to 54 percent for men across all men in the 20 to 69 age group, as well as for and from 42 to 60 percent for women (CINDI 2004). women (Plavinski et al. 1999).While Russia's blood cho- However, self-reported data are often underestimated, lesterol levels are not particularly high compared to especially the female obesity rates. According to data Western averages, strong evidence points to the crucial from Pitkäranta, Karelia Republic, in the 1990s and basic role of blood cholesterol. International experience, 2000, about 35 percent of women had a BMI (meas- even from neighboring countries--Finland and ured) equaling or more than 30 (Laatikainen 2000). Poland--shows how changes in a population's CHD rates seem to follow closely its general blood cholesterol This means that although obesity can be ruled out as a level. primary cause of the mortality crisis in Russia, it can still have a large negative effect on individual and popu- Obesity. Adults who are overweight or obese are at risk lation health as it is related to high blood pressure, high for premature death and disability. Those with a body cholesterol, and diabetes. However, the prevalence of mass index (BMI) of 25 to 29.9 are considered over- obesity has been rising in industrialized countries and weight, while those with a BMI of 30 or more are con- can be expected to do likewise in Russia as its people sidered obese.3 Fontaine et al. (2003) reported that obe- adopt Western lifestyles. In fact, from 1993 to 2000, obesity in Russia increased slightly for both men and women. The trend in obesity prevalence has risen by 3 BMI, a common measure expressing the ratio of weight to about 10 to 40 percent in the majority of European height, is determined by a mathematical formula in which a per- countries over the past 10 years (OECD 2003). son's body weight in kilograms is divided by the square of his or her height in meters (i.e., wt/[ht2]). 39 While genetic predisposition for obesity is a key deter- percent for females. These figures, however, may under- minant, especially after adoption of Western lifestyles, estimate the problem as they are based on self-reported two major lifestyle influences merit mention: urbaniza- data. According to data (with measurements) from tion and women entering the workforce. Urbanization Pitkäranta, Karelia Republic, 57 percent of men and 55 changes diets and leads to a more sedentary lifestyle for percent of women had blood pressure above 140/90 many. Working women are likely to depend on conven- mm Hg and/or treatment for hypertension. With the ience foods--for themselves and their families--that are higher cut-off point of 160/95 mm Hg, the respective typically low in nutrients and high in fat. rates were 30 percent and 38 percent. More than 40 percent of males and 25 percent of females were not Economic and financial impact studies on treating the aware of their hypertension. This lack of awareness morbidity and mortality associated with high choles- clearly affects the self-reported figures on hypertension terol, increasing obesity, and related NCDs report that prevalence (Laatikainen 2000). the cost of medical therapy to screen and reduce choles- terol levels is at least US$10,000 per life year saved Treatment of high blood pressure includes behavior (McKinlay 1993). The direct medical costs of obesity changes and drug therapy. The best approach for insti- are estimated at US$40 billion in the United States gating these activities is a combination of population- alone (Posten and Foreyt 1999), and it accounts for based and intensive, targeted strategies for primary pre- almost 4 percent of DALYs lost in the European Union. vention. Physical activity, a low-salt diet, moderate alco- Because treatment is so expensive and often unafford- hol consumption, potassium supplementation, modifi- able, prevention through dietary modification and cation of eating habits, and weight loss are proven as increased physical activity is the cost-effective way to effective interventions (Whelton et al. 2002). combat overweight's contribution to the rapid rise of Interventions to reduce hypertension may be combined NCDs. with those for lowering blood cholesterol level, because these conditions are interrelated and affected by similar Hypertension. Hypertension or high blood pressure factors, such as diet, stress, and obesity. (systolic blood pressure above 140 mm Hg or diastolic blood pressure above 90 mm Hg) is a major risk factor Sedentary lifestyles. Russians are increasingly adopting for cardiovascular diseases. About 50 percent of people a sedentary lifestyle more typical of industrialized who have a first heart attack and about 66 percent of economies (Table 4.3 illustrates the relationship between people experiencing a first stroke have blood pressure modern lifestyles and low physical activity). From 2000 above 160/95 mm Hg (Hellermann et al. 1997). to 2002, 73-81 percent of surveyed men and 73-86 per- Globally, 26 percent of the adult population has hyper- cent of women aged 25-64 reported having low-levels of tension (Kearney et al. 2005), and of these only 30 per- physical activity (CINDI 2004). They reported that they cent may be aware of their condition (NHLBI). engage in less than 20-30 minutes of rigorous exercise Hypertension has been identified as the leading risk fac- fewer than five days a week. This is an alarming statistic, tor for death and is ranked third as a cause of DALYs since lack of physical exercise increases risks for most lost (Ezzati et al. 2002). Hypertension is preventable, NCDs. Moderate physical activity, for example, per- though, and early detection and treatment are critical: formed on most days of the week can substantially large trials indicate that a 5 mm Hg reduction in dias- reduce the risk of dying from CVD and can reduce the tolic pressure corresponds to a 21 percent reduction in risk of developing colon cancer, diabetes, and high blood heart disease risk (Magnus and Beaglehole 2001). pressure. In addition, regular exercise confers the imme- diate benefits of mood improvement and a sense of well- High blood pressure is clearly the greatest contributor to being and helps prevent weight gain (U.S. CDC 2004). the onset of CVDs, Russia's leading killer. It is closely Both physical and mental health can also help reduce related to high salt intake and other aspects of diet (and accidental injuries and suicides due to depression. to alcohol intake and obesity). The 2002 Russian Behavioral Risk Factor Surveillance System study found Physical inactivity alone is estimated to account for 3.5 that prevalence of hypertension in the male population percent of the DALYs lost in Europe (WHO 2005a). ranged from 34-46 percent for males and from 32-46 This is significant, because it is a risk factor that can be 40 Table 4.3 The Link between Modern Lifestyles and Low Physical Activity Location or activity Modern lifestyle Impact related to obesity Transportation Rise in car ownership. Decrease in walking or cycling. Increase in driving shorter distances. At home Increase in the use of modern appliances Decrease in manual labor. (microwaves, dishwashers, washing Increase in consumption of convenience machines, vacuum cleaners). foods that contribute to obesity. Increase in prepared and processed foods Decrease in time spent on more active and ingredients. recreational pursuits. Increase in television viewing, and computer and video game use. In the workplace Increase in sedentary occupational Decrease in physically demanding behaviors due to technological advances. manual labor. Public places Increase in the use of elevators, escalators, Decrease in daily physical activity and automatic doors. patterns, such as climbing stairs. Urban residency Increase in crime in urban areas. Prevents women, children, and elderly people from going out alone for exercise and leisure activities. Source: AOA 2005. cost-effectively averted by individual effort, since mod- doesn't stop there: people with diabetes are at higher risk erate and regular physical activity can be integrated into for heart disease, blindness, kidney failure, lower-limb daily routines. However, individual effort needs the sup- amputations, and other chronic conditions such as port of public health policies that emphasize and pro- nerve and dental diseases (ADA 2003). They are two- to mote physical activity. For example, in Sweden and four-times more likely to suffer heart attack and stroke, Netherlands, where bicycle riding and walking are pop- and between 50 and 80 percent of them die of CVDs. ular, obesity is less common than in car-bound societies such as the United States, Canada, and the United The WHO considers Russia to be among the top 10 Kingdom (PAHO 2003). countries ranked by number of diabetics (WHO 2004a). At more than 3 percent, the prevalence of dia- Diabetes. Diabetes is a group of diseases that share betes in Russia is only slightly higher than the world high blood sugar levels due to an absolute or relative average of 2.5 percent. It is estimated that nearly half of insulin deficiency. Insulin-dependent diabetes mellitus, U.S. diabetics are unaware of their life-threatening con- or Type 1 diabetes, may appear at any age, is seen most dition--and this may be the situation in Russia. frequently in children and adults under age 30, and requires insulin; its most common cause is the destruc- Table 4.4 (on page 42) compares the prevalence of dia- tion of pancreatic cells. The most common form of dia- betes in Russia with its prevalence in other G-8 coun- betes is noninsulin-dependent diabetes mellitus, or Type tries. As a percentage of total population, its estimated 2, which usually appears after age 40; it is generally prevalence is comparable to that in other European associated with obesity, genetic factors, and modern countries except Italy, and it is substantially below that lifestyles characterized by a sedentary life, dietary factors of Japan and North America. (such as excessive simple carbohydrates), and stress. Reported mortality from diabetes in Russia (10.2 deaths Diabetes was the sixth leading cause of death worldwide per 100,000 population) is only about half that in other for older adults in 2002 and affected at least 171 mil- countries (19-21 deaths per 100,000 population), but lion people in 2000 with an annual attributable mortali- this difference can be explained by the 16.5 percent reg- ty close to 3.2 million (WHO 2004a). But the burden istration rate. Diabetes mellitus is frequently attended 41 Table 4.4 Prevalence of Diabetes Cases in G-8 percent of the average wage by 2000. The change in the Countries, 2000 minimum wage as a share of the average wage is nega- tively and significantly related to the change in mortali- Country Number Percentage of cases of population ty for all causes and strongly negatively related to the increase in CVD deaths. Also, throughout much of the Ireland 86,000 2.2 1990s, the minimum wage in Russia was far below the France 1,710,000 2.9 "subsistence minimum" calculated by the government in United Kingdom 1,765,000 3.0 determining poverty rates. And today the average Russia 4,576,000 3.2 Russian confronts an unpredictable future, a psychoso- Germany 2,627,000 3.2 cial factor influencing health. Japan 6,765,000 5.3 Socioeconomic Disadvantage as a Risk Factor United States 17,702,000 6.1 Canada 2,006,000 6.4 Plavinski, Plavinskaya, and Klimov (2003) examined the Italy 4,252,000 7.4 association between social factors and the increase in mortality in Russia in the 1990s and found that men in Source: Derived from Wild et al. 2004. lower socioeconomic groups were most affected by the sharp increases in mortality. Several health risk factors are associated with the onset of major causes of death by other pathological disorders, including diseases of the and disease in Russia and are prevalent among the poor, circulatory system (66 percent), diseases of digestive increasing the burden of illness and disease in low- organs (32 percent), kidney diseases (20 percent), and income groups. Since risk factors are synergistic in their respiratory diseases (15 percent). A more common diag- effect, the relative risk of developing NCDs and suffer- nosis, disease of the cardiovascular system, is the likely ing injuries increases sharply when various risk factors principal cause of death even among diabetic patients in are combined. Russia. In any case, with the increasing trends in popu- lation aging, unhealthy diets, obesity, and sedentary Ivaschenko (2003) examined the impact of poverty and lifestyles, more people are expected to develop diabetes public expenditures in Russia and found both variables at a younger age in transition economies, such as in to be significant in explaining observed variations in Russia. longevity across regions and over time. This finding must be interpreted with caution, though, because the Other risk factors investigators did not control for the quality of expendi- tures: additional expenditures may have been made on Psychosocial Stress activities ineffective in improving health outcomes. The Mental well-being goes hand in hand with good physi- need is to do better by focusing on effective policies and cal health. Stress is the most commonly reported cause programs, and, in that context, increasing public expen- of sickness absence and a major cause of incapacity in ditures on health as necessary. developed economies. A high level of stress is also posi- tively related to the development of CVD. Psychosocial The concept of inequalities in health outcomes, which stress in Russia is likely brought on by the substantial is commonly acknowledged in Western European coun- changes in the economy and an erosion of the social tries, is less commonly discussed in Russia. Russia's very safety net in the last decade. large regional variations in health status, noted earlier, are related in part to socioeconomic factors. Russian Estimates of the impact of psychosocial stress on mor- adults living in the most affluent region in 2001 could tality in Russia vary, but Brainerd and Cutler believe expect to live on average 20 years longer than those in stress may explain a quarter of the increase in mortality the poorest region. And again, the gap may be widen- (2004, p. 37). Economic reforms in the 1990s changed ing: men in the most affluent region gained approxi- the lives of virtually every Russian citizen. Minimum mately three years in life expectancy between 1990 and wages in Russia have fallen dramatically: to less than 6 2001, while men in the poorest region made no gains. 42 Figure 4.3 Russians' self-reported health Figure 4.4 Russians' self-reported use of status, by consumption quintile, 2003 hard alcohol, by consumption quintile, 2003 Poorest Richest Poorest Richest 60% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Good or Satisfactory Bad or Several Almost Very Good Very Bad Times/wk Everyday Source: NOBUS Survey, 2003. Source: NOBUS Survey, 2003. Furthermore, the NOBUS survey found that individuals consumption of hard alcohol than those in richer quin- in the poorest quintiles in Russia were more likely to tiles (Figure 4.4). self-report bad or very bad health status than men in richer quintiles (Figure 4.3). NOBUS data additionally show that poor people have greater problems accessing health care than richer peo- The NOBUS survey also found that individuals in ple. The percentage of respondents in the poorest quin- poorer quintiles were more likely to report frequent tile who reported they could not afford necessary Figure 4.5 Self-Reported Reasons for Not Seeking Health Care, by Consumption Quintile, 2003 100 Poorest Richest Rural Reason 80 Urban ting Repor 60 40 Respondents of eg 20 centareP 0 No doctors of Difficult to Cannot afford Other reasons medical specialty in area arrange a visit necessary services Source: NOBUS 2003. 43 Figure 4.6 Self-Reported Reasons for Not Following Prescribed Survey of Household Welfare and Treatment, by Consumption Quintile, 2003 Program Participation (NOBUS)4 found that a high portion of drivers do not wear 30 Poorest seat belts. In some regions, the prevalence Richest of Russians refusing to wear seat belts is Reason 25 close to 50 percent for men and 33 per- ting cent for women. This cultural trait of a 20 "long-held contempt for safety rules" was Repor observed by a researcher at Russia's Institute of Ethnology and Anthropology 15 (Abdullaev 2004) and may contribute to the high injury death rate. Pedestrians, Respondents 10 of cyclists, and motorized two-wheel-vehicle eg users also bear a disproportionate share of 5 centareP the injury burden and are at a high risk of crash injury. Finally, inadequate post- 0 crash care, particularly pre-hospital care Only free Partly followed Not followed Not followed prescriptions for lack of money at all due to for other reasons offered by a well-established ambulance for drugs/procedures lack of money and communication network and medical teams, is a major risk factor in some Source: NOBUS 2003. regions that contributes to injury, disabili- ty and death. services was more than three times the percentage of respondents with that report in the richest quintile Worksite Safety (Figure 4.5 on page 43). Those in the poorer quintiles The occupational environment often presents a consid- were also three times as likely to report the lack of a erable risk to the health of employees. Exposure to an local specialty doctor as a reason for not seeking care. unfavorable occupational environment results in fatal Rural respondents reported particular problems with the injuries, nonfatal injuries, and work absences. Data availability of medical specialists. show an increased risk of developing cancer in some industries (e.g., ferrous and nonferrous metallurgy and Lastly, almost 75 percent of households in the NOBUS asbestos-textile production; MOH 1994). Relevant survey reported that they had completely followed pre- 2001 data are in Table 4.5. Creating and sustaining a scribed treatment following a visit to the physician. The preventive safety and health culture at the workplace remaining 25 percent gave several reasons for not doing would help to reduce the number of work-related so (Figure 4.6). These data indicate what could be antic- deaths, ill health, and disability each year. ipated: poorer people are more likely to face issues relat- ed to the affordability of health care. Interaction among Risk Factors Road safety-related risk factors The synergistic effect of several health risk factors acting As noted in Chapter 3, standardized death rates from on any individual has been mentioned previously but traffic accidents are much higher in Russia than Europe. bears repeating. The relative risk of developing NCDs Alcohol impairment is an important factor in crashes and contributes significantly to crash injury (as crash risk starts to rise sharply at levels of 0.04 gram per deciliter 4 [g/dl]), legal blood alcohol concentration [BAC] limits Performed in 2003, NOBUS included a sample of about 44,500 households and is representative both nationally and for 46 of set at 0.10 g/dl allow triple the risk of 0.05 g/dl). Failure the larger subjects of the Russian Federation. It captures a range to use seat belts and child restraints more than doubles of aspects of household welfare and has a strong focus on house- the risk of serious and fatal injury, but the National hold access to social services. 44 and suffering injuries increases sharply when various risk Table 4.5 Occupational Accidents and Work-Related factors combine. As many as two-thirds of all cases of Disease, Russian Federation, 2001 NCDs and injuries are attributable to smoking, Economically active population unhealthy dietary habits, physical inactivity, psychoso- cial risk factors, alcohol abuse, and drug and toxic sub- Number 69,731,000 stance abuse. In Moscow a long-term prospective epi- Percentage 48 demiological study (1977-87) found that when two or Total employment 64,710,000 three major risk factors (smoking, hypertension, and Work-related fatal accidents high cholesterol levels) combine, mortality from all Accident causing 3 days' absence 4,789,749 causes and from CVD doubles compared to only one Work-related mortality 130,533 (MOH 1994). Risk factors more prevalent among the Work-related diseases 117,981 poor contribute to increases in the burden of illness and Deaths caused by dangerous substances disease in low-income groups. 26,015 International Labor Organization estimate of fatal Young Russians add to these health risks by frequently accidents (correction rate 1.126) 6,276 engaging in other risky behaviors, such as unsafe sex, Industry 3,021 injecting drugs, needle sharing, and a general neglect of Service 2,440 safety rules (UNICEF 2004; Abdullaev 2004). All of Agriculture 815 these behaviors point to a worrisome picture of a possi- ble epidemic in HIV infection, other sexually transmit- Source: World Bank 2005b. ted infections, and tuberculosis that will compound the already-negative effect of NCDs and injuries on young Russian adults. are based on self-reports and thus underestimated, including, for example, the prevalence of hypertension, Since risk factors are so prevalent, synergistic, and often hypercholesterolemia, and obesity. combined in individuals, addressing several rather than focusing on one at a time would have a multiplicative Second, the Russian people have been the subject of result in reducing morbidity and mortality. To effective- very few epidemiological studies on the etiology of the ly address these risk factors, as will be argued in the fol- conditions that are driving the exceptionally high mor- lowing chapters, will require societal commitment to a tality among men (and women) of working age. broad health promotion and disease prevention policy, Consequently, current understanding of disease etiology not only to encourage people to make positive lifestyle in Western countries cannot easily transfer to Russia, changes, but also to create a physical and social environ- particularly with respect to cardiovascular disease. For ment supportive of such change. example, recent studies note that the evidence that established risk factors explain most of the worldwide incidence of acute myocardial infarction is less convinc- Policy Implications ing for Eastern Europe and the former Soviet Union. Different researchers have stressed that the understand- Specifically, the pattern of sporadic binge drinking of ing of the current poor state of health of the Russian spirits that appears to distinguish Russian alcohol con- population is contentious and provisional. Why is this? sumption from most other countries may well be associ- ated with many sudden deaths--some of which may be First, compared to many other industrialized countries, erroneously classified as myocardial infarctions. in particular those in the G-8, Russia has very little sys- tematic surveillance and monitoring of these factors at a The paucity of adequate surveillance and monitoring population level. That is, sources of information on data and of etiological evidence is itself a reflection of prevalence of smoking, hypertension, and obesity from the serious lack of research and monitoring capacity. representative samples of the Russian population are few The level of research activity on the major causes of and far between. In addition, most of the existing data morbidity and mortality in Russia is lower that other 45 industrialized and G-8 countries. Epidemiology and environmental risk factors. In addition, knowledge of public health in Russia retain much of the same focus as methods for evaluating the efficacy of interventions, in the Soviet period. Epidemiology is principally about both medical and community based, needs to be the surveillance and control of infectious diseases. Risk strengthened. Attention should be paid to building factors for disease are generally conceived of relative to research capacity in public health and epidemiology to the external environment--such as air pollution or even quickly increase the level of research activity conducted excessive noise levels. Knowledge and expertise in the in Russia use of epidemiological methods for studying non-com- municable diseases common in industrialized countries While there clearly are many open questions on the cur- (including cohort and case-control studies) is not wide- rent health crisis in Russia, and there is much to be spread in Russia. Moreover, modern methods for assess- studied, the policy implications are clear: without any ment of efficacy of interventions or treatments, such as reasonable doubt tobacco, alcohol abuse, and some the randomized controlled trial, are not well known. aspects of national diet and physical inactivity are major causes of premature mortality, ill health, and disability To be able to address effectively the threat from non- in Russia. These are also factors where effective and communicable diseases and injuries, Russia must broad- measurable interventions are available that generate ben- en public health and epidemiological training and sur- efits in years rather than decades as demonstrated in sev- veillance from its current relatively narrow emphasis on eral developed countries, such as Finland and the the monitoring and control of infectious diseases and United States. 46 5 Chapter 5. The Impact of NCDs and Injuries on Health Care Costs N on-communicable disease and injuries impose by diagnosis and by specialty unit in hospitals and out- two types of costs on a society: direct and indi- patient clinics. rect. Direct costs are those that result from the consumption of hospital and other high-cost health care Cost per patient day was used in calculating the cost of services. Russia's total health expenditures per capita inpatient care. The average cost per patient day was cal- were US$535, or 6.2 percent of GDP in 2002--an culated using actual cost data and was adjusted for case amount that represents less than 20 percent of the per mix. Budgeted and actual costs were available for indi- capita health expenditures in France and Germany. vidual health care units based on the following classifi- Indirect costs are costs that result from ill health on cation: salary and payroll taxes, meals, drugs, supplies, individuals' productivity and income-generating poten- rent, and depreciation of equipment and buildings. This tial. For example, the annual cost of cardiovascular dis- approach captured expenditures from the rayon and ease (CVD) in the United States in 2003 was estimated oblast levels, and facilitated the development of esti- at more than US$351 billion (AHA 2003). Of this, 60 mates of national level expenditures. The detailed cost percent (US$209.3 billion) was direct costs (hospital data were linked with aggregate national level numbers and nursing home care, physicians and other profession- on diagnoses, admissions, outpatient visits, and overall als, drugs, medical durable equipment, and home health volumes of care. Interregional variations for input costs care) and 40 percent (US$143 billion) was indirect (lost were used to adjust overall national estimates. The productivity from illness, disability, and death). analysis then adjusted these figures across regions and used national estimates to derive national expenditure This chapter analyzes Russia's direct costs of NCDs and patterns. Outpatient costs were estimated using national injuries and the impact of those costs on the budgets of level ratios of costs incurred for outpatient episodes. the government, insurance organizations, and patients. The methodology, results, and policy implications of the Government and insurance expenditures constitute only analysis are detailed below. a portion of total health expenditures in Russia, which has substantial formal and "informal" payments made by patients. Formal expenditures are mostly pharmaceu- Study Method ticals provided on an outpatient basis, while informal ones are demanded by health service providers. First, costs for NCDs and injuries in Russia were esti- Together, these payments are significant, perhaps as mated for this study (Frid 2005) based on insurance and much as 54 percent of total health expenditures budgetary statistics for two regions--Chuvash Republic (Shishkina et al. 2004). The analysis allocated estimates (an agricultural region) and Kemerovo Oblast (an indus- of informal payments to actual cost categories. These trial region). Each region's insurance company records estimates were derived from the 2003 NOBUS survey covering the previous three years were examined. The and detailed studies by the Independent Social Science entire population is covered by health insurance, so the Research Institute in Moscow, which used household sample is representative for these regions; the data may survey data collected over the previous five years. These also be representative of other regions with similar char- studies provide information on each level of care and acteristics but perhaps are not fully representative of distinguish between formal and informal payments. large urban areas. The databases for Chuvash Republic and Kemerovo Oblast had 904,944 and 347,879 The analysis excluded costs related to emergency care, records, respectively. Treatment cases were summarized expenditures for drugs purchased by patients to take 49 Figure 5.1 Health Care Expenditures to Treat the Four Most Costly · Expenditures for respiratory system dis- Disease Categories, 2003 eases (i.e., acute upper respiratory US$ billions infections, influenza and pneumonia, and chronic lower respiratory diseases) ranked second, accounting for US$1.7 3.0 billion (42 billion rubles) or 13.1 per- Circulatory System cent of the health expenditures. 2.5 Respiratory System External Causes · Expenditures for conditions due to 2.0 Digestive System external causes (i.e., injuries, accidents, poisonings, and burns) ranked third, 1.5 accounting for US$1.2 billion (36 bil- lion rubles) or 9.2 percent of the total. 1.0 · Expenditures for digestive system dis- 0.5 eases (i.e., diseases of the esophagus, stomach, and liver) ranked fourth, 0 accounting for US$1.1 billion (35 bil- Source: Frid 2005. lion rubles) or 8.5 percent of the total. Table 5.1 disaggregates by diagnostic group the total home, and expenditures for targeted programs, such as government and nongovernmental health care expendi- HIV/AIDS. These exclusions render the estimates low. tures for 2003 as a percentage of GDP, showing the rank in terms of expenditures for each diagnostic group. The four groups listed above rank highest in terms of Results Russian health expenditures, followed closely by mental The 2003 cost estimates from two regions in Russia and behavior disorders, cancer, musculoskeletal system were analyzed and extrapolated to the national level for diseases, and genitourinary system diseases. The fact this study using US$13 billion as the total health care that infectious and parasitic diseases ranked 10th is evi- expenditure figure, the widely accepted estimate of dence that Russia is still facing an unfinished agenda of Russia's total health care expenditure, as a denominator, infectious diseases along with a rising burden of NCDs to determine the effect of NCDs and injuries on total and injuries. health expenditures (Frid 2005). The results of those extrapolations indicated that the four broad groups of Estimates of expenditures for circulatory system dis- health conditions--circulatory system diseases, respira- eases, Russia's highest disease group by health care tory system diseases, digestive system diseases, and con- expenditure, are disaggregated in Table 5.1. Cerebro- ditions due to external causes (as classified by the vascular disease is the most expensive condition in this International Statistical Classification of Diseases and group, accounting for US$565.6 million (21 percent) of Related Health Problems, 10th revision [WHO: the total. Also particularly high are expenditures attrib- ICD])--account for an estimated 51.5 percent of total uted to hypertensive (high blood pressure) diseases. health expenditures in Russia in 2003. Figure 5.1 High-cost medical conditions not shown in Table 5.2 depicts these estimates by diagnostic group, with the on page 52 include: highest at over US$2.5 billion and the lowest over a bil- · Asthma: US$286 million (8.8 billion rubles); lion dollars, as follows: · Fractures, injury, poisoning, and certain other conse- · Expenditures for circulatory system diseases (i.e., quences of external causes: US$268 million (8.2 bil- hypertension, ischemic heart disease, and cerebrovas- lion rubles); cular disease) were US$2.7 billion (83 billion rubles) or 20.8 percent of total health expenditures. 50 · Diabetes, malnutrition, and obesity: US$151 million Policy Implications (4.6 billion rubles); This chapter shows the relative importance of NCDs · Gastric and duodenal ulcers: US$106 million (3.3 and injuries on total health care costs in Russia. Many billion rubles); and of the NCDs and injuries that cause death and disability in the Russian population are preventable with changes · Anemia: US$28 million (861 million rubles). in lifestyle to eliminate the risk factors--namely, tobac- co use, alcohol abuse, high blood pressure, high choles- Table 5.1 Estimated Government and Nongovernment Health Care Expenditures (and Percentage of GDP) to Treat Diseases, by Diagnostic Group, 2003 US$ millions and Rub millions Percentage of EXPENDITURES Russia's GDP U.S. (estimated at Diagnostic group Rubles dollars 13.3 trillion rubles) Rank Diseases of the circulatory system 83,094.6 2,708.4 0.62% 1 Diseases of the respiratory system 42,019.8 1,369.6 0.32% 2 Injury, poisoning and certain other 36,478.4 1,189.0 0.27% 3 consequences of external causes Diseases of the digestive system 34,713.3 1,131.5 0.26% 4 Mental and behavioral disorders 28,624.1 933.0 0.22% 5 Neoplasms (cancers) 26,521.8 864.5 0.20% 6 Diseases of the genitourinary system 25,636.4 835.6 0.19% 7 Diseases of the musculoskeletal system 25,429.7 828.9 0.19% 8 and connective tissue Pregnancy, childbirth, and the puerperium 21,600.3 704.1 0.16% 9 Infectious and parasitic diseases 20,983.7 684.0 0.16% 10 Diseases of the nervous system 14,000.5 456.3 0.11% 11 Diseases of the eye and adnexia 9,162.3 298.6 0.07% 12 Diseases of the skin and subcutaneous tissue 8,811.3 287.2 0.07% 13 Endocrine, nutritional, and metabolic diseases 7,623.9 248.5 0.06% 14 Certain conditions originating in the 4,488.5 146.3 0.03% 15 perinatal period Congenital malformations, deformations, 3,577.1 116.6 0.03% 16 and chromosomal abnormalities Diseases of the ear and mastoid 3,233.0 105.4 0.02% 17 Diseases of the blood and blood-forming organs 1,684.8 54.9 0.01% 18 and certain disorders involving the immune mechanism Symptoms, signs, and abnormal clinical and 1,113.3 36.3 0.01% 19 laboratory findings, not elsewhere classified Total 398,797 12,999 3.0% Source: Frid 2005. 51 terol, overweight, low fruit and vegetable intake, and Actuarial analysis would be needed to differentiate physical inactivity--described in Chapter 4. between the most expensive medical conditions at the national level and the most expensive condition to be To compare Russia's highest-cost conditions against treated at the individual level. In the Druss study, for those of the United States, one can look at a study by example, respiratory malignancies are the most costly to Druss et al. (2002). Based on data from 1996, that be treated at the individual level, but they ranked 15th study reports that the 15 highest-cost conditions in the at the national level. Such differentiation between the United States (which include eight circulatory system national and individual levels would allow decision- and four respiratory system diseases) accounted for 44.2 makers to target morbidity more cost-effectively at the percent of total U.S. health care spending. The two preventive and treatment levels, particularly to deal with most costly conditions were ischemic heart disease and future medical inflation that may result from aging and motor vehicle accidents; the most costly condition in from rising expecations of the population, improved Russia--cerebrovascular disease--ranked ninth in the technology, increased care, and expanded scope of care. United States in terms of expenditures. Table 5.2 Estimated Government and Nongovernment Health Care Expenditures to Treat Diseases of the Circulatory System, 2003 US$ in millions and Rub millions EXPENDITURES Diagnostic U.S. Diagnostic group code Rubles dollars Percentage Diseases of the circulatory system I00-99 83,094.6 2,708.4 100.0% Acute rheumatic fever I00-02 1,419.3 46.3 1.7% Chronic rheumatic heart diseases I05-09 6,829.9 222.6 8.2% Hypertensive diseases I10-13 8,635.3 281.5 10.4% Stenocardia (angina pectoris) I20 4,584.1 149.4 5.5% Acute cardiac infarction I21-23 2,380.7 77.6 2.9% Other forms of acute ischemic heart disease I24 128.3 4.2 0.2% Chronic ischemic heart disease I25 3,065.1 99.9 3.7% Cerebrovascular diseases I60-69 17,352.7 565.6 20.9% Subarachnoid hemorrhage I60 336.6 11.0 0.4% Intracerebral and other nontraumatic intracranial I61-62 6.429.2 209.6 7.7% hemorrhages Cerebral arterial occlusion, cerebral infarction I63 2.372.1 77.3 2.9% Stroke unspecified as hemorrhage or infarction I64 370.8 12.1 0,4% Occlusion and stenosis of precerebral and I65-67.1 2.575.9 84.0 3.1% cerebral arteries; other cerebrovascular diseases I67.3-67.9 Cerebral atherosclerosis I67.2 2.983.1 97.2 3.6% Including: Other cerebrovascular diseases 2.462.1 80.3 3.0% Including: Other circulatory system diseases 38,242.4 1.246.5 46.0% Source: Frid 2005. 52 6 Chapter 6. How Do NCDs and Injuries Affect the Russian Economy? R esearch increasingly indicates that a healthy Labor productivity. Healthier individuals could rea- population is not an automatic by-product of sonably be expected to produce more per hour worked. economic development, but can drive economic On the one hand, productivity could be increased growth. Similarly, at the individual level, good health is directly by enhanced physical and mental activity. On an important determinant of economic productivity.5 the other hand, more physically and mentally active Little is known about the correlation of health and individuals could make a better and more efficient use development for the transition countries in Central and of technology, machinery, and equipment. A healthier Eastern Europe and the Commonwealth of Independent labor force could also be expected to be more flexible States that are facing a very particular health challenge, and adaptable to changes (e.g., changes in job tasks and predominantly noncommunicable disease and injuries. the organization of labor), reducing job turnover with This chapter takes a first step toward analyzing the issue its associated costs (Currie and Madrian 1999). in Russia. Labor supply. Somewhat counter-intuitively, economic Earlier chapters cite findings that the health status of theory predicts a more ambiguous impact of health on the Russian population compares unfavorably to those labor supply. The ambiguity results from two effects of other G-8 countries. In light of those findings, this working to offset each other. If the effect of poor health chapter describes the effect adult ill health, in particular is to reduce wages through lower productivity, the sub- that due to non-communicable disease (NCD) and stitution effect would lead to more leisure and therefore injuries, has had on the Russian economy and the eco- lower labor supply as the return for work diminishes. nomic outcomes of its population. On the other hand, the income effect would predict that as lifetime earnings are reduced through lower pro- ductivity, the individual would seek to compensate by Conceptual Framework increasing the labor supply. The income effect is likely Figure 6.1 shows the channels through which health to gain importance if the social benefit system fails to could contribute to an economy and ultimately eco- cushion the effect of reduced productivity on lifetime nomic growth. Four channels are shown, though others earnings. The net impact of the substitution and income may exist: enhanced labor productivity, greater labor supply, education and training fostering higher skills, and more Figure 6.1 From Health to Wealth (and Back) savings available for investment in phys- ical and intellectual capital. The figure Labor Productivity also shows that as an economy grows, health improves. Each channel is Labor Supply described in turn below. HEALTH ECONOMY Education 5 The Commission on Macroeconomics and Saving and Investment Health (CMH 2001) has developed a com- prehensive body of work reviewing and con- tributing to the literature on the economic benefits of improving health in low-income countries. See Suhrcke et al. (2005a) for a lit- Source: Adapted from Bloom, Canning and Jamison (2004). erature review on the economic benefits of health in the high-income context. 55 Figure 6.2 Annual Average Days of Absence Due to Illness, per Employee, Russia and EU-15 Countries 14 Male Female 12 10 8 EU-15: 7.9 (M: 6.9, F: 9.0) 6 4 2 0 2000 2001 2002 2003 Source: Suhrcke et al. 2005b. Russian data are from RLMS rounds 9-12; the Russian figures were obtained by multiplying the RLMS monthly figures by 12. EU-15 data refer to the year 2000 and are from the European Survey of Working and Living Conditions. effects ultimately becomes an empirical question (Currie also contribute to any propensity to invest in physical or and Madrian 1999). intellectual capital (Bloom, Canning, and Graham 2003). Education. Human capital theory suggests that more educated individuals are more productive (and obtain higher earnings). If children with better health and What Effect has Adult Ill Health nutrition attain higher education and suffer less from had on the Economic Outcome of school absenteeism and dropping out of school early, the Population? then improved health in youth would contribute to future productivity. Moreover, if good health is also The overarching answer from an assessment prepared linked to longer life, healthier individuals would have for this study is clear: poor adult health in Russia nega- more incentive to invest in education and training, as tively affects economic well-being at the individual and the rate of depreciation of the gains in skills would be household level (Suhrcke et al. 2005b). If effective lower (Strauss and Thomas 1998). action were taken in Russia, improved health would play an important role in sustaining high economic Savings and investment. The health of an individual growth rates at the macro-level. The main findings of or a population is likely to impact not only the level of this assessment are: income but also the distribution of income among con- sumption, savings, and investment. Individuals in good Work absenteeism. Work absenteeism due to illness is health are likely to have a wider time horizon, so their a widely used, if imperfect, illustration of the effect of savings ratio may be higher than that of individuals in illness on labor supply. Figure 6.2 shows the annual poor health. Therefore, a population experiencing a average number of days of absence from work due to ill- rapid increase in life expectancy may be expected, other ness in Russia, calculated using data from the Russian things being equal, to have higher savings. This should Longitudinal Monitoring Survey (RLMS), compared 56 with the latest-available EU-15 Table 6.1 Estimated Cost of Absenteeism Due to Illness in the data (2000). Although this indica- Russian Federation tor has a disadvantage in that it US$ billions reflects both the burden of ill health and the incentives created by Total wage loss as Total production Total production loss employment policies, it still reflects Total wage loss share in GDP (GDP) loss as share in GDP comparative lost productivity. On 2000 40.33 0.55% 97.38 1.34% average 10 days are lost per 2001 52.01 0.68% 105.17 1.37% employee per year due to illness in 2002 56.62 0.71% 104.03 1.30% Russia, while in the EU-15 coun- 2003 60.96 0.71% 112.87 1.31% tries the average is 7.9 days (Paoli and Merllié 2003). Source: Suhrcke et al. 2005b. Calculations based on RLMS absenteeism data. Note: See appendix B table B1 for details of the calculation. The annual average missed days in Russia The cost of absenteeism are obtained by multiplying the monthly RLMS figures by 12. due to ill health The international experience. Sickness absence incurs a direct cost of the sickness ben- under a range of assumptions regarding firm size, pro- efits paid to absent employees (when applicable) as well duction function, nature of the firm's product, and as the indirect cost of lost productivity. A report competitiveness of the labor market. The conclusion is (European Foundation for the Improvement of Living that the cost of lost work time can be substantially high- and Working Conditions 1997) provides information er than the wage if perfect substitutes are not available on the costs of work absenteeism: in the United to stand in for absent workers, production involves Kingdom in 1994, lost productivity due to sickness teamwork, or a penalty is associated with failing to meet absence was about US$19.7 billion. In Belgium, about an output target. US$3.5 billion was paid in 1995 in sickness benefits and benefits for work-related injuries and occupational The impact on the labor supply. Significant research diseases. In 1993, payments to cover absences from (for example, Currie and Madrian 1999; Suhrcke et al. work were assessed to be up to US$37.8 billion in 2005a) has explored the impact of ill health on the Germany and US$19.3 billion in the Netherlands (out labor market in high-income countries. Such research of which one-fourth was for benefits for sickness shows a negative impact on both labor productivity and absence and three-fourths were for disability benefits). supply. Mitchell and Burkhauser (1990) used the U.S. Survey of Disability and Work in 1978 to find that The cost in Russia. The overall cost associated with arthritis reduced wages by 27.7 percent for men and the reported workdays lost to illness in Russia varies 42.0 percent for women. Moreover, it reduced the num- between 0.55 percent and 1.31 percent of GDP (see ber of hours worked by 42.1 percent and 36.7 percent, Table 6.1), depending on the method of estimation. respectively, for men and women. Stern (1996), using Annual absenteeism rates in Figure 6.2 can be converted the U.S. Panel Study on Income Dynamics of 1981, into a monetary value either by using the average wage shows that limited ability to work due to illness reduced rate (resulting in the lower value) or the GDP per capita wages by 11.7 percent and by 23.8 percent for men and (resulting in the higher value). In 2003, the total wage women, respectively, when a selection correction for loss in Russia is estimated at US$60.9 billion, while the participation in the labor force is introduced. In addi- total production loss is estimated at US$112.8 billion. tion, the probability of staying outside the labor force This is a significant impact, given that the indicator fails increased by an estimated 13 percent. Using the same to capture the many other ways ill health impacts the data, Haveman, Stone, and Wolfe (1994) estimate that labor market. In particular, it does not capture the ill health decreased worked hours by 7.4 percent. effects of reduced productivity and mortality. In a theo- Berkovec and Stern (1991), using data from the retical model, Pauly et al. (2002) examined the magni- National Longitudinal Survey of Older Men (1966-83), tude and incidence of costs associated with absenteeism found that poor health status reduced wages by 16.7 57 percent. Baldwin, Zeager, and Flacco (1994), using data work hours supplied by men per year by 1,030 or to from the Survey on Income Program and Participation raise the probability of leaving the labor force by 42 per- of 1984, found that health limits reduced wages by 6.1 cent. A comparable effect of a 654-hour decrease or a percent for men and 5.4 percent for women. While the 31 percent increase in the probability of leaving the varying percentages from these studies lead to theoreti- labor force was found for women. cal ambiguity, at least in high-income countries there is overall more evidence of a significant negative impact of Turning to evidence from European countries, Jiménez- ill health on labor supply than on productivity (i.e., Martin, Labeaga, and Martínez (1999) found that wage rates). health,6 particularly among men, was a very relevant factor in the decision to retire and for their spouse to Among jobholders in Russia in the recent years, ill retire with them. The authors used information on labor health appears to have had a significant and sizable market transitions between 1994 and 1995 from the impact on labor productivity, but less so on supply. The European Community Household Panel, pooling data impact also seems to be more pronounced among males from across the EU, to analyze retirement patterns of than females. These findings, while slightly different individuals and couples in a sample of men older than from some in Organization for Economic Cooperation 54 years and women older than 49. Strong evidence of and Development (OECD) countries, are not necessari- the influence of health status in the retirement decision ly surprising, since the social welfare system in Russia was also found by Siddiqui (1997), using data from the operates very differently than that those in OECD German Socio-Economic Panel looking at men in West countries, affecting the relationship between health and Germany who have reached the minimum retirement the labor market. In fact, the finding of a significant age (which, given the related policies in the country is impact on the wage rate rather than on hours worked is considered to be 58 years7). Indeed, the degree of dis- evidence of health's particularly strong economic ability seems to be the dominant factor explaining early impact. The following subsection presents evidence of retirement, with the probability of leaving the labor the existence of one labor supply effect of, in particular, force at the earliest possible age for disabled men being chronic illness via its impact on early retirement. four times that of men without disability. As the author notes, these results suggest that improving employees' The impact of chronic illness on early retirement. health could be a highly effective measure to raise the Many studies in industrialized countries have shown actual age of retirement. that ill health, and in particular chronic illness, affects the decision to exit the labor force: healthier people, The impact in Russia. Applying the various approach- other things being equal, tend to retire later than less es used in other countries to Russia's case reveals a statis- healthy ones. Based on a review of various U.S. studies, tically very robust and sizable impact of chronic illness Sammartino (1987) concludes that those in poor health on both age of retirement and on the probability to are likely to retire between 1 and 3 years earlier than retire in the subsequent year. Two different, comple- those in good health with similar economic and demo- mentary approaches were followed: a Cox regression and graphic characteristics. Bound, Stinebrickner, and a panel logit regression. Controlling for other relevant Waidmann (2003), based on the analysis of data from determinants of the decision to retire (e.g., age, gender, the American Health and Retirement Study, estimated that a representative individual in poor health is 10 times more likely than a similar person in average health to retire before becoming eligible for pension benefits. 6 The health variables refer to year 1994 (to minimize the endo- Coile (2003) found that health shocks have a large geneity bias) and include the following indicators: self-reporting effect on labor supply decisions by both men and good health, self-reporting a chronic physical or mental health problem (data available only for 1995), having been admitted as women, mainly when accompanied by major changes in an in-patient during the previous year, having visited a doctor functional status. For example, the onset of a heart between one to five times in the year, and having visited a doctor attack or stroke accompanied by an important deteriora- more than five times in the year. tion in the ability to perform "activities of daily living" 7 The self-employed were withdrawn from the sample due to their (e.g., dressing) was estimated to reduce the number of different pension systems. 58 income), both approaches confirm Figure 6.3 Probability to remain in the workforce with and without chronic the result that chronic illness illness in Russia, by age - based on Cox regression model increases the probability of retiring early. The former approach assesses 1.0 the effect of chronic illness on the probability that an individual will retire in a given year after the first 0.8 year of employment; this method- ological limitation leaves the direc- tion of the causality uncertain: does 0.6 ill health predict retirement or vice function versa? The second approach address- es this limitation by examining the No chronic illness 0.4 effect of chronic illness on the prob- Survival Chronically ill ability of retiring in the subsequent year. 0.2 As shown in Figure 6.3, the Cox regression indicates that a hypothet- 0.0 ical male individual aged 55 on 0 20 40 60 80 median income and having certain Age other average characteristics 8in Russia would be expected to retire Source: Suhrcke et al. 2005. Calculations based on RLMS round 11; Cox regression results in at age 59, while a chronic illness the annex would lower his expected retirement age by two years to 57 years. While the technical details of the regression results can be diffi- than the same individual free of chronic illness. Some of cult to interpret, they are more intuitively understand- the respondents in the RLMS have been followed over able if applied to a hypothetical individual. Similar several survey years.9 This allows the use of a panel logit results are obtained for females. However, only evidence regression to assess the impact of chronic illness in one of an existing association between chronic illness and year on the probability of retirement the next: the earlier retirement is shown, since it is not possible to effects of chronic illness on the probability of entering determine the time of onset of an individual's chronic retirement in the next year are assessed, not the effect disease with the available data. In particular, whether on the probability to retire at a given year after first the illness occurred before or after retirement is employment. Otherwise, the set of explanatory variables unknown. This analysis does not indicate whether the to be controlled for is identical to the Cox model. The statistical association reflects the effect of chronic illness results (presented in appendix table B13) show a pattern on retirement or vice versa. It is, however, possible to similar to those in appendix table B15 based on the Cox address this issue by using a panel logit regression (see regressions, with only minor differences. Chronic illness appendix B table B13). emerges as a highly significant predictor of subsequent The panel logit regression results show that an individ- ual who suffers from chronic illness has a significantly higher probability of retiring in the subsequent year 9 This is the "panel" component of the RLMS, which in principle offers important opportunities for testing hypotheses that involve a causal perspective. One shortcoming of this panel dimension is that it does not feature a true panel design, as both 8 The other characteristics of this hypothetical individual are that entire households and individual members of households are not he is married, has one child, has a high school diploma, was born followed if they move from their dwelling. Nevertheless, the in Russia, and is living in an urban area. effect of attrition is relatively modest. 59 Figure 6.4 Average Predicted Probability to Retire in the Subsequent Period for speaking, this result is Hypothetical Male at Varying Income Levels in Russia (Based on reflected in the statistically Panel Logit Model) significant interaction term between income and chronic 0.7 illness in the regression 0.62 models. Figure 6.4 exempli- 0.6 Not chronically ill 0.56 fies the gradient of the impact on the basis of the 0.5 Chronically ill panel logit model: among 0.43 0.40 0.4 males with a very high 0.32 income, the presence of a 0.3 0.29 chronic illness has no effect 0.23 0.21 on retirement age, while 0.2 0.18 0.18 men just below the average of the income distribution 0.1 (i.e., in the 25th to 50th income percentile) have a 24 0 > 95th 75th - 95th 50th - 75th 25th - 50th < 25th percentage point higher income income income income income probability of retiring early percentile percentile percentile percentile percentile compared to their healthy RICHEST POOREST counterparts. As for the Cox regression, the differential Source: Suhrcke et al. 2005b. Calculations based on RLMS rounds 9-11. impact can be illustrated by Note: Results refer to the hypothetical individual described in the text. comparing the effect of chronic illness in the hypo- thetical individual described above to another with the retirement. Given the different methodology, this result same characteristics but a lower income equal to 50 per- provides a more reliable basis for claiming causality cent of the median: he will retire, on average, at 58.8 between chronic illness and the probability of retire- years without a chronic illness but at 56.3 if a chronic ment. The magnitude of its effect is large compared to illness is present, a gap of 2.5 years and 6 months earlier other variables in the model. than the richer individual (see appendix B for more detailed results). In either approach, the effect of chronic illness is found to vary with income: the lower the income the more The impact of alcohol consumption on the proba- chronic illness affects the decision to retire. This implies bility of being fired. Several studies in other devel- that less-affluent people carry a double burden of ill oped countries have shown that heavy alcohol consump- health: first, they are more likely to suffer from chronic tion has a negative impact on earnings, incomes, and illness, and second, once ill, they suffer worse economic wages, because it reduces individual productivity and consequences than rich people, a feature that tends to may create problems with working arrangements perpetuate socioeconomic disadvantage.10 Technically (Mullahy 1991; Cercone 1994). In this section these results are applied to the available Russian data by exploring whether alcohol consumption in one year 10 (2001, round 11 of RLMS) increased the risk of job loss Note that this approach does not allow exploration of a similar variation of the effect of ill health across the income scale in the in the subsequent year (2002, round 12). The rationale wage and earnings regressions presented above. This would is that job loss would be a natural consequence of an require a different approach, for instance a quantile regression appreciable reduction in individual productivity. (see, for example, Rivera and Currais [1999] for an application of quantile regressions to Brazil). 60 This assessment shows that one negative economic Effect of chronic illness on the household income impact of severe alcohol consumption, arguably an in Russia. Chronic illness has negatively impacted important contributor to premature mortality in Russia, household incomes in Russia, particularly in the period is that it significantly increases the probability of job 1998-2002. In order to address some technical con- loss. Using a panel probit model and controlling for straints on estimating the causal effect of health on eco- gender, age, education, work experience, wage rate, and nomic outcomes, mainly the issue of endogeneity of the the ownership type of the employing organization, it health proxy, a strategy was used here that differs from was found that alcohol has a positive and statistically that of the other analyses undertaken for this study. A significant effect on the probability of being fired (see difference-in-differences estimator combined with a appendix B table B16 for details and results). This may propensity score-matching technique was applied to the reflect the simplified structure of the estimated model: RLMS surveys from 1994 to 2002. Essentially, this further research would disentangle the complex but no technique allowed a comparison of pairs of households doubt important effects of alcohol on the Russian labor that are identical except for the presence of health prob- market. lems (details of the methodology and the results are described in appendix B section 5). The impact on Russian families. The death of a household member was found to increase the probabili- Using a two-step procedure, chronic illness was found ty of suffering depression by 53 percent when control- to contribute to an annual loss of 5.6 percent of per ling for other relevant factors. The panel dimension of capita median income for a hypothetical individual with the RLMS (rounds 11 [2002] and 12 [2003]) were given characteristics (i.e., living in urban areas, with no exploited, which enabled the assumption of a more smokers and no ex-smokers, no people aged over 60 or causal interpretation of the results. The sample included below 14, with at least two workers, and with at least only the individuals who were members of those house- one person who has a high school diploma). The first holds whose composition remained constant in 2002 step confirmed a negative effect of poor health (in gen- and 2003 or was altered because one or more members eral) on household income. This effect is greater in the died. Using probit analysis and controlling for relevant period 1998-2002 than before the financial crisis. A variables, the effect of a household member's death in more detailed logit model then assessed the extent to 2002 on the probability that any surviving household which chronic illness increases the likelihood of experi- member experienced a depression in the subsequent year encing adverse health events. Results show that chronic was explored (detailed results are in appendix table illness increases the risk of health problems. Combining B18). As expected, the probability of depression decreas- the effect of chronic illness and poor health on income es with the age of the deceased. The possible differences then gives the overall indirect impact of chronic illness in per capita income were also controlled for, in order to on household income. check whether depression was related to this factor rather than the death, and it appears that income differ- ences do not affect the probability of depression. Policy Implications The assessment in this chapter has demonstrated various Alcohol consumption was found to increase by about channels through which health has affected various eco- 10 grams per day as a consequence of the death of an nomic outcomes in Russia, aligning with findings from unemployed household member and by about 35 grams an increasing body of literature on health and the econ- if the deceased was employed. Using the same two years, omy in other countries, both rich and poor. In each a tobit model including essentially the same control analysis, the results proved statistically highly significant variables as in the depression model were employed. and, where it could be assessed, of a notable size. Surprisingly, if the deceased was the household head, no independent impact was found, at least not in the short This shows how seriously ill health has negatively affect- term examined (detailed results are in appendix table ed individual and household economic outcomes in B18). Russia. The findings are consistent with studies on the 61 same topic in other countries but improves on them · Policy makers interested in the economic future of since they were largely confined to issues related to Russia and its people would more likely succeed by infectious disease or child and maternal health condi- incorporating health into their portfolio of invest- tions, diseases that characterize developing countries and ment strategies. may have limited relevance to European countries. Some might have thought that diseases that allegedly · Given the magnitude of economic benefits that can strike the individual at a later stage in life would have be expected from improving adult health in Russia, only very minor economic importance, but the analyses any reasonable and well-designed increase in the here show that this hypothesis does not withstand resources devoted to health, both in- and outside the empirical scrutiny. Adult health matters not only intrin- health system, would produce a significant economic sically, but also economically. The policy implications return. The challenge is of course not solely to are: increase resources, but also to put existing resources to better use. · Investing in adult health is a sound investment strate- gy likely to yield tangible economic returns, on top of the human benefits. 62 7 Chapter 7. What Strategies and Interventions Prevent NCDs and Injuries? A number of developed countries have implement- economic determinants in an entire population (e.g., ed a full range of effective strategies to prevent information and communication programs addressing deaths and illnesses from NCDs and injuries. the risks of smoking and the value of smoking avoid- The research on effective policy measures and NCD risk ance, excise and other taxes aimed at reducing smoking factor interventions is growing steadily, enabling deci- uptake and intensity, and restrictions on smoking in sion-makers to make informed policy choices. While the public places and on tobacco advertising). The main level of impact varies by strategy, local circumstances, argument for this strategy is that it targets a high pro- and resource availability, the overwhelming evidence portion of NCD morbidity and mortality: the majority nevertheless suggests that an integrated strategy is most with risk levels near the mode of the distribution, the effective. That is, as demonstrated by the successful 70 percent group in Figure 7.1. This large group would experience of the North Karelia Project in Finland, an benefit from interventions that are far less expensive integrated strategy incorporates all appropriate actions than what high-risk group needs, making the popula- to reduce the burden of disease, including both popula- tion strategy a cost-effective way to reduce NCD rates. tion-based and high-risk prevention strategies (Box 7.1 on page 66). In the "high-risk strategy," individuals at high risk of developing selected diseases are identified, and actions The "population strategy" aims to change disease-related are planned to reduce their disease burden through lifestyles, environmental factors, and their social and provider-based interventions (e.g., clinical interventions Figure 7.1 Distribution of Risk Factors in a Population 5 percent 70 percent 25 percent People with People with average risk People with clinically low risk factor level high risk factor level factor level Individual risk of CHD Distribution of people according to risk factor level Theoretical presentation of the difference between individual risk and the proportional attributable risk Source: Rose 1992, as shown in North Karelia Project Presentation, National Public Health Institute, 2005. 65 Box 7.1 An International Good Practice: North Karelia The results: The North Karelia Project in Finland shows that major changes in mortality from NCDs can be achieved through dietary changes, increased physical activity, and reduced smoking, serum cholesterol, and blood pressure. Coronary heart disease (CHD) in adults aged 65 years and less fell by about 73 percent between 1970 and 1995. In a recent 10-year period, mortality from coronary heart disease declined by about 8 percent a year. Mortality from lung cancer declined more than 70 percent, mostly due to consistent declines in the proportion of men who smoked (from 52 percent in 1972 to 31 percent in 1997). Data on the risk factors from ischemic heart disease and mortality in Finland suggest that the changes in the main coronary risk factors (serum cholesterol concentra- tion, blood pressure, and smoking) can explain most of the decline in mortality from that disease. As a result of targeting important high-risk factors for NCDs, all causes of mortality in North Karelia declined by about 45 percent during 1970­95. In the 1980s, these favorable changes began to develop all over Finland, improving life expectancy by 7 years for men and 6 for women. The largest decline in age-specific mortality was reaped by the 35- to 44-year-olds: men in this age group saw an 87 percent decline in mortality from CHD between 1971 and 1995. Men 35­64 saw age-adjusted mortality rates decline from about 700 per 100,000 popu- lation in 1971 to about 110 per 100,000 in 2001. This rate for all of Finland among men in the same age group was about 470 per 100,000 and fell 75 percent. These improvements in life expectancy are correlated with signifi- cant declines in the amount of saturated fats consumed, coming mainly from milk products and fatty meat (satu- rated fat consumption dropped from about 50 gr/day in 1972 to about 15 gr/day in 1992) and significant reduc- tions in blood cholesterol levels (from about 7mmol/L in 1972 to about 5.6 mmol/L in 1997). Success factors: Appropriate epidemiological and behavioral framework Restricted, well-defined targets Good monitoring of immediate targets (behaviors, process) Flexible intervention Emphasis on changing environment and social norms Working closely with the community Positive feedback, work with media International collaboration, support from WHO Close interaction with national health policy, integration with National Public Health Institute Long-term, dedicated leadership Source: Puska et al. 1995 to treat and counsel individuals about risk factors for a major task for health services, and their per-person CVD--smoking, excessive alcohol consumption, hyper- costs can be high. tension, hyperlipidemia, diabetes and obesity; manage- ment of patients at high risk according to established Data from North Karelia reveal that results from pre- clinical practice guidelines; and therapy for individuals vention efforts may appear in years rather than once overt CVD has occurred). High-risk strategies pose decade--improvements occur some 2-7 years after the 66 elimination of the exposure to a risk factor, and that they are beneficial even for people in older age groups. Measures to control alcohol supply include government monopolies or licensing over sales of alcohol, regulations restricting availability (i.e., limits on the times and con- International Evidence on Effective ditions of sales or service, minimum-age limits), and Interventions to Control Individual restrictions on drinking alcohol in public. About 18 Risk-Factors Associated with the countries surveyed in the WHO Global Status Report Onset of NCDs and Injuries on Alcohol have government monopolies that could raise prices and restrict availability on outlets and hours Interventions to Reduce Alcohol Abuse of sale. About 30 countries have high levels of restric- Epidemiological research has shown that patterns and tions, particularly in the Americas region. The United levels of alcohol consumption, alcohol dependency, and States, for instance, has regulations at state and local lev- alcohol abuse are determined by many factors: availabili- els to license or restrict sales. However, evidence is ty; income per capita; retail process, individual factors unclear on the effects of restricted time and somewhat (genetic and environmental), such as age of first use, mixed on the effects of restricted outlets (Anderson and family history, education, and peer group pressure; psy- Lehto1994). More than 100 countries have a minimum chosocial factors, cultural and historical context; and age limit for alcohol purchases. Research shows that age government policies, such as taxation and restrictions on limits can reduce harm to young people, as in the advertisement and promotion. The most effective United States where traffic accidents declined when the approach to reduce alcohol-related problems would be a age limit was raised (Wagenaar 1993). About 60 coun- comprehensive set of measures aimed at these factors. tries have a total ban on alcohol in official settings such as government offices, education and health institutions, On the demand side, interventions shown to be effec- workplaces, and public transport. Since alcohol-free tive at reducing consumption include price increases, environments are locally enforced, less is known about public education campaigns, and counseling or brief the impact of this approach. Regulations have also been intervention. Price increase at a national level through enacted to ensure product quality and to restrict the sale taxation is the most cost-effective measure and has the of surrogate alcohols. These measures need to be largest and quickest impact on habits, especially for enforcement to be effective. heavy drinkers and young people, while raising revenues with low transaction costs (Babor et al. 2003; Other alcohol harm reduction interventions proven Chaloupka, Grossman, and Saffer 2002). Many Nordic effective are strong laws against drinking and driving. countries have very high taxation rates on alcohols and These laws include low Blood Alcohol Content (BAC) have successfully controlled per capita consumption. limits, visible and frequent enforcement, and license Education campaigns can be useful for raising the suspension in case of an offence (Babor et al. 2003). A awareness of risks and effects of behavior change if sup- combination of these approaches has been shown to plemented by interpersonal counseling or targeted at reduce both the number of cases of drunk driving and specific risk groups (Raistrick, Hodgson, Ritson 1999; alcohol-related traffic accidents (Chaloupka et al. 2002). Holder 1994). For example, media advocacy has been Overall, almost all countries have a defined legal BAC effective in reducing illegal sales to minors, reducing limits for driving, and more than half perform random alcohol availability on campuses. and building support breath tests. for alcohol control policies in the United States. Individual counseling or brief intervention from health Among the mix of interventions, less effective are policies professionals was found effective at reducing consump- to restrict alcohol advertising and school-based education tion or improving alcohol-related problems in a WHO programs. There is no conclusive evidence on either cross-national study, a meta-analysis of 14 data sets, and method due to implementation and evaluation difficul- other studies (Babor and Grant 1992; Babor et al. 1996; ties. More than half of the countries the WHO surveyed Poikolainen 1999). Brief interventions are most effective have no restrictions on advertising in print or billboards. for those seeking to quit, but for the general population, Broadcast media and advertisements on spirits are some- they are not cost-effective. what more restricted. However, since exposures through 67 marketing are likely to shape positive perceptions of has been consistently found to reduce overall consump- drinking, the WHO recommends that countermeasures tion in North America, Australia, Europe, and Israel. An to advertising be part of a comprehensive alcohol policy. analysis of 100 countries found that those countries Similarly, although a systematic review of studies aimed at with a comprehensive ban on tobacco advertising have a educational and psychosocial prevention programs reveals steeper decline in consumption trend over time than a general ineffectiveness (Foxcroft et al. 2002), these pro- those without (Saffer 1995b). Since the 1970s, many grams nevertheless complement other efforts to prevent countries have banned advertising and promotion, and the initiation of drinking among youth. in the mid-1990s, the restrictions on advertising became more comprehensive, such as in the European Union (EU). There is also a growing trend in a number of Interventions to Reduce Smoking Prevalence Western countries to restrict smoking in public settings, A comprehensive approach combining economic and especially workplaces, restaurants, and transport facili- regulatory measures with community-wide programs has ties. In the United States, smoking bans in public areas been identified by the U.S. Surgeon General to have the have led to quit rates between 4 and 10 percent (U.S. greatest long-term population impact (U.S. HHS HHS 2000). 2000). Countries with successful tobacco control poli- cies have focused on measures to reduce both the There is less evidence showing success in reducing the demand for and supply of tobacco. On the demand supply of tobacco. Many countries relied on trade side, many studies concluded that increasing the price of restrictions or agricultural policies, including restricting tobacco and taxes does significantly reduce consump- youth purchases of cigarette products; crop substitution tion, especially among young people, and higher taxes and diversification; controlling imports; and restrictions have reduced cigarette consumption in Canada, South on international trade. These measures are harder to Africa, United Kingdom, United States, Thailand, and enforce due to the black market supplies and the ease of many other countries. It is expected that higher taxes are smuggling. The WHO is leading an effort under a glob- more effective in low- and middle-income countries al treaty on tobacco control. It is expected that countries where smokers are more responsive to price increases joining the treaty will be supported internationally on (World Bank 1999). Government revenues from tobac- their tobacco control policies. co taxes can be applied to anti-smoking activities, such as health education, health services, or counter-advertis- Finally, individual- and community-based programs ing campaigns, as being done in parts of the United such as cessation treatments and counseling are promot- States and China. ed in Western countries and are found to be both effec- tive and cost-effective (West, McNeill, and Raw 2000). Non-price control measures and policies, such as con- Nicotine replacement therapy products can double the sumer information, bans on advertising, and smoking success rates of other cessation efforts (Raw, McNeill, restrictions in public places, have also proven to exten- and West 1999). Telephone help lines or quit lines, sively reduce demand for tobacco. In the United States, along with mass media campaigns, have achieved quit smoking prevalence fell from above 40 percent to about rates of 15.6 percent in England and 28 percent in 25 percent between 1950 and 1990. This period coin- Australia (Australia Commonwealth Department of cided with an increase in people's knowledge of the Health and Aged Care 2003; Owen 2000). On the harmful effects of tobacco from below 45 percent to 95 other hand, community-based programs sometimes have percent. The "information shocks" from publicized difficulty with sustainability; when properly supported research on the harmful effects of tobacco has found by local resources and networks and partnerships, their positive impact in countries such as Finland, Greece, results are much more promising (U.S. HHS 2000). Switzerland, Turkey, the United Kingdom, the United States, and South Africa. Warning labels on cigarette packs caused consumption to fall in Turkey, Australia, Interventions to Prevent Road Traffic and Canada, and Poland. Similarly, mass media campaigns Other Injuries aiming to counter advertising by the tobacco industry Interventions for road safety span a wide range. Proven 68 approaches fall into four broad categories: (1) vehicle fic rules such as speed limits, BAC limits, and, lately, design and safety equipment (i.e., helmets and seat cell phone use restrictions. Speed limit changes have belts), (2) road user measures (e.g., speed limits and been found to reduce the number of fatalities by a range restrictions on drinking and driving), (3) engineering of 6 to 24 percent in Switzerland, Denmark, and measures that improve road design, and (4) traffic man- Sweden, depending on the size of the reduction agement and reducing exposure. To minimize traffic (European Transport Safety Council 1995). injuries for which prevention fails, effective post-crash measures are also important. Engineering technologies and road designs that improve infrastructure, such as better geometric design, intersec- Vehicle safety design helps the driver avoid a crash or tion control, safety or crash barriers, signs, markings, protects drivers and victims against injury. Safety design and maintenance are effective intervention strategies features include daytime running lights and impact pro- widely implemented in developed countries (World tection (air bags or seat belts). Research indicates that Bank 2003a). For example, collapsible lighting columns crash protection design is the most effective strategy for and other devices that break away on impact are widely reducing traffic deaths and serious injuries. A review in used, and cable barriers have been used cost-effectively the United Kingdom found that crash protection in Denmark, Sweden, Switzerland, and the United improvements accounted for 15 percent of reduction to Kingdom (European Transit Safety Council 1998). casualties, compared with 11 percent for drinking-driv- Some of these crash-protective roadside objects, includ- ing measures and 6.5 percent for road safety engineering ing guard fences, crash cushions, and slip base poles measures (Broughton et al. 2000). The European could result in a 30 percent reduction in injuries (Cirillo Transport Safety Council estimated that improved vehi- and Council 1986). Other low-cost traffic engineering cle safety standards could reduce deaths and injuries by measures that are highly cost-effective are skid-resistant as much as 20 percent (European Transport Safety surfacing, central refuges and islands, and signal control Council 1993). or operation of junctions. Strong evidence also shows that laws requiring the use Rules that require safety awareness in engineering and of seat belts and helmets to be effective. Mandatory seat road planning as well as safety audits exist in many belt use laws accompanied by strict enforcement are the countries including the Netherlands, Australia, most successful in preventing deaths, as demonstrated in Denmark, New Zealand, and the United Kingdom. a 30-year study in Finland and the United Kingdom Safety audits and systematic safety impact assessments (European Transport Safety Council 1996; Ashton, were shown to have a cost-benefit ratio of 1:20 in New Mackay, and Camm 1983). The United States, Zealand (Macaulay and McInerney 2002). Australia, New Zealand, the United Kingdom, and many other European countries have seat belt enforce- The least-used and least-evaluated road safety interven- ment laws allowing the police to stop a motorist for not tion strategies are those that aim to reduce exposure to wearing a seat belt (primary enforcement), which risk. It has been estimated in high-income countries that increases the laws' effectiveness (Dinh-Zarr et al. 2001). under certain conditions, for each 1 percent reduction Similarly, studies show that mandatory cycle helmet in distance traveled, there is a corresponding 1.4-1.8 laws reduce the number of head injuries by around 25 percent reduction in crash incidence (Litman 2000; percent (Elvik and Vaa in press). The helmet law was Edlin 2002). It follows that policies that improve trans- responsible for a 56 percent reduction in traffic fatalities port network and efficient land use to reduce distances in Thailand and 30 percent in Malaysia and the United and the necessity to travel make sense. Steps include States (WHO 2002b). providing shorter, safer routes; better management of commuter transport; better mass transit; and prioritiz- International experience and research have proven that ing or restricting access to different parts of a road net- other measures aimed at drivers and pedestrians are also work (preventing pedestrians and cyclists from motor- effective, especially with good enforcement (Elvik and ways or preventing vehicles in pedestrian zones). Many Vaa in press; Finch 1994). These measures include traf- countries are now focusing on transport and land use 69 policies for sustainable mobility; Colombia is a specific comparison with communicable diseases, and much less example (Rodríguez, Fernandez, and Velásquez 2003; than would be justified in terms of their overall contri- Trans-Milenio 2001). bution to the global burden of disease. This mismatch is especially great in eastern Europe, where NCDs pre- Finally, improvements in post-crash care can prevent dominate. deaths and limit the severity of injuries. Areas include faster access to pre-hospital, emergency room, and trau- As the etiology of most NCDs is complex, so risk fac- ma care. In high-income countries, a standard emer- tors should be managed from multiple directions. The gency telephone number is available for public use to government's role in disease prevention should be multi- readily call ambulance, police, and firefighters. faceted, including setting priorities, formulating policies Additionally, many improvements have taken place in and enacting and enforcing regulations,, rallying con- the West over trauma care with regard to clinical capa- stituencies for health promotion and disease prevention, bility as well as equipment and supplies. However, establishing infrastructure and institutions, and educat- investments into emergency or trauma care should be ing the public. Many disease-prevention efforts also decided on the basis of the resources of individual coun- require larger health system reform, for example, creat- tries (Peden et al. 2004). ing financial incentives to strengthen public health pre- vention and develop community-based intersectoral interventions to benefit the majority of the population. The Government's Role in Disease Prevention Role of Private Employers From an economic perspective, government interven- Private sector involvement is particularly important. The tion is justified as a means to achieve a net improvement poor health of employees quickly affects a company's in social welfare, in terms of increased efficiency and for bottom-line and has a longer-term impact on profits. reasons of equity through redistribution (Barr 1994). Businesses have a vested interest in supporting activities Alternative government interventions, such as regula- to improve employee health and can have a strong influ- tions, taxation, and public provision, need to be assessed ence on their employees' behavior and make them aware to determine their costs and benefits, as sometimes gov- of health risks in ways unavailable to the government. ernments lack the capacity to correct market failures. In The involvement of major Russian companies, multina- the health sector, there are strong reasons to support tional corporations, and other stakeholders with experi- government intervention where information is incom- ence in employee- and community-directed health pro- plete or imperfect, and as a result individuals do not grams will be critical in reducing NCDs and injuries. clearly perceive the costs of their actions to themselves or others (Cutler and Kadiyala 1999). Some argue that The concept of employers playing a larger role in in the areas of behavior and lifestyle, a wider govern- improving employee fitness and health is not new. The ment role is justified when health benefits outweigh the U.S. Government is encouraging employers to invest in curtailment or modification of individual choices workplace health promotion, and about 95 percent of (Musgrove 1996). its large employers and one third of its smaller ones offer wellness programs. In the case of NCDs, a strong case could be make for government intervention as a result of a widespread pol- A growing awareness of the costs linked to risk factors icy failure that has often ignored these conditions rela- provides the grounds government promotion of work- tive to communicable diseases. That is, despite the obvi- place-based initiatives. Johnson & Johnson launched a ous epidemiological trends, the policy response, in terms frequently cited model of employer-based health promo- of NCDs prevention and control, has been inadequate tion in the 1970s and reports having saved US$38 mil- and only a very few countries have implemented com- lion in health care costs between 1995 and 1999 by pro- prehensive policies. In addition, official development moting healthy lifestyles (Zeidner 2004). During the assistance earmarked for NCDs is extremely low in 70 1990s, the firm attributed annual savings of US$225 per employee to intervention programs aimed at exer- PacifiCare recently offered US$390 a year to employees cise, smoking, fiber, cholesterol, and blood pressure. to encourage them to eat better, exercise, and reduce smoking or drinking. Participants record their daily Citibank offered employees modest financial compensa- food intake and exercise routine. The company expects tion for completing a health risk appraisal. Participants the program, not yet evaluated, to return more than it with risk factors were selected to receive educational costs within two years. materials and were monitored by a health counselor. The program saved five dollars $5 for each one spent. 71 8 Chapter 8. How Is Russia Responding to NCDs and Injuries? I t is assumed that over the last three decades the ment of federal bodies (e.g., research institutes and Soviet health care system lagged behind that of west- tertiary referral facilities). MOHSD was designated as ern countries in terms of health care outcomes as the responsible for developing state policy and normative country was not able to take full advantage of new med- and legal regulations in such areas as health protec- ical knowledge and technological advances, including tion, social development, labor, occupational health new treatment regimes and drugs (Andreev et al. 2003). and safety, sports, tourism, and consumer rights. A series of health reforms were initiated in Russia dur- · The Agency on Health Care and Social Development ing in the 1990s to address this situation, mainly focus- ing on financing and delivery of services. Although became responsible for providing state services and these reforms changed the health sector's organizational managing state assets in the health and social devel- and financial structures and contributed to some impor- opment sector, including provision of medical and tant legislation for addressing NCDs, the public health resort services, rehabilitation of handicapped persons, system has largely retained its infectious disease focus. and social protection of the population. Recently, the Russian government set up institutional · Two federal services were created to address issues arrangements to respond to NCDs and injuries. related to public health: However, improved institutional capacity is still needed to make health promotion and disease prevention pro- The Federal Service of Surveillance in the Area grams more effective. Indeed many health experts, of Consumer Rights' Protection and Human including Russian government representatives, agree Welfare is to perform control and monitoring that the Russian health system and the Ministry of functions to safeguard sanitary conditions and Health and Social Development (MOHSD) are not yet public health and to protect consumer rights. For equipped to tackle NCDs and injuries effectively.11 example, this service registers imported food and is responsible for identifying causes and conditions for infectious disease and for some non-infectious Recent Organizational Reforms conditions (e.g., poisoning). Government reforms changed the organizational struc- The Federal Service for Surveillance in the ture of the federal health sector in March 2004, espe- Area of Health Care and Social Development is cially at the federal level, as follows: responsible for surveillance, monitoring, and con- trol in such areas as pharmaceuticals, some aspects · The Ministries of Health, Social Affairs, and Labor of occupational health and disability due to acci- were merged in 2004 to create the Ministry of Health dents at productive enterprises, and setting stan- and Social Development (MOHSD). The new dards for and assuring quality of medical care. Ministry gained an enhanced policy-making role while losing many traditional functions, such as epi- These organizational reforms have opened a window of demiological surveillance and control and manage- opportunity for programs aimed at public health as the new MOHSD is expected to assume greater responsibil- ity beyond managing medical care. However, the new 11 Opinions expressed during a World Bank­sponsored senior organizational structures for health and other sectors are policy seminars on public health and disease control for Russian Government and other international and bilateral not yet consolidated, hindering multisectoral arrange- agencies in 2002­03. ments for implementing health promotion, health pro- 73 tection, disease prevention, and control of NCDs and political influence rather than scientific evidence; and it injuries. President Putin has repeatedly criticized health lacks enforcement mechanisms. Implementation of tar- care reform's failure to produce significant improve- geted programs is problematic and hampered by budget- ments in public health, evidenced by Russia's trailing ary legislation, limiting legitimate investment in capaci- many countries in key health indicators, particularly ty development for population-level interventions to among adults. prevent NCDs. As staff costs for these programs often cannot be justified under laws and regulations, funding is usually tied to purchasing goods. The situation is The Legal Framework for Action aggravated by the limited knowledge of health promo- Against NCDs and Injuries tion among curative care-oriented medical and other health professionals. Even ongoing prevention efforts are The health sector is regulated by more than a hundred federal laws.12 Targeted programs (Table 8.1) are very medicalized: an example is the expensive but inef- fective "despansarization" of all children. These routine, anchored in federal laws that mandate program imple- organized programs have all children undergo a day- mentation. Other legal documents supporting these long examination by 5 to 12 specialists. Children are programs are (a) "Concept of Health Care and Medical screened for all diseases in a process lacking standards Science Development in the Russian Federation," and reference to disease prevalence, and medication is approved by the Russian Government in November often prescribed without appropriate diagnosis. 1997 and the first legal document specifically targeting health that was approved by the Government; (b) Issues related to occupational health are addressed in a "Concept of the Russian Population Health Protection," wide range of ordinances, including the Labor Code, a approved in August 2000; and (c) "Concept of federal law "On Mandatory Social Insurance from Demographic Policy of the Russian Federation to the Accidents at Industrial Enterprises and Occupational year 2015," which is also directly linked to adult health, Diseases," and health ministry orders issued in the health prevention, and promotion and which was dis- 1980s and 1990s. Some of these ordinances are now cussed at a government meeting in 2001. outdated. A number of new ordinances are under dis- cussion, including the draft of a proposed program, Although the legal framework covers most major areas "Health of Working Population in the Russian of public health, it has several weaknesses: it is vague Federation for 2005." and declarative, its legislative bases were often driven by Russia's 2001 Federal Law on Tobacco is weak by inter- national standards, and compliance and enforcement are 12 Legislative bases on citizens' health protection include laws known to be low generally. Some MOHSD and non- "Sanitary and Epidemiological Welfare of the Population"; "On Health Insurance of the Population in the Russian Federation"; governmental organization programs endeavor to edu- "On Prevention of Spread of Disease Caused by HIV in the cate and inform the public, discourage tobacco use, and Russian Federation"; "On Narcotic and Psychotropic encourage cessation, but they are small relative to the Substances"; "On Pharmaceuticals"; "On State Regulation in the Area of Genetic Engineering Activities"; "On Radiation public health importance of this risk factor (Ross 2004; Safety of the Population"; "On Protection of Population and Gerasimenko and Demine 2001; McKee et al. 1998). Territories from Emergency Situations of Natural and Low prices and taxes make cigarettes more affordable in Technological Causes"; "On Mental Health Care and Russia than in most other countries (Guindon, Tobin, Guarantees for Citizens' Rights When Delivering It"; "On Blood and Its Components' Donation"; "On Transplantation of and Yach 2002). In December 2004, President Putin Organs and/or Human Tissues"; "On Natural Treatment signed a law strengthening the Law on Tobacco Resources, Treatment and Rehabilitation Territories and Consumption to make it illegal to sell tobacco products Resorts"; "On Safe Circulation of Pesticides and Agricultural in health, cultural, and educational facilities and to Chemical Substances"; "On Atmospheric Air Protection"; "On Main Guarantees of Children's Rights in the Russian smoke in workplaces, public transport, closed sport Federation"; "On Immunological Prevention of Infectious facilities, health and educational facilities, and govern- Diseases"; "On Quality and Safety of Food Products"; "On ment buildings except in designated smoking areas. Prevention of TB Spread in the Russian Federation"; "On Limitations of Tobacco Smoking"; and "On Temporary Ban of However, Russia has not signed yet the World Health Human Being Cloning." Organization (WHO)-sponsored Framework 74 Table 8.1 Active Federal Targeted Programs in Public Health in the Russian Federation Program and budget Objectives Activities Main program developers "Prevention and Control To decrease morbidity, disability, and Prevent, diagnose, and treat socially Ministries of Health; of Justice; of of Socially Significant mortality due to socially significant dis- significant diseases Agriculture; of Industry, Science Diseases" (2002-06) eases (diabetes, tuberculosis, HIV/AIDS, and Technology; of Education; and Budget: US$122.2 million sexually transmitted diseases, etc.) of Nuclear Energy; Russian Academy of Medical Science "Prevention and To improve the prevention, diagnosis, Establish hypertension prevention, moni- Ministries of Health; of Education; Treatment of Arterial and treatment of hypertension and reha- toring, and reporting systems; strength- and of Industry, Science and Hypertension in the bilitation of patients with complications en cardiology facilities; train medical Technology; Russian Academy of Russian Federation" personnel Medical Science (2002-08) Budget: US$1.6 million "Ecology and Natural To protect natural resources, stabilize Establish a national environmental Ministries of Natural Resources Resources" (2002-10) and improve the ecological situation in monitoring system, introduce meas- and Agriculture, State Committee Budget: US$42.8 million the country, and prevent environmen- ures to reduce pollution, and develop on Fishing, Federal Hydrometeor- tal degradation national parks ology and Environmental Monitoring Service, Federal Service for Geodesy and Maps "Nuclear and Radiation To reduce the risk of radiation to people Establish a national nuclear monitoring Ministry of Nuclear Energy Safety of the Russian and their homes to a socially acceptable and reporting system, train personnel, Federation" (2000-06) level, resolve nuclear waste disposal and establish occupational health pro- Budget: US$7.1 million issues, and introduce safety measures grams at nuclear enterprises at nuclear enterprises "Risk Reduction and To increase protection from emergency Implement measures to forecast and Ministry of Emergency Situations Mitigation of Impacts situations for populations and territories, monitor emergency situations and train and the Russian Academy of from Emergency (Natural develop a methodological and scientific the general population and specialists to Science and Technological) basis to manage risks of emergency sit- manage emergency situations Situations in the Russian uations, and strengthen training pro- Federation until the Year grams 2005" Budget: US$3.3 million "Russia's Youth" To develop legal, economical, and orga- Create job opportunities for youth and (2001-05) nizational conditions to bring up youth support youth sports, civil organizations, Budget: US$29.1 million as part of democratic society and as educational programs, and health pro- active participants in transformation grams; promote and strengthen the fam- processes in the country ily support system "Children of Russia" To improve children's quality of life and Oversee maternal and child health pro- Ministries of Labor and Social (2003-06) health status grams, provide social services to dis- Development, of Health, and of Budget: US$74.5 million abled children, train and retrain special- Education ists, and develop and implement an information system "Comprehensive Measures to Counteract Drug Abuse and Their Illegal Turnover" (2005-09) Budget: US$17.9 million Source: Russian MOHSD documents. 75 Convention on Tobacco Control, an international treaty to limit tobacco use. In addition, President Putin has The Centers for Medical Prevention are charged with not signed the law "On Limitation of Sales and NCD prevention and control in the regions. Despite Drinking of Beer and Drinks Produced on its Basis"; being a wide network and often nearly the only agency comments have been submitted to Parliament to active in NCD prevention, they are not active in many strengthen it, which may cause delay. regions. Separated from the rest of the health care com- munity, they suffer from having a low priority status and from an inability to influence regional policy, with Institutional Weaknesses in some notable exceptions (e.g., in Vologda and Preventing and Controlling NCDs Primorsk). Their ability to influence populations and communities is limited by funding constraints and Several factors hinder Russia's progress in developing spending regulations. Restrictions on the degree to institutions and in applying effective approaches to which local policies can diverge from national ones also health promotion, health protection, and disease pre- hinder progress. Regional health authorities often see vention for NCDs. These factors are largely associated their responsibility as running public health schools, with insufficient action to include public health as part which are predominantly treatment oriented. The local of ongoing health system reforms. This failure implies level largely lacks structures systematically designated to that the country has not shifted from the system of prevent NCDs. Furthermore, medical institutions are infectious disease control to a system of NCD preven- not reimbursed for providing preventive services. tion and control that requires new approaches and methods (MOPH 1997). What Are the Main Problems? The federal level is responsible for "the organization of the State Sanitary Epidemiological Surveillance Although established structures and ongoing efforts (SanEpid) System, development and approval of federal could contribute to a solid foundation for NCD preven- sanitary regulations, norms and hygienic standards, sani- tion and control in Russia, progress is hindered by sev- tary-epidemiological surveillance, and organization of eral factors: the system for sanitary protection." Traditionally, the SanEpid System has concentrated on infectious disease Insufficient coordination of NCD prevention and prevention through an environmental approach (e.g., control activities. The roles and responsibilities in enactment of regulations, inspection system, and penal- NCD prevention and control at the federal, regional, ties to prevent mainly food- and water-borne diseases and local levels, as well as among the national institutes, and limit environmental pollution and workplace haz- are not well defined or communicated. Coordination is ards). This approach was justified in the past, but especially important, since anti-NCD programs require changing disease patterns call for new forms of involve- multi-sector collaboration. ment with populations and communities served. Limited workforce for NCD prevention and con- Both health care and academic institutions are also trol. Human resources well trained in NCD prevention poorly equipped to address NCDs. Overspecialization and control are lacking. Knowledge and skills relevant of both doctors (there are over 120 narrow specialties in to NCD prevention and control need to be taught in Russia) and facilities hinders disease prevention and medical academies, schools of public health, preventive control efforts following population-based, multisectoral medicine and nursing, post-graduate courses, and degree approaches. Several federal research institutions have programs such as MPH and DrPH. Urgently needed are been tasked with various aspects of NCD prevention, public health courses of various formats and foci target- including the National Center for Preventive Medicine ed to a range of trainees, including (but not limited to) under the MOHSD, the Institute of Information and MOHSD staff, national institutes' staff, primary health Organization of Health Care, and the Institute of Public care providers, specialist physicians, nurses, sociologists, Health. However, the role and influence of these institu- psychologists, administrators and managers, journalists, tions in policy formulation and program development health statisticians, and public relations professionals at are limited. all levels. 76 and limited in scope and coverage. Monitoring and Inadequate access to information on NCD preven- reporting surveillance data, coupled with research and tion and control. Regional and local staff have had evaluation activities, are necessary in developing relevant limited training in modern approaches to NCD preven- health policies and effective programs at the federal and tion and control and lack ready access to information. regional levels. Lack of English proficiency is a common barrier to international sources of information. Numerous exam- ples of scientific methodologies, programs, and evalua- Some Good Practices in Russia tion results at regional and local levels exist, but infor- The CINDI Experience mation about experiences and good practices in health promotion is not well shared. Many regional and local The Countrywide Integrated Noncommunicable staff start from scratch and rely on self-learning, and Disease Intervention (CINDI) Program is a multi-coun- some regional staff have developed materials and inter- try initiative of the WHO and has been actively imple- ventions on their own. Feedback or mentoring from the mented in Russia since the mid-1980s. The National federal level, now limited, would propel such efforts: the Center for Preventive Medicine under the MOHSD lack of federal leadership leads to high variability in manages CINDI, which covers 18 regions13 in Russia content and approaches, whereas tested, effective, well- with four new regions having applied to join. The cen- focused, and consistent messaging for the target popula- ter works actively in the areas of disease prevention and tions is required. Federal institutes could be natural health promotion. Its most recent publications and nodes for collaboration and information sharing. studies include (a) guidelines for monitoring behavioral population risk factors of NCDs, (b) a reference book Treatment bias. Health care providers tend to treat for physicians on adult nutrition, (c) methodological patients only for the condition that stimulates the recommendations on health care development for patient's visit. Providers need to be trained to watch for strengthening health and preventing NCDs, and (d) and ask about other symptoms that would signal anoth- research on smoking in Russia. Other organizations er condition(s) and to talk to patients about risky contributing to this work are the Science and Research behaviors and exposure. Institutes on Occupational Health; on Human Ecology and Environmental Hygiene; on Nutrition; on Social Limited funding for NCDs prevention activities. Hygiene; on the Economics and Management of Health As the resources allocated to NCDs prevention activities Care; and on Epidemiology and Microbiology; the are low, the development of a financial policy is required Scientific Center for Mental Health; and the Scientific for supporting integrated prevention programs and Center for Rehabilitation and Resorts. activities at the federal and regional levels. Given the magnitude of economic benefits that can be expected CINDI aims to support the reduction of smoking, from improving adult health in Russia, any reasonable unhealthy nutrition, alcohol abuse, physical inactivity, and well-designed increase in the resources devoted to and psycho-social stress; enhance preventive practices of health, both inside and outside the health system, would health professionals; and ensure success by exchanging produce a significant economic return. The challenge is information, sharing experiences, and building interna- of course not solely to increase resources, but also to put tional networks. One of CINDI's major contributions is existing resources to better use. data collection and analysis on cardiovascular diseases (CVDs), which are not otherwise routinely performed Inadequate NCD surveillance capacity and in Russia. In addition, the program recently introduced research efforts. A surveillance system that tracks new methods for the collection and analysis of data, trends and reports cases is a key public health tool for enabling evaluation of the process of development and preventing and controlling NCDs, but Russia's capacity for implementing, monitoring, and evaluating NCD 13 prevention and control is weak. Several regions have The CINDI regions are Chelyabinsk, Electrostal, Krasnodar, Kostomuksha, Mirnyi, Novosibirsk, Orenburg, Pitkyaranta, tried to collect data on behavior and risk factors, but Rostov on Don, Pontonnaya,Tomsk, Tver, Perm, Bijsk, Ufa, epidemiological surveys are intermittent, underfunded, Murmansk, Verkynyaya Salda, and Vologda. 77 implementation of preventive measures (MOPH 1999; Regional Efforts CINDI 2004). A good example of the application of Beside the CINDI projects, some innovative cross-sec- these research tools is in Pitkaranta, Karelia Republic, toral approaches are starting to be developed in different where a health surveillance program was developed and regions of the country. Among the leading regions in implemented over the last 10 years through collabora- terms of promoting healthy living is the Chuvash tion among the Health Ministry of Karelia, the Central Republic, where the regional government has taken a Hospital of Pitkaranta, the National Public Health leading role in establishing an enabling environment for Institute in Finland, and the North Karelia Project. health promotion and disease prevention involving pub- Surveys of risk factors (a self-administered question- lic and private institutions, community organizations naire, physical measurements, and laboratory tests) are and individual citizens (Box 8.1). The Tula Oblast pro- undertaken by trained teams of physicians and nurses vides another good example (WHO 2005). As a result every five years. In addition, health behavior surveys are of efforts initiated in 1998 and supported by the conducted by a mailed, self-administered questionnaire Central Public Health Research Institute of the Russian every two years (Laatikainen et al. 2005). These efforts MOHSD and the Tula Oblast Health Authority, togeth- have developed a comprehensive, reliable database on er with international partners, a 70 percent success rate chronic disease risk factors and related health behavior in controlling high blood pressure, a 85 percent reduc- among Pitkaranta adults. Changes in CVD risk factors tion in admissions for high blood pressure, and net sav- and health behaviors were observed in this region over ings for overall high blood pressure care costs of 23 per- the 1992-97 period (Laatikainen et al. 2002b). cent, were achieved. Other activities developed in the CINDI regions NGOs Efforts include: In Russia, a number of NGOs, such as the Open Healthy nutrition policy. Capacity development for Society Institute and its spin-off, the Open Health nutrition assessment among adults and children. Institute, have since 1998 implemented activities aimed at developing public health training. Nearly 30 special- Smoking. The population-based Quit&Win smoking ists have been trained as Masters of Public Health. cessation campaign was launched in 2002, covering 28 Graduates of international programs and other interna- centers at regional and local levels and more than 28 tionally recognized researchers have formed a number of million people, including 7 million smokers. More than associations, e.g., "For Public Health Support" and 40,000 smokers and more than 3,000 health profession- "Society of Evidence-Based Medicine Specialists." The als participated. Moscow Medical Academy is running a full-time, two- year program in Public Health and Health Behavioral risk factors. Guidelines to monitor behav- Management. The St. Petersburg Medical Academy of ioral risk factors have been developed and adopted by Postgraduate Training, Tver Medical Academy (TMA), the MOHSD for implementation countrywide. and Urals Medical Academy of Additional Education (UGMADO in Chelyabinsk) are implementing post- Hypertension management. The 11-year collabora- graduate training programs and courses similar to tion between CINDI Pitkaranta (Karelia) and CINDI European schools of public health. All four academies Finland on blood pressure control first showed signs of provide training for specialists with different back- success in 2002, with lower systolic blood pressure in grounds. The Moscow Medical Academy has set up these populations. departments of Law, Economics, and Sociology with the involvement of non-medical professionals. TMA and The CINDI network has also facilitated international UGMADO are providing problem-oriented courses for collaboration, clearly beneficial in many ways: in build- representatives of their regional legislative authorities ing research capacity, in health promotion planning, and and other non-medical professionals. in stimulating participation in international activities and funding. 78 Box 8.1. Chuvash Republic: Innovative Approaches to Healthy Living Chuvash Republic is a leading region in Russia in terms of promoting healthy living. In February 2004, its president signed a decree entitled "On Organizing a Movement 'Chuvash--Healthy Region.'" This decree was intended to create optimal social and economic conditions to improve the population's quality of life, labor growth, and healthy living. It declared several objectives as the priorities for state policy: (1) strengthening population health, (2) environ- mental improvements, (3) forming high moral values and culture, and (4) personal responsibility for health. The decree tasked the Cabinet of Ministers of Chuvash with developing the Concept of Health Protection for Healthy People in the republic for 2004­10 and a health promotion strategy for youth. The documents promote a multisectorial approach to lifestyle, environment, and health protection and call upon local and municipal self-gov- ernance bodies to develop long-term programs for health protection and improvement of living conditions. A management council for "Chuvash--Healthy Region" was established to coordinate activities at the Government level and self-governance bodies for implementing activities in the area of health protection and health promotion. The president declared 2004 the "Year of Youth and Healthy Living." Various activities were conducted throughout the Republic at different levels, all aimed at health promotion, development of socially significant youth initiatives, and broad participation in physical activities and sports. The republic also spends significantly on prevention and health promotion programs. According to its Ministry of Health, 93 million rubles (about US$3.3 million) were spent on these activities in 2002 and 102 million rubles (about US$4 million) in 2003. The Healthy Living Concept focuses on: Prevention of infectious diseases. This work seeks to reach full coverage (95 percent or more) with vacci- nation programs and increase personal responsibility for health. Significant improvement of environmental conditions. Chuvash is currently implementing such pro- grams as "Provision of Chuvash Population with Drinking Water: 1999­2010" and "Environmental Safety of Chuvash Republic for 2001­2003 and for the period till 2005," and "Healthy Nutrition." Health promotion. The program "Development of Physical Exercises, Sports and Health Promotion for Chuvash Republic Population for 2003­2006" is being actively implemented throughout the republic. Different regional entities are also involved in the project "Healthy Cities, Rayons, Settlements" under which they perform health promotion activities at the local level reaching people regionwide. Improvement of social and economic conditions. Chuvash has adopted a comprehensive Program of Economic and Social Development for 2003­2010 under which the authorities are strengthening education, increasing employment, and improving housing and health. Occupational health. Chuvash adopted a law on "Labor Protection in Chuvash Republic" as well as pro- grams on improving working conditions and "Health of Working Population: 2004­2010." Medical check- ups are being conducted regularly at regional enterprises. Shifting primary care focus to prevention. Chuvash is a leader in promoting general practice. In 2003 the number of general practitioner offices reached 200 with more than 100 being opened in rural areas. Health schools, such as an "Asthma School," "Hypertension School," and "Diabetes School" are being organized. This program is successful not only through a very strong commitment at the highest level of Government but also through its years-long multisectorial approach. Establishment of intersectorial teams to deal with various tasks enables them to work with different agencies and entities involved at all levels, from municipalities and settlements and to the regional level. Sources: Various documents from the Chuvash Ministry of Health. 79 Steps to Advance NCD Prevention ment; Ministry of Transport, police, and international and Control Efforts agencies; well-known public figures; and the public. Alcohol, tobacco, pharmaceutical, and food industries As part of its ongoing dialogue with the Russian are stakeholders, but partnerships are often impossible Government, the World Bank collaborated with the due to conflicts of interest. MOHSD in 2002-03 by supporting a series of senior policy seminars on public health and disease control. Seminar participants generally agreed that the following These seminars provided opportunities to share scientif- steps must be taken to move Russia forward in the pre- ic information and health system reform experiences vention and control of NCDs and injuries: from countries in Eastern Europe and the Organization for Economic Co-operation and Development and to · Upgrade the training of human resources; put forward options for official consideration by the · Increase financial and technical resources; MOHSD and other agencies. Seminar participants · Develop comprehensive legislative and policy frame- included representatives of the Russian Government, relevant public institutions, and Russian scientists, as works; well as external resource persons and partner agencies, · Upgrade the SanEpid's technical capability to carry such as the WHO, the U.K. Department for out epidemiological surveillance, and improve the International Development, the U.S. Agency for quality of the data for use in monitoring, reporting, International Development, the Canadian International and communicating trends in risk factors; Development Agency, and the Soros Foundation. · Establish adequate organizational and institutional Participants generally agreed that the Russian health structures in health protection, NCD prevention, care system and MOHSD are not well equipped to health education, and health promotion, with clearly tackle NCDs effectively. They also agreed on a number defined and coordinated roles, responsibilities, and of activities that must be implemented to move Russia functions. forward in the prevention and control of NCDs and injuries. They identified a wide range of potential stake- holders and partners in NCD prevention and control: Finally, seminar participants noted that many actions that could improve Russians' quality of life and extend Potential stakeholders in the health sector include life expectancy require intersectoral policies and inter- MOHSD, Health Committee of the Duma, regional ventions beyond the scope of the health services system. and local health authorities, SanEpid, centers for disease They also recognized that Russia's current health system prevention, primary health care facilities and providers, is neither designed for promoting health at the popula- the federal Health Insurance Fund, professional health tion level nor well positioned for delivering cost-effec- organizations (i.e., the All Russia Scientific Society of tive interventions to control NCD risk factors. Cardiologists, the Russian Academy of Medical Sciences, and the Hypertension Society); and research Policy Implications institutes. Strengthening Russia's existing public health infrastruc- Potential stakeholders outside the health sector include ture and redefining its agencies' roles is needed to government and legislative structures, such as the advance NCD control. It is important to involve President's Office, the Duma, oblast legislatures, and national research institutes, including those responsible political parties; the Ministry of Education, educational for various medical specialties, e.g., cardiology and trau- organizations, school teachers, and various faculty; the matology. General practitioners and other medical pro- Ministry of Culture and Sport and physical training and fessional groups of primary contact (district sports organizations; the mass media; public organiza- [uchastkovii] physicians and nurses) also need to be tions and associations; the Ministries of Economy, involved, as they see patients before onset of a disease or Finance, Trade; state and private industry; religious in its early stages. Research institutes and centers of pre- organizations; employers and worker associations; min- ventive medicine need to lead in formulating health istries and organizations for agriculture and environ- policy and implementing preventive programs, but first 80 they must improve their ability to independently and scaled up and replicated in other regions to achieve effectively carry out the task of NCD control. long-term health improvements. In particular, the devel- Involvement of civil society, including citizen-based opment of health surveillance through population sur- organizations--such as patient associations and diabetes veys, the planning and implementation of pilot inter- associations--and research groups outside the tradition- ventions, and several education activities for health per- al health field would boost opportunities for advocacy sonnel and the public are critical building blocks for and promotion of intersectional approaches. successful NCD prevention and control programs. CINDI's educational activities have played a key role in Although the projects carried out under the CINDI increasing the knowledge of health personnel. Its Program in Russia, such as in Pitkaranta, have been emphasis on prevention--and its promotion by political small, the results show success and suggest that they authorities--stimulated resource allocation and the could serve as powerful stimulants for the development shifting of health services away from a purely medical of national policies and programs. They should be care approach. 81 9 Chapter 9. What Additional Actions Can Russia Take? A s noted in the previous chapter, several efforts to A nationwide mortality reduction program should reform the Russian health care system are under- include three focus areas: (a) federal-level policies and way. Health care organization is being restruc- strategies, (b) priority sub-programs in regions, and (c) tured to achieve a better balance between hospital and measures to improve road safety and emergency services: ambulatory care and between prevention and treatment. Revised standards and norms are being used to assure A. Federal Level Policies and Strategies the quality of care. Meanwhile, demographic and health National subprograms must be defined in light of the trends signal the absolute and relative importance of magnitude of each health threat, the political commit- adult health problems caused by noncommunicable dis- ment required to solve or manage it, and feasibility and eases (NCDs) and injuries. New paradigms and inten- cost-effectiveness. The support mechanisms required for sive efforts are recognized as needed to meet new, com- success-legislative frameworks, institutional capacity, and plex, and multisectoral challenges. Experience in G-8 federal oversight also need to be considered. countries shows that in the medium and long term, reducing the social costs associated with NCDs and Select priority subprograms injuries and improving the population's overall health will require comprehensive and multisectoral health pro- Subprograms can target the population at large ("pri- motion and disease prevention efforts. mary" or "population-based" efforts) or individuals who need clinical intervention ("secondary" or "individual" The Russian Government is cognizant of the experience efforts). These latter individuals need medical attention, in other countries and is willing to invest in a federally treatment, and follow up to prevent their existing condi- mandated program under the MOHSD to reduce pre- tion(s) from worsening. Population-based subprograms mature mortality and avoidable disability, especially in should be developed to address alcohol, tobacco, road working-age adults. It has indicated its commitment to safety, and diet/physical activity, while Individual sub- launching a coherent and comprehensive program to programs would help people with hypertension, elevated combat NCDs and injuries. levels of cholesterol, and/or diabetes. Develop legal, policy, and strategy support for The Rationale for a priority subprograms Comprehensive Program Success in the alcohol, tobacco, road accident subpro- A well-defined and structured national program of pop- grams requires a strong framework of laws, regulations, ulation-based and clinical interventions to confront and enforcement mechanisms. The legal framework NCDs and injuries in Russia would help improve social would set out the national strategy and selected subpro- welfare and contribute to sustainable economic growth grams. Passing laws, issuing regulations, and developing by (a) bettering the health of the economically active strategies require strong political and technical leader- population; (b) reducing labor supply and productivity ship and broad support for the national program from losses from preventable deaths, illnesses, and disabilities myriad stakeholders. Good communication is essential, due to NCDs and injuries; and (c) minimizing regional including stakeholder consultations, public relations, disparities by reducing social risks for NCDs and working with the mass media, educating the public, injuries in the most vulnerable regions. and developing strategies to convey behavior change messages. 83 Build institutional capacity to implement at the inpatient level, in ambulatory settings, in work priority subprograms places, and at homes. The subprograms should support the development and national implementation of NCD Building institutional capacity entails (a) strengthening prevention and control, incorporating Russian expertise national technical leadership in developing a national and WHO guidelines. Both population and clinical strategy, action plan, and health goals; creating a popu- interventions should be covered at the federal, regional, lar and political consensus on strategy and goals; and and municipal levels. establishing partnerships with international stakehold- ers; (b) developing mechanisms for creating multisec- toral links with other ministries, NGOs, and the private The subprograms should build on the CINDI Program sector; (c) implementing professional training to remedy experience and promote healthy behaviors by awarding the shortage of trained public health professionals and "health promotion initiative grants" to stimulate innova- researchers; (d) creating a surveillance system integrated tive health promotion and disease prevention initiatives with a federal databank to generate valid, reliable, and at the regional and municipal levels and to build capaci- timely data; (e) overseeing empirical research; and (f ) ty at these levels for implementing integrated approach- partnering with the private sector and civil society to es. Technical assistance should be provided to assess the leverage resources and increase program effectiveness. capacity and readiness of various partners and organiza- tions to undertake health promotion and prevention Assure federal oversight of and accountability activities, to establish efficient organizational and mana- for priority subprograms gerial structures for health promotion, and to conduct process and outcome evaluation studies to measure fed- The MOHSD should be the coordinating ministry, and eral program success. a high-level working group in the Presidential Administration should ensure that the actions of partici- Population-based interventions pating ministries and the private sector are executed timely. This group's members could include MOHSD, Control of Alcohol Consumption Ministries of Transport, Justice, Agriculture, Education, Priority actions for controlling excessive alcohol con- Industry, Science and Technology, Economic sumption should target both supply and demand. Development and Trade, and Finance. To ensure the program's technical quality and relevance to the Russian · Supply: Alcoholism can be prevented by limiting context, the group should be supported by technical the availability of alcohol through the regulation of working groups responsible for coordinating and/or exe- production, quality assurance, distribution, prices, cuting specific subprograms. access (particularly for minors), and advertising. Supply can be regulated through higher taxes and B. Priority Subprograms in Regions prices and by reducing the number of sales outlets, This area should support implementation of national limiting their operating hours, and prohibiting sales priority programs in the regions, executing the national- to minors. These policies must be coupled with ly defined priority subprograms, but allowing for restricting advertising and broadcasting messages regional differences in health status and choosing compatible with healthy lifestyles. Particular atten- region-specific, appropriate, effective interventions. tion must be paid to prevention of (a) the sale of Activities should include primary prevention (before surrogate alcohol, which is highly toxic, and (b) problems arise) through population-based interventions home-produced alcohol. Stronger laws against drunk targeting alcohol, tobacco, and diet/physical activity and driving are also needed. secondary prevention (to control disease in its early stages and prevent progression) through clinical inter- · Demand: It could be modified through education ventions targeting individuals with high blood pressure, and information to reduce the onset of excessive elevated cholesterol, and diabetes. drinking, particularly among youth, and excessive drinking among heavy drinkers; promoting moder- Activities would seek to control multiple risk factors and ate alcohol consumption; and increasing awareness include prevention, diagnosis, treatment, and follow-up of the adverse consequences of alcohol abuse. Efforts 84 should be tailored to the diverse circumstances of the · Large, strongly worded health warnings should be general population, adolescents (especially students), required on all cigarette packs sold in Russia. pregnant women, drivers, family members of alco- Advertisers should be required (with strong enforce- holics, and worksite supervisors. Health professionals ment and large fines for noncompliance) to display and volunteers can provide key support in achieving large warning labels covering about 20 percent of all these educational goals through training and adviso- print advertisements and to pay for counter-advertis- ry assistance to teachers and community leaders. The ing in a required proportion to all radio and TV effectiveness of Russian narcology services and inno- advertising. These recommendations are low cost for vative methods, such as motivational interviewing, the MOHSD, but they are made with awareness of should be reviewed. previous attempts to implement tobacco advertising bans and restrictions that were successfully resisted Tobacco Control by the media and cigarette companies. · The highest priorities should be for the Government to sign the WHO-sponsored Framework · Legislation banning all tobacco advertising, promo- Convention on Tobacco Control and to enforce tion, and sponsorship and banning tobacco product existing policies for smoke-free worksites and public sales to minors should be considered. places, and, if allowed by the existing regulatory It should be clear, however, that alcohol abuse and environment, to extend their scope to all workplaces tobacco control measures will depend on the capacity of and enclosed public places. This may require aggres- law enforcement agencies to implement and enforce sive education and information efforts so that people related laws and regulations. understand the risks of second-hand smoke. Considerable international experience shows how to Changes in Diet and Promotion of implement and enforce such policies. Physical Activity · Second-tier priorities should be to identify successful Five of the ten leading risk factors contributing to high tobacco control and smoking cessation efforts and levels of mortality are influenced by diet and sedentary expand and build on them. It is important to evalu- lifestyle: high blood pressure, elevated cholesterol levels, ate program impact to learn which programs are and high body mass index. Priority actions for this pop- most cost-effective and worthy of expansion. ulation-based intervention should include: · Substantial tax increases are needed and should be · Public health policies promoting dietary guidelines designed to steadily raise cigarette prices and reduce for healthier eating; their affordability. Simultaneous efforts must reduce smuggling and informal market cigarette sales. Fines · Celebrities and well-known athletes serving as cham- that are consistent with the very large profits being pions for an active lifestyle, low body mass index, made should be used when people are convicted of and healthy diets; selling cigarettes on which taxes have not been paid. · Corporate and social responsibility of the food man- This measure may deter some violators, would gen- ufacturing industry in manufacturing and marketing erate revenue for enforcement, and would cut into healthier food; informal market profits, perhaps causing increases in · School programs on the importance of diet and the prices of these cigarettes. physical activity and their contributory causal rela- · Russian physicians who still smoke set a bad exam- tionships to cardiovascular disease, cancer, and dia- ple and should be encouraged to stop. All physicians betes; should be trained to ask patients whether they · Public health policies promoting regular physical smoke, provide advice and written materials on quit- activities; and ting, and increase availability of cessation support · An outdoor environment that invites physical activi- services. ty: bicycle paths, sidewalks, and crime-free parks. 85 Secondary prevention interventions multiplier effect on basic actions carried out among patients and their family members. Clinical interventions occur at the patient level in a health care setting. The physician is responsible for (a) Regional support subprograms for implementing early detection and diagnosis through laboratory testing the national mortality reduction program: of hypertension, high total and/or LDL cholesterol, and Analogous to the institutional capacity-building subpro- early signs of diabetes; (b) prescribing correct treatment; grams defined at the federal level in the first focus area (c) following up by checking patient compliance with for implementing priority subprograms, the regions and treatment, periodic retesting, and adjusting treatment municipalities should (a) conduct professional training regimen; and (d) informing patients of health risks and programs and empirical research studies; (b) develop where to find more information and help. These efforts mechanisms to link the various stakeholders and form should build upon ongoing reforms of the primary care partnerships; (c) implement a surveillance system to system at the regional level, such as in the Voronezh monitor the epidemiological evolution of the incidence Oblast and the Chuvash Republic supported under the and prevalence of NCDs and injuries targeted in the Health Reform Implementation Program. priority programs; and (d) develop regional databases Hypertension: Individuals with uncontrolled hyperten- compatible with the structure and content of the federal one and capable of data sharing and reporting progress sion have a three to four times greater risk of developing towards achieving the national objectives. coronary disease and a seven-fold greater risk of having a coronary event than those with normal blood pressure. The federal level should provide regions and municipali- Early detection, timely treatment, and monitoring of ties with technical assistance to establish, implement, treatment compliance would reduce cardiovascular mor- and/or strengthen these support systems. Regions and tality but must be balanced against the financial burden municipalities with strong support infrastructures on people needing long-term care (e.g., in many parts of should provide technical assistance to regions with Russia drugs must be paid for out-of-pocket). weaker ones. Also, regions and municipalities could Cholesterol: High cholesterol levels are related to diet work in clusters to share experiences and collaborate. and have a significant impact on cardiovascular mortali- ty. Strategies for reducing mortality from CVDs include C. Improved Road Safety and (a) screening people with multiple risk factors at the Emergency Services local level and at worksites; (b) expanding accessibility, As these efforts involve many more stakeholders than coverage, quality, and timeliness of care, including fol- does the control of other risk factors, they warrant a low-up programs; and (c) health education and commu- separate area of focus. nity activities to control risk factors and increase treat- ment compliance. Federal Level Responsibilities Experience from other countries shows that improving Diabetes: Primary prevention of diabetes is feasible in road safety requires a consistent, 20- to 30-year effort to part through physical activity, appropriate diet, and con- develop and implement comprehensive, integrated safe- trol of obesity. The importance of this pathology rests ty programs, including enhancement of road safety, on its complications (e.g., blindness, renal insufficiency, auditing processes, and funding of interventions in pri- and cardiovascular and neurological complications), and ority high-risk road corridors. Business planning its control is critical in the prevention of such complica- processes and performance monitoring systems must be tions. A secondary prevention program is therefore nec- created to support such a long-term strategy, and the essary at the primary care level. Primary care strategies first phase of such strategy should lead to consensus must include early detection and diagnosis in high-risk among stakeholders and public sector agencies on a groups (e.g., obese and pregnant women) and effective long-term action plan based on demonstrably successful follow-up and monitoring of all cases diagnosed. interventions. A financing plan for implementing the Orientation for nurses, teachers, and community health strategy should also be specified. Support should be pro- workers on proper management of diabetes can have a vided to develop a program of road-user education, traf- 86 fic safety enforcement, and emergency preparedness framework with indicators and national and regional along a first set of demonstration corridors. targets; and (h) provide technical assistance to regions. Motorization in Russia has increased rapidly, and the Regional Responsibilities public sector response has emphasized basic traffic man- Each region should develop its own road safety strategy agement functions, such as driver licensing and vehicle and action program following the federal program. inspection. International experience suggests that such Regions should be responsible for implementing nation- programs, although important, have only a marginal al laws, standards, and guidelines; implementing emer- impact on accident and fatality rates. To curb rapidly gency medical services; and developing an inter-institu- rising accident and fatality rates, targeted, cost-effective tional emergency medical network, including the ambu- measures that have an immediate impact, such as speed- lance network and emergency communication system. limit and drunk-driving enforcement, separation of traf- fic, and motorcycle helmet law enforcement, must be Regional responsibilities for training, capacity building, identified, implemented, and monitored under a com- developing partnerships and collaborative arrangements prehensive program. among all sectors, monitoring implementation of action plans, evaluating the outcomes of the road safety strate- Public awareness of road safety must change. For exam- gy, and developing a regional database compatible with ple, alcohol is a big traffic safety problem, but there is the federal one are similar to those described in the pre- virtually no public or political support for countermea- vious focus area on alcohol, tobacco, and diet/physical sures or a sufficient legal basis to deter drunk drivers. activity. Also analogous is that regions implementing Furthermore, laws requiring seat belts and motorcycle road safety strategies and programs should receive tech- helmets and imposing speed limits are ineffective or nical assistance from the federal level to establish or seen as conflicting with personal freedom. Finally, in the strengthen the support systems. Regions that already face of poor traffic behavior, traffic police lack knowl- have a strong support infrastructure should provide edge of--and incentives for developing--more effective technical assistance to weaker ones, and, again, regions policing strategies, have limited resources, and see their could work in clusters to share experiences and collabo- responsibility as apprehending offenders rather than rate. deterring unsafe behavior. Cost-effective interventions for preventing road traffic Specific federal responsibilities would be to (a) ensure injuries should include (World Bank 2004e): that an enforceable legislative framework is in place by reviewing and revising existing laws and reviewing and · Road environment: safe design of new infrastruc- adjusting regulations that implement them; (b) formu- ture; retro-fitting current infrastructure with of low- late a national road safety strategy and subprogram, cost safety design features (medians; roundabouts; identifying risk factors as a basis for planning and separation for motorcyclists, cyclists, and pedestri- improving effective prevention of injuries through a ans); systematic maintenance of all safety features, combination of education, regulation, enforcement, systematic reviews to identify and remediate road engineering, and technology; (c) establish mechanisms hazards, systematic safety audits of road designs to to forge links with public sector entities, industry, and ensure compliance with safety standards and regula- NGOs; (d) identify dangerous road corridors for early tions; action; (e) set standards and disseminate clinical proto- · Speed management: general deterrence-based cols specifying procedures on clinical management of police enforcement and education to ensure compli- patients during emergency medical services on the road, ance with speed limits; while in transport, and in trauma centers and hospitals to reduce pre-admission death rates and disability from · Safety belts and helmets: general deterrence-based accidents; (f ) create a national road accident database police enforcement and education to ensure compli- and establish/upgrade a management information sys- ance with child restraints, child auto safety, and hel- tem on road traffic accidents; (g) establish, implement, met standards and rules; and maintain a robust monitoring and evaluation 87 · Drunk driving: general deterrence-based police · Commercial vehicles: general deterrence-based enforcement and education to ensure compliance police enforcement and education to ensure compli- with legal alcohol limits; ance with safe loads, driving hours, and vehicle stan- · Novice drivers: graduated driver-licensing system dards; to control and reduce the risk exposure of young · Child safety: traffic safety education in the core drivers; school curriculum; and · Vehicle safety: harmonization with international · Emergency medical services: pre- and post-hospi- best practices and systematic inspection and certifi- tal care and victim recovery targeting high-risk corri- cation to ensure industry compliance with standards dors, including communication and ambulance net- and rules; works, well-trained teams, and improved emergency care wards in hospitals. 88 10 Chapter 10. What Health Improvements Could Result from an Intensified Program of Action? A s discussed in Chapter 7, Figure 10.1 Decline in Age-Adjusted Mortality from CHD in N. Karelia: Finland's North Karelia Project 1969-71 to 1993-95 provides strong evidence of what Russia could expect from a carefully designed, rigorously implemented pro- 35-44 gram to reduce NCDs and injuries. This project achieved major changes in mortality from NCDs through dietary changes; increased physical activity; 45-54 and reduced smoking, serum choles- terol, and blood pressure. By targeting e Ag important high-risk factors for NCDs, mortality by all causes in North Karelia 55-64 declined by about 45 percent during 1970-95. In the 1980s, these favorable Women changes began to develop all over Men Finland, improving life expectancy by 65-74 7 years for men and 6 for women. As shown in Figure 10.1, the 35-44-year- olds reaped the largest decline in age- 0 20 40 60 80 100 specific mortality: men in this age Percentage Decline in age adjusted mortality rate of CHD in N Karelia group had an 87 percent decline in from 1969-71 to 1993-95 mortality from CHD between 1971 Source: Adapted from Puska et al. (1995). and 1995. Men 35-64 had their age- adjusted mortality rate decline from about 700 per 100,000 population in 1971 to about 110 per 100,000 in 2001. expectancy in Russia. While the North Karelia experi- ence cannot automatically be extrapolated to the Russia More importantly for Russia, data from North Karelia experience, it does signal the possibilities. The data in reveal that results from prevention efforts may appear in Tables 10.1 (males) and 10.2 (females) on page 92 must years rather than decades--improvements occur some 2- be interpreted with caution and be mindful of the 7 years after the elimination of the exposure to a risk assumptions underlying this exercise. Further analysis factor, and that they are beneficial even for people in could add context to this assessment by availing of older age groups. 14 Age-specific and age-and-cause-specific death rates and cause of Mortality Reduction death ratios were computed, and an ordinary life table using age-specific death rates for all causes combined was developed. The Multiple Decrement Life Table14 approach was used Next, death rates were distributed by cause, indicating the total number of deaths in each age group. Probabilities of death by to estimate the effect reductions in cardiovascular, diges- cause were then computed by cause elimination, after which tive, and external causes of diseases could have on life the associated decrement tables were calculated. 91 Table 10.1 Estimated Improvements in Life Expectancy if Mortality Rates of mortality (road accidents, intentional Are Reduced by 20 and 40 Percent, Russian Males, 2000 self-harm, and assaults) signal that this as another high impact avenue for raising life Current Circulatory Digestive External causes expectancy in Russia (Table 10.1). Life expectancy 20 40 20 40 20 40 Birth 57.77 62.5 63.9 58.3 59.8 61.3 62.5 Although women in Russia already live 20-24 40.08 44.0 46.2 40.6 41.9 45.4 46.2 longer than men, significant gains are pos- 25-29 35.74 41.7 44.2 36.3 37.5 40.4 42.0 sible for women, too, although the magni- 30-34 31.82 37.8 40.6 32.3 33.4 35.7 36.1 tude of the improvement is not as large (Table 10.2). 35-39 27.78 33.8 34.6 28.3 29.6 30.9 32.9 40-44 24.19 30.0 30.8 24.6 25.1 26.6 27.9 As shown in Figure 10.2, WHO estimates 45-49 20.40 26.1 28.3 20.8 20.9 22.2 25.3 show that controlling the major risk fac- 50-54 17.33 22.5 24.8 17.6 17.7 18.6 19.2 tors could result in major improvements in 55-59 13.92 18.7 21.5 14.2 14.2 14.7 15.4 healthy life expectancy (HALE), an indica- 60-64 11.48 15.1 17.0 11.6 11.6 11.9 14.1 tor that captures both mortality and mor- bidity. The region that includes Russia, Source: Bakilana 2005. EUR-C,15 could gain more than 10 years of HALE by controlling the 20 leading Table 10.2 Estimated Improvements in Life Expectancy if Mortality Rates Are Reduced by 20 and 40 Percent, Russian Females, 2000 risk factors. This is a very important find- ing because increasing life expectancy but Current Circulatory Digestive External causes not HALE would spike the health system's Life expectancy 20 40 20 40 20 40 burden. Birth 72.3 73.9 75.2 73.2 74.1 72.8 73.1 20-24 54.0 55.6 56.9 54.9 55.8 54.5 54.8 The Impact on Achieving the 25-29 49.3 50.9 52.2 50.2 51.1 49.8 50.1 Millennium Development 30-34 46.6 46.2 47.5 45.5 46.4 45.1 45.4 Goals (MDGs) 35-39 40.0 43.6 42.9 40.9 41.8 40.5 40.8 40-44 NCDs and injuries have not been consid- 35.4 37.0 38.3 36.3 37.2 35.9 36.2 ered sufficiently important to be included 45-49 30.9 32.5 33.8 31.8 32.7 31.4 31.7 among the health-related targets of the 50-54 26.7 28.3 29.6 27.6 28.5 27.2 27.5 MDGs, leading the relevance of these goals 55-59 22.6 24.2 25.5 23.5 24.4 23.1 23.4 for countries in eastern Europe and Russia, 60-64 18.8 20.4 21.7 19.7 20.6 19.3 19.6 where NCDs predominate, to be called into Source: Bakilana 2005. question (Marquez and Suhrcke 2005). Russian morbidity and mortality data and data from epi- 15 demiologic transitions models. EUR-A (countries with very low child and adult mortality): Andorra, Austria, Belgium, Croatia, Cyprus, the Czech Republic, Denmark, Finland, France, Germany, Greece, If mortality from preventable or treatable components Iceland, Ireland, Israel, Italy, Luxembourg, Malta, Monaco, the of circulatory and digestive diseases and external causes Netherlands, Norway, Portugal, San Marino, Slovenia, Spain, can be reduced in working-age adults in Russia, impor- Sweden, Switzerland, and the United Kingdom; EUR-B (coun- tries with low child and adult mortality): Albania, Armenia, tant improvements would result in life expectancy for Azerbaijan, Bosnia and Herzegovina, Bulgaria, Georgia, both men and women. For example, life expectancy at Kyrgyzstan, Poland, Romania, Serbia and Montenegro, birth for men could increase as much as 5 years by Slovakia, Tajikistan, the former Yugoslav Republic of reducing CVD by 20 percent. The impact of reducing Macedonia, Turkey, Turkmenistan and Uzbekistan; and EUR-C (countries with low child and high adult mortality): Belarus, the burden of these diseases among men is more evident Estonia, Hungary, Kazakhstan, Latvia, Lithuania, the Republic in adulthood. The estimates of reducing external causes of Moldova, the Russian Federation, and Ukraine. 92 A recent World Bank study (Rechel, Figure 10.2 Estimated Gains in Healthy Life Expectancy with Elimination of Shapo, and McKee 2004) examined the 20 Leading Risk Factors by Subregion the potential impact of achieving the 18 MDGs in Russia and concluded that achieving the goals for child mortality 16 AFR-E and maternal mortality (67 percent 14 and 75 percent reductions, respectively, from the 1990 values) would con- 12 EUR-C: Including AFR-D tribute very little to improving life Russia HALE 10 expectancy at birth (0.96 years). By in EMR-D EUR-B contrast, reduction of adult mortality 8 AMR-D AMR-B AMR-A due to CVDs and external causes of Gain 6 EUR-A death (injuries, violence and poisoning) SER-B 4 to the level found in the European WPR-A Union, would yield an increase in life 2 expectancy at birth of more than 10 0 years in the Russian Federation. 35 40 45 50 55 60 65 70 75 80 Current HALE years It is clear, therefore, that health Source: Adapted from WHO 2002b. improvements in Russia will require a substantial effort in NCDs prevention and control in the upcoming years. However, progress towards restoring a balance in the global health agenda is likely to be furthered by capturing and translating the impacts of NCDs into economic valuations as it is done in Chapter 11 of this study. 93 11 Chapter 11. What Are the Potential Benefits of Mortality Reduction in Russia? I f the excessive burden of adult ill Figure 11.1 Three Scenarios for Russian Adult Mortality Rates due to health and premature death in Noncommunicable Diseases and Injuries, Age 15-64, 2002-25 Russia due to NCDs and injuries Deaths per 100,000 population were reduced, what economic bene- fits could result? The overarching 1000 Scenario 1 message in this chapter is unambigu- 900 ous: if effective action were taken in 800 Russia, improved health would play an important role in sustaining high 700 economic growth rates at the macro- 600 Scenario 2 level. This is despite the fact that the 500 assessment concentrates only on the 400 effect of mortality reductions, setting Scenario 3 aside the impact of the likely associat- 300 ed morbidity reduction. 200 Benchmark EU-15 (2001) 100 Benefits of Reducing 0 NCDs and Injury 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 Mortality: A Simple Source: Suhrcke at al. 2005. Static Calculation Note: Scenarios are based on the assumptions outlined in the text. The potential impact of policies that promote reduction in preventable adult mortality on the Russian economy was assessed for NCD- and injury-mortality could be expected to this study by comparing three scenarios (Suhrcke at al. decline simply as a response to the rapid economic 2005). Adult mortality itself was defined as mortality in growth observed in the last five years (a trend that may the age group 15-64, between 2002 (the year for which well continue into the future), even without specific the latest Russian data from the WHO Mortality additional efforts undertaken to improve adult health. Database are available) and 2025. Since this report focus The future trends cannot be predicted, but what is on the impact of NCDs and injuries, only changes in known, as shown in the previous chapters, is that (a) the adult mortality driven by the evolution of adult NCD- long-term, increasing trend of non-communicable (in and injury-mortality rates were assessed, understating particular CVDs) disease and injury-related mortality the greater health impact that eventual broad-based over the past decades leaves very limited hope for a health interventions would likely have. sudden or even gradual reversal of these trends, and (b) these cause-specific mortality rates have increased These scenarios, as illustrated in Figure 11.1, are: significantly even in recent years when economic growth was particularly strong (see Figure 1.11 in chapter 1). Scenario 1: Status quo. In this scenario, the current For these reasons, a benchmark scenario in which the (2002) level of adult mortality from NCDs and injuries relevant cause-specific mortality rates would remain at is assumed to remain constant until 2025. This might their 2002 levels is considered modestly optimistic. seem to be an unnecessarily pessimistic scenario as 95 Table 11.1 Economic Benefit Estimation for Scenario 2 NCD Injuries NCD + Injuries CVD A) Zero growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 15.217 9.637 24.854 15.474 Share of benefits in 2002 GDP 1.3% 0.8% 2.1% 1.3% B) 3% p.a. growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 22.974 14.490 37.464 23.146 Share of benefits in 2002 GDP 1.9% 1.2% 3.2% 2.0% C) 5% p.a. growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 30.454 19.158 49.612 30.506 Share of benefits in GDP 2.6% 1.6% 4.2% 2.6% Source: Suhrcke at al. 2005b. Note: Future benefits are discounted to the present at a 3% rate p.a. Scenario 2: Intermediate scenario. This scenario The static economic benefit of gradually bringing the assumes that policies are adopted that achieve half the adult NCD- and injury-mortality rates down to current improvement seen in the most optimistic scenario, EU-15 rates by the year 2025 (scenario 3) is estimated number 3. This second scenario corresponds to an to be between 3.6 percent and 4.8 percent of the 2002 annual reduction of 2.3 percent for NCDs and 3.3 per- Russian GDP. The analysis of scenarios 2 and 3 distin- cent for injuries. guishes between three sub-scenarios A, B, and C, each varying with its own assumed future growth path. The Scenario 3: Optimistic scenario. This scenario higher the future GDP, the more production would be assumes that policies are adopted that cause a decline in foregone due to a life year lost, and by implication the Russian mortality rates from NCDs and injuries to the benefits of reducing mortality would be greater. Under most recent available EU-15 level. This corresponds to this approach, each year saved (compared to the first an annual rate of reduction of 4.6 percent for NCDs scenario) was valued by the projected per capita GDP and 6.6 percent for injuries. for the year in which the "saving" occurs. To be able to compare the different future income streams, the 2002 None of these scenarios is based on the detailed model- present value of future values was calculated by applying ing of the impact of specific policy interventions, a sub- the commonly used discount rate of 3 percent. Table ject of further research. What is important here is that 11.1 and Table 11.2 report the benefits in both absolute the chosen scenarios can be considered as plausible, dollars and as a share in GDP for scenario 2 and 3, based for instance on the mortality reductions that other respectively. A column that examines the reduction in Western and Northern European countries have adult CVD mortality is also included in the tables.16 achieved in the past decades. While the optimistic sce- nario is no doubt very ambitious, it is within reach of It is highly probable that the actual economic gain from what those other countries (e.g., Finland) have achieved. reducing future mortality is larger than the static gains The main findings of the assessment prepared for this 16 study (Suhrcke et al. 2005b) are: As in the case of non-communicable diseases and injuries it is assumed that Russian CVD rates would reach the latest EU-15 rates by 2025. 96 Table 11.2 Economic Benefit Estimation for Scenario 3 NCD Injuries NCD + Injuries CVD A) Zero growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 28.654 13.975 42.629 17.631 Share of benefits in 2002 GDP 2.42% 1.18% 3.60% 1.49% B) 3% p.a. growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 34.126 16.544 50.671 20.753 Share of benefits in 2002 GDP 2.88% 1.40% 4.27% 1.75% C) 5% p.a. growth in GDP p.c. Present value of benefits 2002-2025 (in billion US$ PPP) 38.707 18.685 57.391 23.341 Share of benefits in GDP 3.3% 1.6% 4.8% 2.00% Source: Suhrcke at al. 2005b. Note: Future benefits are discounted to the present at a 3% rate p.a. calculated above. If dynamic effects exist, they are bound does not mean health has no value. When asked, people to be larger than any static effect, as even a marginal are ready to pay substantially for better and longer dynamic impact will outgrow any static gain over time. health, so there must be an implicit value that people There is substantial empirical evidence that health does attribute to health. While this value is high, it is not impact positively on economic growth and, hence, does infinite, since people are not willing to give up every- entail positive dynamic effects on the macroeconomy. thing in exchange for better health.18 One way to make the high value attributed to health Static Welfare Effects more explicit is by measuring the extent to which a per- The Theoretical Framework. Several prominent econ- son is willing to trade off health with specific market omists, as well as multilateral financial organisations goods for which a price exists. Willingness-to-pay (World Bank, IMF) have measured the economic cost (WTP) studies undertake this measurement. WTP can of mortality using a broader concept than GDP per be inferred from risk premiums in the job market: jobs capita. The new approach starts from the recognition that entail health risks, such as mining, pay more in the that GDP is an imperfect measure of social welfare: it form of a risk premium. A large number of WTP stud- fails to incorporate the value of health. The true purpose ies now make it possible to calculate a "value of a statis- of economic activity is the maximisation of social wel- tical life" (VSL), which can be used to value changes in fare, not necessarily of the production of goods by itself. mortality. Usher (1973) first introduced the value of Since health is an important component of properly mortality reductions into national income accounting. defined social welfare, measuring the economic cost of mortality only in terms of foregone GDP leaves out a potentially major part of its `true economic' impact, 17The health care inputs included in the measurement of GDP defined as its impact on social welfare. represent only a small share of the true value of health, as argued here. Health is not incorporated in the measurement of GDP 18Referred to here are situations where people face marginal trade- because it is a non-market good and consequently has offs between health and other goods, not the far less representa- no quoted market price.17 Yet having no market price tive situation where people face immediate death, the prospect of which would increase the readiness to pay. 97 Table 11.3 Welfare benefits of Scenario 3 and 2 NCD Injuries NCD + Injuries CVD SCENARIO 3 Present value of benefits per capita (in US$ PPP) 1,512 866 2,377 1,242 Share of benefits in 2002 GDP 18.4% 10.5% 28.9% 15.1% SCENARIO 2 Present value of benefits per capita (in US$ PPP) 919 565 1,484 876 Share of benefits in 2002 GDP 11.2% 6.9% 18.0% 10.6% Source: Suhrcke at al. 2005b. This was done by generating estimates of the growth in many times greater than that of other forms of "full income" (or "wealth"), a concept that captures investment. changes in life expectancy by including them in an assessment of economic welfare, for six countries The Results in Russia. It is straightforward to apply (Canada, Chile, France, Japan, Sri Lanka, and Taiwan) the approach to assess the welfare benefits of reducing during the middle decades of the 20th century. For the adult mortality in Russia. The critical input is a value of higher-income countries, about 30 percent of the a statistical life for Russia. The principle in developing growth of full income resulted from declines in mortali- such estimates is to ensure that the lower boundary of ty. Estimates of changes in full income are typically gen- plausible estimates cannot be challenged. Real values erated by adding the value of changes in annual mortali- will certainly be higher; however, the key issue is the ty rates (calculated using VSL figures) to changes in minimal plausible figure. For the purpose of the present annual GDP per capita. Even these full-income esti- calculations, a very conservative estimate of US$ mates are conservative, including only the value of 500,000 was used for the value of a statistical life in changes in mortality while excluding the total value of Russia as of 2002.19 changes in health status. 19 To assess how conservative this is, see Miller (2000). Miller For the United States, Nordhaus (2003) rediscovered assembled a collection of VSL studies and estimated an equa- Usher's pioneering work and found that the economic tion that would predict the VSL in terms of GNP per capita and some other factors. Applying the parameters to Russia, he value of increases in longevity in the last hundred years obtained a range of US$ 300,000 to US$ 800,000 with the roughly equals the value of measured growth in non- best estimate being US$ 370,000. However, these figures were health goods and services. Nordhaus tested the hypothe- based on 1997 GDP data and expressed in 1995 dollars. By sis that improvements in health status have made a 2003/04, Russian GDP had increased by 30 percent since 1997. An updated VSL would be US$ 500,000 in 1995 dol- major contribution to economic wealth (defined as full lars. Inflation in the United States between 1995 and 2004 has income) over the 20th century. A more detailed assess- an accumulated value of 18 percent, so that in 2004 dollars, ment reveals that "health income" probably contributed VSL in Russia would be US$ 590,000. Hence, the US$ to changes in full income somewhat more than non- 500,000 used here is certainly a lower bound. Based on a review of existing VSL studies, Crafts (2003) assumes that a health goods and services before 1950 and marginally conventional estimate of a country's VSL equals 132 times less than nonhealth goods and services afterwards. If the GDP per capita. For Russia this would give a 2002 VSL of 132 results of this and other related papers (e.g., Cutler and x US$ 8,230 = US $ 1,086,360, about double the estimate in this report. This calculation does, however, assume a unitary Richardson 1997; Miller 2000; Costa and Kahn 2004; income elasticity of VSL, a result that other authors reject in Crafts 2003; Viscusi and Aldy 2003) are confirmed, favor of an income elasticity below 1 (see, for example, Viscusi then the role of health should be reconsidered: the social and Aldy 2003), which would tend to reduce the VSL of coun- productivity of spending on health (via the health sys- tries with a lower GDP per capita. Yet even in this case our estimates remain the lower bound of the range of possible tem and other sectors that impact on health) may be estimates. 98 Table 11.3 summarises the results from the Figure 11.2 GDP per Capita Forecasts in the Three Scenarios welfare benefits estimation of scenarios 3 1996 constant U.S. dollars and 2, assuming a Russian VSL of US$500,000. In the calculations the same 17,000 discount rate (3 percent) for future benefits Scenario 3 (reach EU-15) was used. It was assumed that the VSL 16,000 Scenario 2 remains constant over the period 2002 to 15,000 Scenario 1 (status quo) 2025, which is in line with the literature, if 14,000 GDP per capita also remains constant. If GDP per capita grows over time, this will 13,000 increase the VSL in future years, thereby 12,000 even further increasing the welfare benefits to health. 11,000 10,000 Not surprisingly, the estimated welfare ben- 9,000 efits are a multiple of the more narrow returns in the previous GDP-based calcula- 8,000 2000 2005 2010 2015 2020 2025 tions. Specifically, the accumulated effects of reductions in mortality from non-com- Source: Suhrcke at al. 2005b. municable diseases and injuries reductions is approximately six times higher than when using the narrow concept (i.e., the share of parameter estimates of the effects of life expectancy on welfare benefits is about 28.9 percent of 2002 GDP economic growth have been remarkably comparable and under Scenario 3 as compared to the static economic robust across studies, notwithstanding the observation benefits of about 4.8 percent of 2002 GDP under that the empirical growth regression results are generally Scenario 3). not very robust, given the high degree of multicollinear- ity between many of the explanatory variables used Dynamic Effects: The Impact of Adult (Levine and Renelt 1992; Sala-I-Martin, Doppelhofer, and Miller 2004). In some studies, initial health status, Health on Economic Growth typically proxied by life expectancy or adult mortality, The Theoretical Framework. Recent worldwide proved to be a more significant and more important empirical evidence strongly suggests that health is a predictor of subsequent growth than the education indi- robust determinant of economic growth. Such growth cators employed (Barro 1997). Bhargava, Jamison, and is driven by effects on savings (Bloom, Canning, and Murray (2001), for instance, show in the context of a Graham 2003), on human capital investment (Kalemi- panel regression that the 5-year growth rate of GDP per Ozcan, Ryder, and Weil 2000), on labor market partici- capita depends on a country's adult mortality rate, pation (Thomas 2001), on foreign direct investment among other factors. They also show that the direction (Alsan, Bloom, and Canning 2004), and on productivi- of causality runs unambiguously from adult mortality to ty growth (Bloom, Canning, and Sevilla 2002). The growth. In this section , this empirical relationship is combined effects of health on economic growth are con- applied to Russia and then employs the empirical results firmed in theoretic and empirical work by Barro (1996); to project different future pathways in GDP per capita Bhargava, Jamison, and Murray (2001); Bloom, under the above three scenarios. In doing so, an Canning, and Seville (2001); Jamison, Lau, and Wang assumption is made that the empirical regularities that (2004), among others. Studies examining the impact of hold in a representative world sample of countries also health on income levels or income growth differ sub- hold for Russia (see appendix B section 8 for details). stantially in terms of country samples, time frames, con- trol variables, functional forms, data definitions and The Results in Russia. Applied to the specific configurations, and estimation techniques. Nevertheless, Russian context, the dynamic benefits of improving 99 adult health, i.e., the effect on economic growth rates, of the estimated loss in Russia is better appreciated are massive and growing over time. Even if the future when it is compared with the significant lower estimat- returns are discounted to the starting year value (2002), ed losses in the United Kingdom: US$1.6 billion in they represent a multiple of the static GDP effects. One 2005 and US$6.4 billion by 2015. As these losses accu- conservative estimate indicates that while in 2005 the mulate over time because each year more people die, the difference in the per capita GDP between the status estimated accumulated loss in Russia during the 2005- quo scenario and the most optimistic scenario is only 15 period amounts to US$303.2 billion as compared to US$105-324 (depending on the estimation methodolo- only US$32.8 billion in the United Kingdom. gy used), by 2025 this difference would have grown to US$2,856-9,243. Even if these future returns are dis- When these losses are translated into percentage reduc- counted to the starting year value, they represent a mul- tion in GDP, WHO estimates that in 2005 tiple of the static GDP effects. Figure 11.2 on page 99 1 percent of the Russian GDP was reduced, and by illustrates the predicted path of GDP per capita under 2015 the percentage reduction in GDP would be over 5 the three scenarios, using the very conservative lower percent of GDP, far higher than the estimated reduction bound of the growth estimates calculated. The area of 1 percent of GDP in other countries such as Brazil, between the scenario 1 and scenario 3 lines indicates China and the United Kingdom. In large measure the the likely economic benefit under the optimistic estimated large losses in Russia are due to higher rates of scenario. CVDs (the leading killer of the Russian working-age population as discussed in chapter 3) than in other countries. Additional Estimates of the Macroeconomic Consequences of NCDs and Injuries in Russia Conclusion In addition to the economic assessment prepared for Reducing NCDs and injury-related mortality rates this study and discussed in the previous sections of this among Russian working-age adults will have a major chapter, recent estimates undertaken by WHO (2005b) macroeconomic and poverty reduction impact, regard- also show a very dire picture. According to the WHO less of how this is measured. Based on the results of the estimates20, the Russian Federation will lose in 2005 assessment conducted for this study, as well as the recent US$11.1 billion of national income as a result of the WHO estimates described above, the expected econom- impact of deaths from heart disease, stroke and diabetes ic benefits are of a magnitude that easily outweighs the on labor supplies and savings. This figure is estimated costs of health promotion and disease prevention pro- to increase to US$66.4 billion by 2015. The magnitude grams. Given the significant positive effect on economic growth from investing in health (Barro 1997, Suhrcke at al. 2005a), governmental intervention is urgently need- 20 Three approaches were used: (i) systematic review on chronic ed in Russia to develop health-enhancing policies and disease costs of illness; (ii) elucidation of the human capital programs to address the alarmingly high mortality rates impact of chronic diseases through their impact on labor sup- among the working-age population. These efforts ply--the Solow growth model using the Cobb-Douglas func- should be seen as key investments to help improve the tion; and (iii) elucidation of the impact of chronic diseases on and growth in economic welfare--the full-income approach. general welfare of the population and secure sustainable (WHO 2005b; Annex 4 Economic Analysis Methods). economic growth in the future. 100 EPILOGUE Epilogue R ussia is experiencing a major and complex demo- control, including strengthening the country's disease graphic and health crisis characterized by prema- surveillance, research, and evaluation capacity; (d) an ture mortality; ill health and disability among upgraded workforce and intersectoral institutional young adults, particularly males; dramatic decreases in arrangements and structures to meet current needs; and life expectancy; and a reduction in population size, cou- (e) funding policies to support the scaling up of health pled with the aging of its population. The population at promotion and NCDs and injuries prevention. large, particularly in the less developed regions of the country is affected by risk factors associated with higher Much of the NCD and injury burden can be avoided disease and disability levels due to non-communicable with the adoption of good practices that have been conditions and injuries. These conditions require treat- proven at the international level. Similar to the signifi- ment and care on a continuing basis to control them cant and sustained achievements under the North and prevent, if possible, disability. The health care sys- Karelia Project in Finland, major health gains can be tem, individuals, families, and communities bear a achieved in Russia in a short period by modest lifestyle heavy burden due to the chronic nature of these condi- changes if they are practiced by major segments of the tions. Poor adult health negatively affects economic population. In addition, interventions during the clini- well-being at the individual and household level, and, if cal stages of the disease have more impact and are less effective action were taken in Russia, improved health expensive to the health care system if treated early. would play an important role in sustaining high eco- nomic growth rates at the macro-level. The time has come, therefore, to transform the concept of health care in Russia by acknowledging the limits of This report argues that most NCDs and injuries could medicalization and the breadth of social and economic be prevented and controlled through a comprehensive factors that affect health, including the responsibility of national program operating at the federal, regional, and individuals for their own health and government municipal levels. Program features would include (a) responsibility for creating an enabling environment for addressing health as a multisectoral concern; (b) its population to make healthier choices (Califano demand- and supply-side incentives to encourage 1994). healthy behaviors; (c) modern approaches to disease 103 APPENDIX A to taloT earsy tality declin- NCDs specific popula- 1,203,442 1,176,707 1,310,638 1,369,912 1,392,552 1,444,112 990,533 1,082,691 1,076,730 1,145,213 1,125,095 1,186,100 mor and beene ,y omfr ariousv omfr used h,t the ariantv hav is ASDRs Pregnanc ildbirhc and 503 502 497 485 454 427 for puerperium (ASDRs) "normal"a the complications medium ussiaR and in if ariantv rates the and population to Assault 30,127 29,597 29,896 29,173 27,962 26,733 10,206 10,306 10,241 10,068 9,714 9,398 ojection death numbers pr ojected assumption ojected pr tility appliedear pr age-specific fer population 47,520 46,440 47,478 46,408 45,152 43,977 9,099 9,354 9,233 9,168 8,978 8,932 population Intentional self-harm ussia'sR ASDRs the accurately 2000-2025, UN should to dium" 0 been ndereGyb "mea 200 ecauseb has hichw 45,935 45,016 46,188 45,158 43,859 42,397 12,855 12,887 13,114 12,912 12,648 12,279 deaths applied on Accidental poisoning of uses ussia'sR (it if simply population ussiaR afficrT number thenerew eases 28,738 28,081 28,138 27,178 26,024 24,933 10,182 10,266 10,010 9,737 9,480 9,285 depending chosen NCDs. accidents atabase. deaths D of incr the in ASDRs was ofe erwlo of potential then siz or risks The number External causes the and the 245,771 240,207 245,549 239,995 232,740 225,272 67,730 68,588 68,503 67,808 66,148 64,889 njuriesI opulationP option of if higher erwlo or UN atabase. be and D ariantv expected declines deaths the and of higher iseasesD of liver 17,278 16,468 18,280 18,660 19,087 19,206 11,860 12,066 12,508 12,797 12,970 12,926 the would w to data NCDsyb y demonstrationa talityro medium actually M sho number NCDs as The leading 71,679 68,571 78,565 83,138 86,283 90,554 29,825 32,083 31,880 33,571 33,208 34,765 WHO)( WHO usedaC Respirator NCDs the atabase). y udelycryrev NCDs omfr omfr expected omfr omfr D the deaths behavior of rganization 2025 deaths of culator eaths 543,219 536,960 621,337 672,820 692,365 742,762 685,609 766,578 759,765 821,739 799,531 855,495 O to of D Cir deaths aboute up patterns of ealth of s H der obtainederew opulationP number UN number Mental 5,712 5,651 5,939 5,911 5,800 5,758 2,655 2,808 2,840 2,935 2,815 2,849 orld NCDs umberN isord informativ W number the total changing on the NCDs omfr omfr expected the but the the for based deaths cases udecr that 3,469 3,256 3,698 3,850 3,945 4,108 8,010 8,282 8,269 8,535 9,026 9,321 is ojectedrP Diabetes mellitus estimate of account to ASDRs some , in used A clearyrev into and der used obtainederew number takes Endocrine tritional,un and 4,217 3,965 4,450 4,600 4,754 4,853 8,676 8,953 8,943 9,207 9,695 9,984 Calculations ussia'sR istI method metabolic disor was expected, el. lev (2005). same. obtained As The method populations Estimated the The ears.y 2000 Neo- 2025. methodology plasmas 163,994 152,494 181,118 193,021 204,582 213,559 133,826 140,018 140,927 146,251 150,429 155,552 the A1. akilanaB simple A mainer ojected to some at the ppendixA (pr bleaT ce: up for earY MEN 2000 2005 2010 2015 2020 2025 WOMEN 2000 2005 2010 2015 2020 2025 Sour ote:N causes 2025 assumption). tions ing mainer whether 105 APPENDIX B Appendix B. Data Sources, Methodologies, and Detailed Results for Chapters 6 and 11 Brief Description of Datasets Used household response rate in the beginning of the second for the Economic Analysis phase of the RLMS exceeded 80 percent, and individual questionnaires were obtained from about 97 percent of Russian Longitudinal Monitoring Survey the individuals listed in the household rosters. (RLMS) The Russian Longitudinal Monitoring Survey (RLMS) This dataset lacks a true panel design, as household are was conducted with the support and assistance of the not followed if they move from their dwelling, and like- World Bank, the United States Agency for International wise individuals who leave a household are not followed. Development (USAID), the National Science The effect of attrition is relatively modest and has been Foundation, the National Institute of Health, and the highest for the respondents from Moscow and St. North Carolina Population Center. Petersburg. RLMS covers the period from 1992 to 2003, but the The information is rich on income and expenditures of survey changed considerably throughout this period: in households, labor force participation, health conditions, a first phase (from 1992 to 1994), the main RLMS and individual risk factors. accomplishment was the creation of the first national sample frame allowing surveys to be representative at National Survey of Household Welfare the national level. More recently, this sample frame has and Program Participation (NOBUS) been extended to develop samples representative at the While the RLMS has the advantage of being repeated regional and oblast levels (RLMS 1998). annually, which allows some comparison over time, the NOBUS survey, so far only held once in 2003, covers a For the second phase, covering the period 1994-2003, far more comprehensive portion of the population. the emphasis changed from institution-building to pro- With a sample of about 44,500 households, it is repre- viding timely, high-quality information. The survey's sentative both nationally and for 46 larger subjects of main unit of observation is the household. RLMS cov- the Russian Federation. It captures differing aspects of ers primarily the European part of Russia, but the distri- household welfare and focuses on household access to bution of household size in the sample within urban social services. Its health measurement component, and rural areas corresponds well to the figures from the however, is small compared to the RLMS, so a direct 1989 census (for a detailed comparison of the 1989 cen- comparison to the RLMS results is not possible. sus and the RLMS, see RLMS [1998]). At each round, data are collected on the household, each household member, and the residential community. Technical Details and Specific Results on the Analysis of the Labor Households were selected on the basis of a multi-stage Supply and Productivity Effect process, with the households being clustered into pri- mary sampling units ("sites"). Although the target sam- of Health ple size was 4,000 households, the number of house- The methodologies that can be applied are largely deter- holds drawn into the second phase sample was 4,728 in mined by data availability and by the informed evalua- order to allow for a 15 percent nonresponse rate. The tion of the importance of the endogeneity problem, 107 Table B1. Calculation for Costs of Absenteeism Year Gender Annual Average Average GDP per Average Average Active Total Total average annual annual capita (in wage loss production population income loss production working wage wage current local for a loss for a (billion) (GDP) loss days (among all (among currency person person who (billion) missed jobholders) those units, in absent was absent due to absent at constant the average the average illness least once) 2000 prices) number number of days of days 2000 Male 10.8 26,268 24,576 777 1,480 36,639,000 28.48 54.24 Female 9.24 15,648 15,864 396 1,266 33,822,000 13.40 42.83 Total 10.08 20,724 19,992 50,028 572 1,382 70,461,000 40.33 97.35 2001 Male 9.48 32,501 33,994 844 1,373 36,788,000 31.05 50.52 Female 10.92 20,335 20,046 608 1,582 34,402,000 20.93 54.42 Total 10.2 26,145 26,062 52,876 731 1,478 71,190,000 52.01 105.19 2002 Male 8.64 37,448 37,929 886 1,318 36,937,000 32.74 48.70 Female 10.32 23,891 25,146 675 1,575 34,982,000 23.63 55.09 Total 9.48 30,309 30,763 55,699 787 1,447 71,919,000 56.62 104.04 2003 Male 9.6 40,514 36,851 1,066 1,583 37,087,000 39.52 58.72 Female 9.36 25,552 25,544 655 1,544 35,125,000 23.02 54.22 Total 9.48 32,503 30,570 60,195 844 1,563 72,212,000 60.96 112.90 Sources: RLMS RLMS RLMS IMF Note: The population average wage was used in the cost calculations since there were no systematic patterns when comparing population average wage and absentees' average wage. which tends to afflict many if not all efforts to establish use a self-rated health indicator, medically diagnosed a causal relationship in economic and social empirical diseases, or workdays missed due to illness. research. In the given context the endogeneity problem means that there could be a simultaneous relationship 1. Ordinary least squares (OLS) regressions. This between the chosen health proxy and labor market out- approach is based on a seminal paper by Bartel and comes that would bias the statistical relationship that Taubman (1979), which uses a Mincerian wage equa- would be measured using the most common economet- tion by adding to the usual variables (age, work experi- ric technique (i.e., ordinary least squares estimation). ence, years of schooling, family background) and indica- The proposed solutions to the endogeneity problem also tors of diseases, both physical and mental (heart disease critically depend on the health indicator used and the and hypertension, psychoses and neuroses, arthritis, potential measurement error associated with the given bronchitis, ulcers diseases of nerves, diseases of liver, and health indicator, because in some cases the particular bone diseases). In particular, they analyze the effects of kind of the measurement error can offset the bias result- such diseases on the basis of their year of onset, in order ing from the endogeneity problem. to disentangle short-term from long-term effects. A sim- ilar exercise was performed for this study by regressing The following three methods were used, all adopted wage rates (in natural logarithms and at 2000 prices) from the existing literature. The methodologies were and the number of hours worked per week (in natural applied to the main data source, the RLMS, in particu- logarithms) on a large set of the individual-specific lar for the four years from 1999 to 2002. The second health and non-health variables and environmental vari- methodology (instrumental variable estimation) was ables (see Table B2 for a list of these variables). The applied to the one-time (2003) NOBUS household sur- assumption of this approach-corroborated by a number vey, too. As health proxy, the different methodologies of statistical tests-is that endogeneity does not really 108 Table B2. Independent Variables Used in the Regression Analysis Using the RLMS Data Instrumental Variable Description variables* gender gender (male=1) age Age age2 squared age highsc high school diploma tecdp technical or medical diploma insdp institute or university diploma gradp doctoral degree married Married tenure experience at current workplace tenure2 squared experience at current workplace pjemps number of employees in enterprise ncat number children under 7 y.o. private private sector region_2 Northern and North Western region_3 Central and Central Black-Earth region_4 Volga-Vaytski and Volga Basin region_5 North Caucasian region_6 Ural region_7 Western Siberian region_8 Eastern Siberian and Far Eastern urban urban area occupation_2 professionals (ISCO-88 code) occupation_3 technicians and associate professionals (ISCO-88 code) occupation_4 clerks (ISCO-88 code) occupation_5 service workers and market workers (ISCO-88 code) occupation_6 skilled agricultural (ISCO-88 code) occupation_7 craft and related trades (ISCO-88 code) occupation_8 plant and machine operators and assemblers (ISCO-88 code) occupation_9 elementary (Unskilled) occupations (ISCO-88 code) round_10 year 2001 round_11 year 2002 round_12 year 2003 cheart chronic heart disease X clungs chronic lungs disease X cliver chronic liver disease X ckidny chronic kidney disease X cgi chronic stomach disease X spine chronic spine disease X cother other chronic diseases X (continued on page 110) 109 Table B2. Independent Variables Used in the Regression Analysis Using the RLMS Data (continued) Instrumental Variable Description variables* diabetes_10 diabetes diagnosed between 10 and 5 years ago X diabetes_20 diabetes diagnosed between 20 and 10 years ago diabetes_5 diabetes diagnosed less than 5 years ago diabetes_b20 diabetes diagnosed more than 20 years ago heart_10 heart attack diagnosed between 10 and 5 years ago X heart_20 heart attack diagnosed between 20 and 10 years ago heart_5 heart attack diagnosed less than 5 years ago heart_b20 heart attack diagnosed more than 20 years ago hepatitis_10 hepatitis diagnosed between 10 and 5 years ago X hepatitis_20 hepatitis diagnosed between 20 and 10 years ago hepatitis_5 hepatitis diagnosed less than 5 years ago hepatitis_b20 hepatitis diagnosed more than 20 years ago stroke_10 stroke diagnosed between 10 and 5 years ago X stroke_20 stroke diagnosed between 20 and 10 years ago stroke_5 stroke diagnosed less than 5 years ago stroke_b20 stroke diagnosed more than 20 years ago tbc_10 tuberculosis diagnosed between 10 and 5 years ago X tbc_20 tuberculosis diagnosed between 20 and 10 years ago tbc_5 tuberculosis diagnosed less than 5 years ago tbc_b20 tuberculosis diagnosed more than 20 years ago healthGOOD self-reported good health status misseddays missed work days due to ill health school_1 high school diploma completed before 2000 school_2 technical or medical diploma completed before 2000 school_3 institute or university diploma completed before 2000 school_4 doctoral degree completed before 2000 *Instrumental variables have only been used in the regressions summarized in Tables B5 and B6. matter given the specific health indicators used, so the tuberculosis reduces wage rate, as expected. Hepatitis use of OLS becomes justified. diagnosed very early reduces labor supply, while recently diagnosed tuberculosis increases labor supply. Indeed, Table B3 and Table B4 on page 113 report the results of respiratory and lung-related diseases (such as asthma four models that differ by the date of medical diagnosis and bronchitis) seem to have a positive effect on labor for diabetes, heart attack, stroke, tuberculosis and hepa- supply. Given the fact that respiratory diseases cause rel- titis (the only diseases for which the diagnosis date is atively little work limitations, a possible hypothesis available in the dataset). As expected lung, kidney, and explaining these findings could be that individuals seek spine chronic diseases reduce the wage rate (and hence to augment their revenues in order to compensate for productivity). Surprisingly, chronic lung disease increas- the additional medical care expenditures they require. es labor supply. Recently diagnosed heart attack and 110 Table B3. OLS - Dependent Variable: Log Hourly Wage Rate (Measured at 2000Prices) Variable Before20 Y20_10 Y10_5 Y5_0 gender .30254066*** .30310181*** .3024037*** .30367693*** age .03272136*** .03260822*** .03273228*** .03251867*** age2 -.00041325*** -.00041165*** -.00041368*** -.0004103*** highsc .07731209*** .07729698*** .0775283*** .07760015*** tecdp .08662943*** .08694422*** .08624272*** .08602663*** insdp .32191213*** .32172709*** .32240742*** .32086648*** gradp -.07311596 -.07601234 -.07762188 -.07221849 married .04515979*** .04471361*** .04424292*** .04513566*** tenure -.00126128 -.00124136 -.00126923 -.00110397 tenure2 .00011182* .00011086* .00011178* .0001044* pjemps 9.158e-06*** 9.184e-06*** 9.161e-06*** 9.201e-06*** ncat -.04025733*** -.04012284*** -.03946076*** -.04009372*** private .17656016*** .17648686*** .17705347*** .17630873*** region_2 -.02601835 -.02536182 -.0264554 -.0255168 region_3 -.46472316*** -.46428774*** -.46488513*** -.46488793*** region_4 -.71409733*** -.71366399*** -.7137759*** -.71324021*** region_5 -.61041382*** -.60970428*** -.61063961*** -.60931095*** region_6 -.48056355*** -.48006629*** -.48088991*** -.48145873*** region_7 -.48499262*** -.48461688*** -.48570588*** -.48480409*** region_8 -.29421497*** -.29363089*** -.29479044*** -.29217805*** urban .43861682*** .43867082*** .4389986*** .44019666*** occupation_2 -.01549473 -.0169319 -.01764095 -.01722858 occupation_3 -.1018942*** -.10313616*** -.10426042*** -.102774*** occupation_4 -.16137001*** -.16203158*** -.16283756*** -.16217807*** occupation_5 -.41726362*** -.41845074*** -.41930993*** -.41848629*** occupation_6 -.46935269*** -.47401677*** -.47384018*** -.47511778*** occupation_7 -.04230204 -.04379586 -.04411402 -.04298466 occupation_8 -.11553389*** -.11695315*** -.11749264*** -.11677098*** occupation_9 -.48967173*** -.48989482*** -.49126905*** -.49107733*** round_10 .17638925*** .17525387*** .17556241*** .17504897*** round_11 .38113833*** .38000061*** .38030253*** .37951903*** round_12 .47109966*** .46966934*** .47030324*** .46988307*** cheart -.02067898 -.01857795 -.01968821 -.01338115 clungs -.08023211** -.07860568** -.07878113** -.07764093** cliver -.00480458 -.00782106 -.00376398 -.01182401 ckidny -.04546527* -.04487214* -.04552355* -.0444479* cgi .01611436 .01571097 .01533843 .01483718 cspine -.03773294** -.03885295** -.0386692** -.03875688** cother -.02434006 -.02327219 -.02333522 -.02540852 (continued on page 112) 111 Table B3. OLS - Dependent Variable: Log Hourly Wage Rate (Measured at 2000Prices) (continued) Variable Before20 Y20_10 Y10_5 Y5_0 diabetes_b~0 .08708819 heart_b20 - stroke_b20 -.12886329 tbc_b20 -.11782447 hepatitis_b20 -.02362581 diabetes_20 -.08324869 heart_20 -.06870232 stroke_20 -.23865608 tbc_20 -.04481312 hepatitis_20 .00727449 diabetes_10 -.03340999 heart_10 .0153402 stroke_10 -.2775952 tbc_10 -.12228027 hepatitis_10 -.04278534 diabetes_5 .05831311 heart_5 -.13975016* stroke_5 -.10652745 tbc_5 -.23336728** hepatitis_5 .10332314 constant 1.2241776*** 1.2269543*** 1.226739*** 1.2266473*** R2 .3803084 .38032227 .38038093 .3806654 N 11297 11297 11297 11297 Note: Legend: * p<.1; ** p<.05; *** p<.01 Although this approach was used in the literature, its variables that are correlated with the endogenous vari- underlying assumptions are controversial. The following able but uncorrelated with the error term. The predicted two methods address the endogeneity issue in more values will then contain part of the information of the direct ways. original variable, but they will be purified from the cor- relation with errors. This approach was applied to both 2. Instrumental variables (IV) estimation. When endo- the RLMS and the NOBUS data. Since the surveys dif- geneity is explicitly taken into account, simultaneous fer, the precise specification of the estimation methodol- equation or instrumental variables approaches are typi- ogy differs slightly, too. cally the preferred option. Following this method, the endogenous variable (here: the health indicator) should RLMS be substituted by the predicted values coming from its Individual self-reported health status was used as the own regression over a set of instrumental variables plus health proxy in the first set of regressions, and the all the exogenous variables that are part of the model. reported number of workdays missed due to illness in The researcher must choose as instruments one or more 112 Table B4. OLS - Dependent Variable: Log Weekly Hours Variable Before20 Y20_10 Y10_5 Y5_0 gender .1082822*** .10816178*** .10869426*** .10820324*** age .01699662*** .01691379*** .01676215*** .01688689*** age2 -.00020535*** -.00020482*** -.00020268*** -.00020421*** highsc -.01192034 -.01158593 -.01202024 -.01188906 tecdp .00299955 .0030822 .00286354 .00316107 insdp .00574539 .00556812 .00571997 .00574958 gradp .01750542 .01471094 .01609397 .01712725 married -.02446514*** -.02487114*** -.02475416*** -.02468505*** tenure -.00206881** -.00206111** -.00207175** -.00207253** tenure2 .00005766** .00005736** .0000583** .00005743** pjemps -8.690e-07*** -8.490e-07*** -8.641e-07*** -8.489e-07*** ncat .00040164 .00039929 .00030107 .00046486 private .07633224*** .07656981*** .07671271*** .07617914*** region_2 .06115134*** .06051219*** .0609761*** .06085042*** region_3 .02044861* .02015783* .02036473* .02046827* region_4 .03674088*** .03693256*** .03682934*** .03713913*** region_5 .07975371*** .07938687*** .07960869*** .08008555*** region_6 .01340273 .01321734 .01336299 .01393258 region_7 .04322431*** .04314116*** .04295671*** .04327367*** region_8 .05036055*** .05051396*** .05041537*** .05130143*** urban .02271182*** .02296056*** .02270854*** .02260605*** occupation_2 -.17578267*** -.17522752*** -.17577442*** -.1761433*** occupation_3 -.07078628*** -.07051406*** -.07101583*** -.07137268*** occupation_4 -.0686374*** -.0680209*** -.06813237*** -.06915404*** occupation_5 .08926562*** .0897478*** .08947398*** .08936801*** occupation_6 -.02700664 -.02734797 -.02807609 -.02808858 occupation_7 -.10144504*** -.10061856*** -.10121411*** -.10116903*** occupation_8 -.01019046 -.00971794 -.0104076 -.01010831 occupation_9 -.12541047*** -.12508277*** -.12513555*** -.12552281*** round_10 .0042041 .00258998 .00310756 .00281578 round_11 -.00524675 -.00687934 -.00644211 -.0065833 round_12 -.00674471 -.0082841 -.00784437 -.00816691 cheart -.01282462 -.0135959 -.0118314 -.01302344 clungs .03941723*** .03918327*** .04054747*** .03694418*** cliver .01767289* .01477509 .01618572 .01378418 ckidny .0009254 .00122756 .00160304 .00111697 cgi -.00062025 -.00079791 -.00106807 -.00058534 cspine -.00464508 -.00489828 -.00504277 -.00521482 cother -.00086547 -.00073189 -.00019993 -.00135145 (continued on page 114) 113 Table B4. OLS - Dependent Variable: Log Weekly Hours (continued) Variable Before20 Y20_10 Y10_5 Y5_0 diabetes_b20 -.03719927 heart_b20 - stroke_b20 -.03803866 tbc_b20 -.01698457 hepatitis_b20 -.02917758** diabetes_20 -.02750776 heart_20 .05839574 stroke_20 .24507382* tbc_20 .01060056 hepatitis_20 .00925964 diabetes_10 -.04903766 heart_10 -.01289033 stroke_10 -.03361457 tbc_10 -.11833582 hepatitis_10 -.02558451 diabetes_5 .02017598 heart_5 .00539258 stroke_5 -.01842532 tbc_5 .19298307*** hepatitis_5 .01203015 constant 4.8475018*** 4.8493382*** 4.8524877*** 4.8500274*** R2 .14135195 .14153691 .14132912 .14158014 N 12009 12009 12009 12009 Note: Legend: * p<.1; ** p<.05; *** p<.01 the second. The latter is self-reported, too, and thus tus and missed days due to ill health. Tables B4 and B5 may be affected by measurement errors that are also sys- report estimates for both the logarithm of wage rate and tematically related to individuals' characteristics. This labor supply, separately by gender. Both indicators nega- indicator was used because it could be considered a tively affect the wage rate, but they do not have a signif- more specific indicator of work limitations than the icant influence on labor supply. A reported good health overall health status. Schultz and Tansel (1995) used the status increases the wage rate by 22% for women and by same indicator in another country context, interpreting 18% for men, compared to those who were not in good it as an "objective" measure of health status. Two kinds health. Similarly, a workday missed due to illness of estimations were performed for this study, and both reduces the wage rate by 3.7 percent in the male sub- follow Stern (1989) in the choice of instruments. Stern sample and by 5.5 percent among females. used medically diagnosed diseases to instrument for self- reported health indicators. The Sargan test of overidentification does not reject the hypothesis of exogeneity of the selected instruments. The variables in the third column of Table B1 are used Although this result must be interpreted only as an indi- as instruments for respectively self-evaluated health sta- cation of exogeneity, because the Sargan test has only 114 Table B5. IV­Dependent Variables: Log Deflated Wage Rate (WR) at 2000 Prices and Log Weekly Worked Hours (LS) Variable WRfullsample WRmale WRfemale LSfullsample LSmale LSfemale healthGOOD .20261634*** .1806543** .22419709*** -.01000299 .02130741 -.02027266 gender .27585464*** .1101072*** age .03614345*** .02352038*** .04389459*** .01683673*** .01700254*** .01448804*** age2 -.00043666*** -.00030836*** -.00051943*** -.00020515*** -.00020695*** -.00017566*** highsc .07089832*** .07871912*** .04943325* -.01161474 -.01891072** -.00217997 tecdp .08644842*** .10692356*** .06691908*** .0033794 -.01350005 .01483595* insdp .31426173*** .2392171*** .34637742*** .00540103 .02079802* -.00123836 gradp -.08415951 -.04194061 -.10088591 .01388915 -.02216501 .04175885 married .05141418*** .15022216*** .01163241 -.02538585*** .01937082* -.03886909*** tenure -.00081092 -.00546021* .00218432 -.00211222** -.00426801*** -.00063797 tenure2 .00009809 .00019499** .0000289 .00005859** .00011212*** .00001423 pjemps 8.804e-06*** 7.632e-06*** 9.004e-06*** -8.013e-07 -9.378e-07 -2.854e-07 ncat -.04521146*** .00215761 -.10828932*** .00057746 .01806025** -.02033957** private .17277442*** .09955215*** .24806507*** .07645701*** .06493953*** .07887115*** region_2 -.0133603 .09148693* -.0958643** .05869207*** .06096713*** .05765205*** region_3 -.44289043*** -.40460824*** -.48610782*** .01849206* .01404615 .01895777 region_4 -.70229557*** -.66428493*** -.7348911*** .03537858*** .01953694 .04064509*** region_5 -.61434647*** -.53340948*** -.67970519*** .07870559*** .05313381*** .09620142*** region_6 -.45873715*** -.33987848*** -.55962049*** .01230895 -.01643169 .03275026** region_7 -.46645139*** -.45783094*** -.47691641*** .04152007*** .08148437*** .00841795 region_8 -.2853641*** -.2289622*** -.32352706*** .0502703*** .06153782*** .03448115** urban .43761558*** .62312852*** .2831659*** .02274333*** -.01505292* .04760405*** occupation_2 -.01146613 -.02094595 -.00256181 -.17491303*** -.1825607*** -.17610764*** occupation_3 -.09973417*** -.04334927 -.10912741*** -.07085586*** -.0616262*** -.07649988*** occupation_4 -.15451163*** -.03652103 -.16638103*** -.06779682*** -.05827246* -.06896023*** occupation_5 -.41975459*** -.31318111*** -.48718413*** .08981792*** .0592768*** .10308978*** occupation_6 -.47198821*** -.37263434*** -.84237456*** -.02883948 -.04088518 -.03736733 occupation_7 -.03552898 -.05105415 -.05908826 -.10097471*** -.10827515*** -.02649565 occupation_8 -.10721544*** -.12569034*** -.04698364 -.01031805 -.01859227 .00006369 occupation_9 -.48071172*** -.55627943*** -.42828469*** -.12501984*** -.0406391** -.18950645*** round_10 .17584632*** .20261341*** .15252564*** .00302838 -.0035675 .00816274 round_11 .37643375*** .35788906*** .39169577*** -.00642209 -.00452699 -.00850983 round_12 .46681949*** .49302316*** .44638437*** -.00806578 -.01231146 -.00638784 constant 1.0524022*** 1.3785507*** 1.0700368*** 4.8578013*** 4.9607769*** 4.8823667*** R2 .38005142 .37336365 .37954554 .14009513 .10294687 .13476493 N 11297 5081 6216 12009 5425 6584 sargan 13.573047 11.927898 12.401589 17.049472 13.678117 19.081833 sargan p .25752479 .36908726 .33422615 .10642072 .25131896 .0596403 Note: Legend: * p<.1; ** p<.05; *** p<.01; health measure: self-reported health status 115 Table B6. IV­Dependent Variables: Log Deflated Wage Rate (WR) AT 2000 Prices and Log Weakly Worked Hours (LS) Variable WRfullsample WRmale WRfemale LSfullsample LSmale LSfemale misseddays -.05380539*** -.03690035* -.05546552*** .00821319 -.00709594 .01402738* gender .29772294*** .10998022*** age .03114485*** .01850996*** .04046707*** .01721108*** .01632242*** .0147364*** age2 -.00040003*** -.00026616*** -.00050769*** -.00020857*** -.00020076*** -.00017564*** highsc .0764457*** .08810853*** .04997693* -.01159807 -.01808813* -.00146113 tecdp .08665504*** .1073604*** .06226547*** .0031956 -.01300065 .01608852* insdp .31530499*** .24677968*** .33882159*** .00571132 .02204791* .00141949 gradp -.0787445 -.036784 -.08761471 .01393563 -.02159879 .04096062 married .05356679*** .15083713*** .01173622 -.02579732*** .01930354* -.03936297*** tenure -.00122834 -.00524941 .00082434 -.00211588** -.00420651*** -.00045703 tenure2 .00011241* .00019362** .00006644 .00005821** .00011106*** 8.823e-06 pjemps 9.280e-06*** 8.407e-06*** 8.816e-06*** -8.500e-07 -8.565e-07 -3.074e-07 ncat -.04023962*** .00362648 -.09662369*** .000693 .01770139** -.02175956*** private .16724902*** .09387912*** .24856076*** .07756117*** .06402571*** .08025452*** region_2 -.01474598 .09875703** -.11009453*** .0577796*** .06201305*** .057782*** region_3 -.4558448*** -.42070126*** -.49921143*** .01890107* .01165807 .01929232 region_4 -.71430958*** -.66690059*** -.75631568*** .03665627*** .01863438 .04386867*** region_5 -.61081329*** -.5267249*** -.68255719*** .07879792*** .05402857*** .09821436*** region_6 -.47894979*** -.34875802*** -.59416784*** .01363934 -.0175098 .03771347** region_7 -.48772765*** -.46787374*** -.51264077*** .04312625*** .08002961*** .01396134 region_8 -.29421679*** -.23611033*** -.33333955*** .05040509*** .06036483*** .03431264** urban .44175078*** .62699818*** .28492593*** .02193919*** -.01431369 .04659559*** occupation_2 -.0275641 -.04925416 -.00844466 -.17300785*** -.18745121*** -.1760129*** occupation_3 -.12230053*** -.05608059 -.1344305*** -.06794825*** -.06362624*** -.07169013*** occupation_4 -.17091698*** -.0763228 -.18042962*** -.06618904*** -.06534309** -.0677314*** occupation_5 -.43565594*** -.2985637*** -.52124317*** .09266941*** .06085781*** .11050034*** occupation_6 -.48343043*** -.38829406*** -.7552105*** -.02690202 -.04387496 -.05866459 occupation_7 -.04122031 -.05320954 -.07698807 -.10081479*** -.10856194*** -.02472499 occupation_8 -.12073531*** -.13498605*** -.0576109 -.00918769 -.01997502 .00046696 occupation_9 -.50227005*** -.56088127*** -.46485105*** -.12261359*** -.04186336** -.18370621*** round_10 .17854361*** .20508354*** .15430605*** .00259199 -.0038077 .00628939 round_11 .38042079*** .36002555*** .39748229*** -.00650397 -.00496189 -.00990774 round_12 .46644317*** .49288934*** .44663449*** -.00755301 -.01312552 -.00643482 constant 1.3008961*** 1.6037212*** 1.3262024*** 4.8369351*** 4.9923983*** 4.8495011*** R2 .32233376 .34607243 .31337964 .13220628 .09785919 .11655025 N 11297 5081 6216 12009 5425 6584 sargan 10.582327 13.854043 8.4962567 15.267294 13.358761 15.893294 sargan p .47888791 .24117537 .66828023 .17058417 .27052497 .14513876 Note: Legend: * p<.1; ** p<.05; *** p<.01; health measure: missed days due to ill health. 116 little power, it does support the NOBUS (IV) Bartel and Taubman (1979) Table B7. IV - Dependent Variable: Log Monthly Wage Rate assumption of exogeneity of the health conditions they used in their Variable Full Male Female OLS analysis. healthGOOD .23073613*** .29161317*** .18554934*** age .00194805 .00285506 .00034706 NOBUS male .2827457*** The NOBUS was used exclusively children -.0186142 .01235114 -.05409407*** for the instrumental variable proce- private .04593329** -.02217283 .16266443*** dure. Again, self-reported health sta- schooling2 .17295232*** .18103981*** .14660409*** tus served as a health proxy: the dummy healthGOOD comprises schooling3 .42042849*** .40874823*** .44509322*** both excellent and good self-rated experience 2 .15488742*** .21468458*** .09025464** health status (as was done in the case experience 3 .27605528*** .33826986*** .19469783*** of the RLMS analysis). A two-stage experience 4 .29482454*** .3339668*** .24849332*** least squares (2SLS) regression of the experience 5 .30288889*** .28737294*** .36047057*** logarithm of monthly wage rate and [98 omitted regional dummies] the logarithm of worked hours per urban .36058887*** .45050028*** .20029591*** week were used, respectively, on age, gender, number of children, private constant 6.3669247*** 6.468474*** 6.7210779*** sector employment, secondary school and university, length of work R2 .35884352 .34130484 .41305857 experience, location indicators, and N 4139 2410 1729 urban/rural indicator. Secondary sargan 2.3231368 4.2421652 .15670567 school and university are represented sargan p .12746276 .03943185 .69220781 by the values 2 and 3 of the categor- ical variable schooling derived from Source: NOBUS Dataset round 1: sample of jobholders whose family includes the parents. a NOBUS categorical variable that is healthGOOD instrumented by father and mother health status. ordered in 8 levels. Work experience Note: Legend: * p<.1; ** p<.05; *** p<.01 length comes directly from a NOBUS categorial variable ordered in 5 levels. The indicator urban assumes value 1 for all places with more that 20,000 inhabitants. For this study, one location The results, presented in Table B6 and Table B7, show indicator was included for each region. that health impacts wages more than labor supply (recall: among individuals who participate into the labor force). Individual health status has been instrumented by the In particular, males in good health earn about 30 percent mother's and father's health status. This may be justified more than the others (i.e., males with fair, bad, and very because many chronic diseases are inter-generationally bad health) and females 18 percent more. transmitted-either biologically or socially. Therefore, parents' health can be correlated with the health of their The Sargan tests reported in the bottom of Table B6 offspring without necessarily being correlated with the and Table B7 generally support the choice of instru- child's individual-specific omitted variables absorbed by ments used here (especially for females). Other instru- the error term. This choice, determined by data avail- ments-such as location indicators or the number of ability, meant having to limit the analysis for this study inhabitants to capture differences in the prevalence of to the sub-sample of jobholders who lived in households communicable diseases, differences in the availability of with their parents. Clearly, this may have caused a selec- medical facilities, differences in the prices of health tion bias, which is not easily addressed. inputs, and differences in environmental conditions- 117 Table B8. IV - Dependent Variable: Log of Weekly Worked Hours 3. Panel regressions. The third Variable Full Male Female approach exploits the longitudinal healthGOOD .03167153 .03403846 .02639951 dimension of the dataset by using age .00021789 -.00028859 .00090962 panel regression methods. Few studies on the relationship male .04823373*** between health and labor market children .01161265** .01740233** .00588029 outcomes have explicitly adopted private .04238258*** .02611874** .06846073*** panel data estimators. Recently, schooling 2 -.00202497 -.00121966 .0023968 Pelkowski and Berger (2004) have schooling 3 -.0291298*** -.02807691* -.02598168 studied the impact of health on experience 2 .02950427** .03812825** .01756524 employment, wages, and hours experience 3 .04732545*** .05443608*** .04033184* worked distinguishing between experience 4 .04869325*** .06112273*** .03543496 temporary and permanent impair- ments by using fixed effects esti- experience 5 .04743424** .0745132*** .01398805 mators. Here, another recent study [98 omitted regional dummies] was followed: it extensively adopts urban .00093956 -.01060702 .02084233 panel data analysis, i.e., constant 3.4491043*** 3.4881173*** 3.460679*** Cotoyannis and Rice (2001). The authors suggest the use of R2 .0451653 .04935113 .07885763 Hausman-Taylor estimators. In N 4488 2655 1833 terms of the previous problem of finding "good" instruments, the sargan 2.9013272 1.909446 .56854037 main advantage of this procedure sargan p .08850665 .16702481 .45083952 is that it does not require finding Source: NOBUS Dataset round 1: Sample of jobholders whose family includes the parents. valid instruments outside the healthGOOD instrumented by father and mother health status model, because it uses the already- included exogenous variables to Note: Legend: * p<.1; ** p<.05; *** p<.01 instrument the relevant endoge- nous variable. The only require- ment is the inclusion of both were tried but were rejected by the Sargan test. Also, the time-varying and time-invariant variables, each of inclusion of parents' age in addition to parents' health which has to be separated into exogenous and endoge- status increased the probability of the instruments' nous ones. Moreover, Hausman-Taylor estimators have endogeneity. the advantage over the usual within (fixed effects) esti- mators of allowing the effects of time-invariant variables Despite the positive signal previously offered by the to be consistently estimated. The disadvantage lies in Sargan test, concerns remain about the actual exogeneity the strong exogeneity assumptions to ensure consisten- of the chosen instruments. For instance, it seems reason- cy. For this reason, as in Cotoyannis and Rice (2001), able to think that high levels of labor supply may such exogeneity assumptions were tested for this study increase the probability of stomach diseases and hyper- by means of a Hausman (1978) test. Moreover, to fur- tension, because of the prolonged stress. Moreover, one ther improve the precision of the estimates found here, may think that heart attacks, strokes, or chronic heart the Amemiya-MaCurdy estimators were applied; they diseases are perhaps linked to individuals' risky lifestyle share the same spirit as Hausman-Taylor but make use choices (smoking, drinking, little physical exercise), of a more efficient set of instruments (essentially trans- which may be correlated with individual specific error formations of the HT instruments). A Hausman test components. Addressing these concerns necessitated between HT and AM estimators favored use of the moving from cross-sectional to panel analysis in the next latter. approach. (continued on page 123) 118 PANEL REGRESSIONS Table B9. PANEL - Dependent Variable Log Deflated Wage Rate at Prices 2000: MALES Variable OLS RE FE HT AM age .02117373* .03060971* .03371538 .01813034 .0234186 age2 -.02543067** -.03546528* -.02382196 -.01311323 -.0213325 tenure -.00772551 -.01517426** -.0206151*** -.01909465** -.01806072*** tenure2 .01210427 .04237933** .06945062*** .06422693*** .0584895*** pjemps 9.530e-06*** 9.788e-06** 9.429e-06 .0000118** .00001127** private .04570286 .03499996 .03313663 .0309431 .02631349 married .14033812** -.03676585 -.22294367** -.20775497** -.15196781* ncat -.05326126 -.01793909 .00538132 .01683112 -.00097233 healthGOOD .13197755*** .09158229*** .07569402** .07786367** .07551662** occupation_2 -.01218971 .00225306 .0043603 .00617929 .00224176 occupation_3 -.05167111 .02008562 .05246994 .05219877 .05232641 occupation_4 -.19906217 .2181705 .39427491** .39282628** .37307211** occupation_5 -.24020787** -.02854871 .125755 .13997179 .11371703 occupation_6 -.94112994** -.17614337 .10702769 .10056151 .0679691 occupation_7 .03866432 .10132112 .12744936 .12405609 .11974571 occupation_8 -.0429258 .00489442 .03180318 .03342481 .02176699 occupation_9 -.57218885*** -.28322658*** -.07798007 -.07972034 -.10124062 region_2 .27892274** .29314799 - .25565869 .26318595 region_3 -.29012799*** -.2805072* - -.41834883 -.32384047 region_4 -.48320866*** -.50339542*** - -.62553843** -.56938138** region_5 -.39498039*** -.40471495** - -.37981395 -.37930179 region_6 -.13100975 -.13158781 - -.26354191 -.16251081 region_7 -.65294516*** -.67053312*** - -.70585965** -.72745543*** region_8 -.08770838 -.05826851 - -.12961441 -.09358942 urban .53909976*** .52193122*** - .32506656 .39969081*** round_10 .168135*** .16313149*** .14594161*** .15260468*** .15498627*** round_11 .38907013*** .37849082*** .34758368*** .36074622*** .36456176*** round_12 .49593055*** .485547*** .44142583*** .46017735*** .46584738*** school_1 .25757494*** .32212736*** - .81691085 .83514246** school_2 .40336141*** .49213283*** - 1.9765633** 1.2196359*** school_3 .63699184*** .7444805*** - 1.4499061 1.3256559*** school_4 .66247112*** .72739066*** - .17519122 .57196294 constant 1.160447*** 1.0129677** 1.3497267 .74809819 .74063887 N 1096 1096 1096 1096 1096 Note: Legend: * p<.1; ** p<.05; *** p<.01; FE is fixed effect Hausman test fixed effects vs. random effect: chi2(20) = 40.65; Prob>chi2 = 0.0041 Hausman test fixed effects vs. Hausman-Taylor: chi2(19) = 1.12; Prob>chi2 = 1.0000 Hausman test Hausman-Taylor vs. Amemiya-Macurdy: chi2(19) = 3.08; Prob>chi2 = 1.0000 119 Table B10. PANEL - Dependent Variable: Log Weekly Worked Hours: MALES Variable OLS RE FE HT AM age .00968936** .01011673 .0014124 .00773793 .00550749 age2 -.01130532** -.01229069* -.00820846 -.01072686 -.0061599 tenure -.00395814* -.00350348 -.00269069 -.0029343 -.00371731 tenure2 .01254271* .01316617 .01322701 .01425271 .0161937* pjemps -1.109e-06 -5.444e-07 1.023e-06 2.967e-07 -1.640e-07 private .05022197*** .02641639 .00345284 .00441571 .00623633 married .09359559*** .07396736** .04708204 .04284235 .05958309 ncat .01429715 .02212276 .02753488 .02416649 .03011741 healthGOOD -.01965858 -.01468967 -.0137866 -.01445042 -.0149034 occupation_2 -.21092642*** -.12738774*** -.04395166 -.04416019 -.04812057 occupation_3 -.1103376*** -.10051006*** -.09634389*** -.09603576*** -.0996362*** occupation_4 -.08852414 -.17208368*** -.20984353*** -.20838166*** -.20470985*** occupation_5 .03435495 .02383147 .01944352 .0153265 .01634753 occupation_6 -.00552382 -.04861333 -.0707335 -.06588934 -.05051898 occupation_7 -.17901824*** -.1455056*** -.11574215*** -.11468832*** -.11785297*** occupation_8 -.09831434*** -.08156896** -.06571441 -.06633536 -.0683112 occupation_9 -.03111297 -.01936531 -.01017347 -.00940763 -.00781331 region_2 .07384698 .07319388 - .08981761 .06052615 region_3 -.0642961* -.06370392 - -.07252672 -.07501232 region_4 -.07546378* -.07729309 - -.10336157 -.08865404 region_5 -.03169123 -.02565908 - .03149358 -.04685389 region_6 -.08640749** -.08861418 - -.08514215 -.11174203 region_7 -.0443195 -.04286526 - -.04661855 -.03856058 region_8 -.06315885 -.07427061 - -.11929904 -.07699556 urban -.01937896 -.01609842 - -.07301517 -.01544788 round_10 .00777308 .01059797 .01816888 .0139167 .01233232 round_11 -.00697861 -.00330935 .00917258 .00074844 -.00239133 round_12 -.00645034 -.00263858 .01457219 .00227273 -.00210441 school_1 -.05903512** -.0635665 - .12727169 -.14966983 school_2 -.08106482*** -.08653917* - .22689688 -.03809073 school_3 -.05420356 -.0688878 - .3848647 -.24634658 school_4 -.07648919 -.14565492 - .39316961 -.92683147 constant 5.2020154*** 5.1969917*** 5.3382847*** 5.0510247*** 5.3499498*** N 1096 1096 1096 1096 1096 Note: Legend: * p<.1; ** p<.05; *** p<.01 Hausman test fixed effects vs. random effect: chi2(20)= 28.21; Prob>chi2 = 0.1046 Hausman test fixed effects vs. Hausman-Taylor: chi2(19)= 0.55; Prob>chi2 = 1.0000 Hausman test Hausman-Taylor vs. Amemiya-Macurdy: chi2(19) = 1.71; Prob>chi2 = 1.0000 120 Table B11. PANEL - Dependent Variable: Log Deflated Wage Rate at Prices 2000: FEMALE Variable OLS RE FE HT AM age .04884346*** .06736967*** .10066973*** .10790077*** .10325259*** age2 -.05421362*** -.07580477*** -.15605471*** -.15072374*** -.13079675*** tenure -.00005831 -.00346258 -.00820986 -.00729327 -.00662808 tenure2 -.00103402 .00719763 .02147504 .01986644 .01643053 pjemps .00001359*** .00001239*** -5.603e-06 1.181e-06 1.233e-06 private .22376717*** .072317** -.01437716 -.0136186 -.01310365 married .00028125 .00738042 .01341194 .01071813 .01188477 ncat -.10801699*** -.07377408** -.04957205 -.05391545 -.05380167 healthGOOD .00899523 .03121741 .02602269 .02761743 .02923847 occupation_2 .05577093 .07850739 .06758284 .06643888 .06439402 occupation_3 -.0153171 .02146124 .03148909 .03083616 .03104208 occupation_4 -.11040535* -.06521419 -.04835037 -.04741679 -.04775724 occupation_5 -.54344458*** -.2801505*** .01557992 .02031464 .01952723 occupation_7 .08008339 .06526647 .00884341 .01068766 .01395186 occupation_8 -.05201295 -.03782227 -.07510576 -.06741964 -.0661714 occupation_9 -.45957353*** -.20444463** .09278365 .09481375 .09614351 region_2 -.13999192* -.20911144 - -.48590165 -.34880907 region_3 -.50231503*** -.53116243*** - -.45795033 -.53031251 region_4 -.74096226*** -.77493104*** - -.89981922** -.873309** region_5 -.63540426*** -.69525462*** - -.6915516 -.78563166* region_6 -.5473698*** -.57912027*** - -.78280566 -.71864947* region_7 -.62834388*** -.66565366*** - -.90982369 -.80104975* region_8 -.37340449*** -.43559004*** - -.56746261 -.56762899 urban .15423442*** .18812914*** - .36136134 .25022075 round_10 .18149427*** .19284666*** .2380016*** .22420629*** .21243407*** round_11 .45971719*** .47197908*** .55111916*** .52550048*** .50231554*** round_12 .51680043*** .53104055*** .64696711*** .60956004*** .5742841*** school_1 .19448843** .18916976 - -6.0365962 -1.5936188 school_2 .25540653*** .30060589* - -6.093138 -1.2300934 school_3 .57598761*** .6274978*** - -4.4909463 -.75936396 school_4 .681436*** .74557585*** - -.8708852 1.1328886 constant .84034209*** .41563824 .38980734 5.8148068 1.4537398 N 1904 1904 1904 1904 1904 Note: Legend: * p<.1; ** p<.05; *** p<.01 Hausman test fixed effects vs. random effect: chi2(20) = 64.56; Prob>chi2 = 0.0000 Hausman test fixed effects vs. Hausman-Taylor: chi2(19) = 2.23; Prob>chi2 = 1.0000 Hausman test Hausman-Taylor vs. Amemiya-Macurdy: chi2(19) = 2.39; Prob>chi2 = 1.0000 121 Table B12. PANEL - Dependent Variable: Log Weekly Worked Hours: FEMALE Variable OLS RE FE HT AM age .01405474*** .00713686 -.02876607** -.02768314** -.02504998** age2 -.01520261*** -.00682048 .04564225*** .04468031*** .03637539*** tenure .00018417 .0015545 .00491668* .0046434* .00428144* tenure2 -.00177468 -.00700912 -.01839697** -.01760554* -.0160395** pjemps -7.685e-07 -2.912e-07 -2.633e-06 -2.617e-06 -2.063e-06 private .0723087*** .02815004* -.00652035 -.00634699 -.00584964 married -.06523011*** -.05414285*** -.01885375 -.01718639 -.02338578 ncat -.04694642*** -.05493409*** -.06310345*** -.06328587*** -.06176603*** healthGOOD -.01529577 -.02104664 -.02394817 -.02400828 -.02430819* occupation_2 -.13872762*** -.07864605*** -.01027135 -.01049274 -.01060721 occupation_3 -.04832293** -.02975507 -.02232382 -.02192721 -.02245031 occupation_4 .00178738 .01370745 -.01396924 -.01365074 -.01291123 occupation_5 .20324933*** .1690263*** .04756556 .04767363 .04907036 occupation_7 .00249755 .02887341 .02006936 .01969135 .01862799 occupation_8 .06178278** .07571579** .0622016 .06197632 .06119144 occupation_9 -.06500797** -.05805535 -.11941641** -.11816436** -.11771333*** region_2 .04269644 .03478438 - .14238722 .06589317 region_3 .00976103 .00053723 - -.01808684 -.00896455 region_4 .05898428** .05603467 - .24275572 .09263906 region_5 .08703557*** .07217828 - .19255138 .0737897 region_6 .01526735 .00665344 - .26225193 .07547088 region_7 -.02668769 -.02780075 - .2432143 .03633142 region_8 .0230401 .02041398 - .21246341 .07272393 urban .06394156*** .06786983*** - .01321326 .07288475 round_10 -.00400421 -.00251684 -.00893178 -.00912039 -.004995 round_11 -.0074286 -.00476587 -.01726122 -.01767266 -.00950599 round_12 -.01834118 -.01732908 -.0406428* -.04128489* -.0286993* school_1 .21837139*** .20248681*** - 5.6885155 .99684116 school_2 .2593469*** .23333548*** - 4.8112981* .91200172 school_3 .2436505*** .194317*** - 5.1919824 .7156517 school_4 .298923*** .23065273** - 2.1060542 -.56722012 constant 4.6019202*** 4.7393288*** 5.5207691*** .27965245 4.5953589*** N 1904 1904 1904 1904 1904 Note: Legend: * p<.1; ** p<.05; *** p<.01 Hausman test fixed effects vs. random effect: chi2(20) = 59.37; Prob>chi2 = 0.0000 Hausman test fixed effects vs. Hausman-Taylor: chi2(19) = 0.60; Prob>chi2 = 1.0000 Hausman test Hausman-Taylor vs. Amemiya-Macurdy: chi2(19) = 2.47; Prob>chi2 = 1.0000 122 (continued from page 118) To perform this study, the sample of all individuals who Table B13. Random Effects Panel Logit were followed in all the rounds 9-12 and who provided Regression Results answers to all the questions of the subject survey were used. This means only the sub-sample of jobholders Variable Coefficient could be considered. Due to attrition and the relatively Age -0.492 *** high frequency of missing responses, the sub-sample of Age squared 0.003 *** males is composed of only 274 individuals, each one Reference: male observed four times, while the sub-sample of females Female -0.423 *** has 476 individuals. To address the problem of an even- tual selection bias, similar estimations were performed, Age*female 0.013 *** whenever possible, on a significantly larger unbalanced Married -0.275 *** panel that produced similar results, giving even greater Cohabit -0.129 credence to the underlying findings. Widow or divorced -0.262 *** Chronic illness 0.228 *** In general this study found that good health status increases wage rate for males, but it does not substan- Poverty status 0.495 *** tially affect labor supply. This result is in line with what Household income -0.012 *** obtained in the cross-sectional IV estimators of the pre- Income*Chronic illness -0.014 ** ceding subsection. However, now the effect of good High school diploma -0.447 *** health is reduced: being in good health increases the Number of children in hous. -0.123 wage rate by about 7.5%. Surprisingly, good health Female*N. children 0.378 *** impacts neither wage rate nor labor supply among female workers, differring from what was obtained in Born in Russia -0.141 *** the cross-sectional instrumental variables estimations, Living in village 0.113 ** where the effect on female was even larger than the Constant 4.192 *** effect on male wage rate. Rho 0.141 ** For the sake of completeness, the "missed days due to Note: *** 1%-significance level, ill health" variable was used as alternative measure of ** 5%-significance level. health status. However, its coefficient was statistically insignificant both in the wage rate and in the labor supply model. Table B14. Average Predicted Probability to Retire in the Subsequent Period, by Sex and Income Level MALES FEMALES Not chronically Chronically Not chronically Chronically Income level ill ill ill ill Over the 95th percentile 0.18 0.18 0.26 0.20 Between 75th and 95th percentile 0.21 0.29 0.24 0.35 Between 50th and 75th percentile 0.23 0.40 0.27 0.46 Between 25th and 50th percentile 0.32 0.56 0.35 0.58 Below the 25th percentile 0.43 0.62 0.52 0.75 Source: Calculations based on RLMS rounds 9-11.Note: Results refer to the hypothetical individual described in the text. 123 Technical Details and Specific Table B15. Results of Cox Regression Model Results on the Impact of Chronic on Age to Retirement Illness on Retirement Variable Coefficient A Cox regression offers a means to estimate the precise Age -.492*** moment that an event takes place, as time proceeds. It squared age .003*** is usually employed in survival analysis, where the out- Female -.423*** come considered is death, but it can also be used to age*female .0132*** estimate the timing of retirement. For this study, a Cox Married -.275*** regression model was estimated on the age at retire- ment, using data from the 11th round of the RLMS Cohabit -.129 (2002), which provides retrospective information on widowed or divorced -.262*** job retirement. chronic illness .228*** poverty status .495*** Estimating a Cox regression model on the age to retire- household income -.0116*** ment. This is a model of a hazard regression where the hh income*chronic illness -.014** log hazard function of retirement log[h(t)] is assumed high school diploma -.447*** to be a linear function of a baseline hazard function and the effect of p covariates, formally n. children under 7 y.o. -.123 female*n. children under 7 .378*** log[h(t)] = log[h0(t)] + b1x1 = b2x2 + K + bpxp born in Russia -.141*** Thus, the parameters estimated represent a proportional living in village .113** shift of the baseline hazard function due to the covari- ates. A positive parameter means an increase of the risk Note: *** 1%-significance level, ** 5%-significance level. of retiring during the overall time period (since first "hh" is household head. Breslow method for ties employment). The results are shown in Table B15. The No. of subjects = 8266 reported coefficients should be interpreted as follows: a No. of failures = 3225 positive coefficient means an increase in the risk of Time at risk = 349639.2499 experiencing the event (retirement in this case) and a LR chi2(16) = 3706.30 negative coefficient is associated with a decrease in the Prob > chi2 = 0.0000 risk of experiencing the event. (The test based on Log likelihood = -22630.901 Schoenfeld residuals showed that the null hypothesis- chronic illness effect is proportional-is not rejected.) A set of demographic and socioeconomic indicators the effect decreases with age. The effect of weight is (e.g., age, gender, income, education) was controlled interesting: those who are below the normal weight (in for. The health variable of particular interest is the pres- terms of body mass index) retire earlier from the labor ence of a chronic illness. A positive coefficient on the market, whereas those who are above it (overweight and chronic illness variable indicates an increase in the obese) are more likely to retire later. Reported drinking probability (i.e., the hazard) of retiring, relative to the does not have any significant effect, but chronic illness baseline first year of employment. has a positive and highly significant effect. This means that after having controlled for the other factors, in Those who are married are more likely to retire later contrast with findings from the Kaplan-Meier estimates, from the job market than those who never married. the research here indicates that those suffering from any Those who are widows or divorced also retire later than chronic disease are more likely to retire earlier. the never married. The effect of age is U-shaped. Moreover, the effect of chronic illness interacts with Females retire later but the effect is weak and decreases income: the higher the income level, the weaker the with age. Smoking brings a higher risk of retiring, but effect of chronic illness. In addition, the research finds 124 that workers below the poverty line retire earlier and Technical Details and Specific that income has a negative effect (i.e., the higher the Results on the Analysis of the income level the later a worker retires). The number of Impact of Chronic Illness on Income children has no significant effect for males, but it has a positive effect for females. Finally, the estimates from In order to address the endogeneity problems involved the Cox model suggest that people born in Moscow are in estimating the effect of health on economic out- more likely to retire later, and those living in a village comes, a strategy that does not employ instrumental are more likely to retire earlier. variables was used here. A difference-in-differences esti- mator combined with a propensity score matching tech- Technical Details and Specific nique (Rosembaum and Rubin 1983; Heckman, Results of Panel Probit Model on the Ichimura, and Todd 1997) was used. With this approach, every household experiencing a health prob- Probability of Being Fired lem is matched to a similar household that did not have A probit model was estimated for this study of the health problems. Similarity is defined in terms of a probability of being fired, which was made dependent propensity score, i.e. the propensity of experiencing a on gender, age (in months), wage rate, possession of health adverse event given the household characteristics high school diploma, post-secondary years of schooling, (for instance whether the household members suffer work experience, type of enterprise ownership (state, from chronic illness). In this way, by comparing the foreigners, or private Russian owners), and, finally, daily experiences of two similar households, the causal effect alcohol consumption (in grams of pure alcohol) and of health on income can be identified. The logic is squared daily alcohol consumption. The dummy vari- essentially that of comparing two groups that differ only able "fired" was defined such that it takes the value 1 if an individual was employed in Table B16. Panel Probit Results on Alcohol as Determinant of Being Fired round 11 (2002), he was not employed in round 12 (2003), Robust and yet he participated the Variable dF/dx Std. Err. Z P>z x-bar workforce in round 12. An gender -.0020767 .0023638 -0.89 0.373 1.54277 alternative definition embody- age .0000553 .000028 2.10 0.036 472.353 ing the condition of being wage rate -1.53e-06 6.95e-07 -2.26 0.024 3422.47 unemployed in round 12 pro- diploma -.0042912 .0035533 -1.25 0.213 1.14186 duced a very similar identifica- yrs sch. -.0011318 .0005492 -2.04 0.042 3.28421 tion. Through the chosen set- up, alcohol consumption was experience -.001044 .0003576 -3.36 0.001 19.0252 assumed to have a nonlinear state* -.0020864 .0032861 -0.66 0.511 .679367 effect on the probability of foreingn* .0085151 .0085096 1.37 0.169 .047688 being fired. This supposition private* .0050818 .0032222 1.72 0.086 .426312 was confirmed by other analy- alcohol .0002961 .0001112 2.25 0.025 15.5616 ses. The Huber/White/sand- alcohol2 -2.84e-06 1.10e-06 -2.04 0.042 1818.48 wich estimator of variance was applied in place of the tradi- tional calculation to obtain obs. P .015816 robust standard errors. The pred. P .0081124 (at x-bar) detailed results are shown in Note: dF/dx is for discrete change of dummy variable from 0 to 1 Table B16. z and P > |z| are the test of the underlying coefficient being 0 Number of obs = 4173 Pseudo R2 = 0.0812 Wald chi2(11) = 60.89 Log likelihood = -311.60966 Prob > chi2 = 0.0000 125 in relation to the variable of interest. Table B17. Results from Difference-in-Differences Estimator This makes it possible to assess the Combined with Propensity Score Technique: impact of intervention free from most The Effect of Adverse Health on Total Income for other contingent effects. Different Periods The results, reported in Table B17, Total income Total income Total income show the effect on total income of two 1994-1998 1998-2002 full period different events related to poor health: Health problems -22.255 -135.98*** 83.147*** generic health problems and hospitali- Hospitalization -136.19*** -105.83*** 82.30*** sation. Two separate estimates are made for the periods 1994-98 and Source: Suhrcke et al (2005). 1998-2002, taking into account the Note: *** 1%-significance level, ** 5%-significance level. economic crisis in Russia that began in 1998. The results confirm a negative effect of poor health on household economic well-being. illness that corresponds to 5.6 percent of median per This effect is greater in the later period. capita income. To estimate the specific impact of chronic diseases, a logit Technical Details and Results of the model was used to assess whether and to what extent Effect of a Household Member's chronic illness increases the likelihood of experiencing adverse health events. The corresponding results are Death on Depression reported in the accompanying technical papers. It can be The first analysis was performed using a probit model. concluded that chronic illness increases the risk of health The dependent variable is a dummy, which indicates the problems, hospitalization, and a surgical procedure. status of depression. The explanatory variables are: The results confirm that chronic illness does indirectly 4. Gender (male = 1) and negatively affect the economic well-being of 5. Age (in months) Russian households, especially since the economic crisis 6. Jobholder (yes = 1) in 1998. But what can be said about the magnitude of 7. Difference in per-capita income (after and the effect? It is not possible to provide a comprehensive before death) answer, since the risk of health problems depends not only on the presence of chronically ill persons in the 8. High school diploma (yes = 1) household, but also on other factors (number of smok- 9. Number of dead members throughout the past year ers, household size, number of older people, etc.). 10. Number of dead members who were However, a specific answer for a specific population can household-heads be provided: if there are households in urban areas with no smokers and no ex-smokers, no people aged over 60 11. Number of dead members who held a job nor below 14, with at least two workers and at least one 12. Age of dead members person who has a high school diploma. For this restrict- ed population the average difference in the probability of having health problems between households with Technical Details and Results of the chronically ill members and households without such Effect of a Household Member's people is 0.219. The difference in the probability of Death on Alcohol Consumption being hospitalized is 0.038, and the difference in the probability of undergoing a surgical procedure is 0.018. This analysis on alcohol consumption uses a tobit Multiplying these differences by the effect of health model. Pure alcohol consumption in grams per day was problems, hospitalization, and surgical procedure on regressed on: economic outcomes gives the indirect effect of chronic 1. Gender (male = 1) illness on income. This gives an impact of chronic 126 Table B18.Probit Results on the Effect on Depression Variable dF/dx Std. Err. z P>z x-bar gender -.0973314 .0085112 -11.05 0.000 .432762 age .000119 .0000197 6.02 0.000 525.495 job hold -.0199527 .0096232 -2.08 0.037 .554542 diff inc 6.60e-07 9.13e-07 0.72 0.470 675.682 high sch .0232161 .0095614 2.39 0.017 .67891 N death .5328651 .1086558 4.91 0.000 .032047 D hh head .0377452 .0509147 0.78 0.433 .012819 D worker .1132259 .1072779 1.19 0.233 .002712 dead age -.0004461 .0001198 -3.73 0.000 26.545 obs. P .1863676 pred. P .1777732 (at x-bar) Note: Number of obs = 8113 LR chi2(9) = 321.50 Prob > chi2 = 0.0000 Pseudo R2 = 0.0412 Log likelihood = -3740.8969 Table B19. Tobit Regression Result on Alcohol Consumption in Response the Household Member's Death Variable Coef. Std. Err. t P>t gender 36.46903 1.533006 23.79 0.000 age -.0100391 .0036239 -2.77 0.006 jobholder 23.21008 1.68241 13.80 0.000 diff income .0004966 .0001596 3.11 0.002 high school 10.74956 1.762026 6.10 0.000 n. deaths 10.54596 5.278959 2.00 0.046 d. hh head 4.400213 9.017779 0.49 0.626 d. worker 25.18829 14.66649 1.72 0.086 constant -44.9459 2.830359 -15.88 0.000 se 60.72394 .6605658 (Ancillary parameter) Obs. Summary 3677 left-censored observations at alcohol<=0 4493 uncensored observations Note: Number of obs = 8170 LR chi2(8) = 1002.07 Prob > chi2 = 0.0000 Pseudo R2 = 0.0183 Log likelihood = -26843.276 127 2. Age (in months) tor, well above the 2 percent that is well-known in the 3. Jobholder (yes = 1) empirical growth literature. However, as Islam (1995) noted, convergence rates increase dramatically in a panel 4. Difference in per-capita income (after and before data context. The long-run convergence rate is then death) mixed with business cycle effects. Concerning the vari- 5. High school diploma (yes = 1) able of interest in this study, the lagged adult mortality rate is found to be highly significant for both estimators 6. Number of dead members throughout the past year with negative sign as expected. Hence, the larger the 7. Number of dead members who were household- mortality rate, the lower the GDP per capita growth. heads To assess further the economic significance of this effect, 8. Number of dead members who held a job these alternative growth regressions were used here to predict Russian GDP per capita up to the year 2025. Technical Details and Specific This requires an assumption about the future path of Results of Economic Growth Impact the fertility rate, which was taken from the UN Estimates Population Division forecasts. The openness status of the Russian economy is assumed to stay constant over This estimation starts by running a standard pooled the next 20 years as the key question relates to the dif- ordinary least squares (OLS) panel growth regression for ferent mortality scenarios. An increase in openness the period 1960 to 2000. The dependent variable is the would not change results dramatically, although the annual average of the five-year growth rate of real GDP growth path would become somewhat steeper. per capita. The other explanatory variables are the five- year time lag of GDP per capita, the lagged fertility As for the adult mortality scenarios, the three different rate, the lagged working age mortality rate,22 and the scenarios described in chapter 11 were used to separately Warner-Sachs index of openness.23 The fertility rate is carry out a forward prediction on the OLS and FE esti- from the World Development Indicators and the adult mates. The results are shown in Figure B1. (The three mortality rate is constructed from the WHO mortality scenarios with the more conservative results are identical database. to the results presented in Figure 11.2). Since OLS panel growth regressions yield downward As Figure B1 illustrates, the predicted per capita GDP biased estimates on the projected growth rate (Trognon path is highly dependent on the choice of estimation 1978), a fixed effect (FE) estimator is applied to the methodology. As expected, the FE estimates produce a same regression equation. The FE regression is known to yield upward biased estimates Table B20. Growth Regression Results on the projected growth rate (Nickel 1981). Hence, the Dependent variable: GDP per capita. OLS FE unbiased growth path is bounded by the OLS and FE Lagged-GDP p.c. .86*** (.02) .65*** (.05) estimates. The regression Lagged fertility growth rate -.05 (.03) -.17*** (.06) results of the OLS and FE Openness .16*** (.02) - regressions are shown in Table Lagged adult mortality rate -.08** (.04) -.18*** (.06) B20. R2 0.97 0.98 No. of observations 302 332 The results in Table B20 show a convergence rate of 14 per- Notes: Heteroscedasticity-consistent standard errors in parenthesis. *, **, *** denote sig- cent with OLS or even 35 per- nificance at the 10%-, 5%-, and 1%-level, respectively. Constant terms are not reported. cent with fixed effect estima- 128 steeper growth path than the OLS Figure B1. GDP per Capita Forecasts Based on OLS and FE Regression estimates, and the "true" effect will US$ PPP lie somewhere in between. In 32,000 either type of estimate, however, there is a sizeable impact of the 30,000 reduction of mortality rates on 28,000 future incomes, and the effect Scenarios 1, 2, 3 Fixed Effects 26,000 grows over time. While in 2005 the difference in the per capita 24,000 GDP between the first and the 22,000 third is only US$105 in the OLS 20,000 estimation (and US$325 in the FE estimates), by 2025 this difference 18,000 would have grown to US$2,856 16,000 (respectively US$9,243). Even if 14,000 these future returns are discounted to the starting year value, they do 12,000 Scenarios 1, 2, 3 OLS make the static GDP effects calcu- 10,000 lated in the more narrow approach 8,000 of the previous section appear very 2000 2005 2010 2015 2020 2025 small. Source: Source: Suhrcke et al (2005). Calculations based on model presented in Table B8. 129 BIBLIOGRAPHY Bibliography Abelson, P. 2001."Returns on Investment in Public Andreev, E., E. Nolte, V. M. Shkolnikov, E. Varavikova, Health." An Epidemiological and Economic Analysis and M. McKee. 2003. "The Evolving Pattern of Prepared for the Department of Health and Ageing, Avoidable Mortality in Russia," International Australia. Journal of Epidemiology 32: 437-46. http://www.health.gov.au/internet/wcms/ AOA (American Obesity Association). 2005. "Obesity: A publishing.nsf. Global Epidemic." AOA Fact Sheets. Updated May 2. Abdullaev, N. 2004. "Men Show Little Regard for Their http://www.obesity.org/subs/fastfacts/obesity_global_ Own Safety." The Moscow Times Newspaper. epidemic.shtml (checked August 2005). December 8. Arrow, K. J. 1963. "Uncertainty and the Welfare ADA (American Diabetes Association). 2003. Economics of Medical Care." The American "Economic Costs of Diabetes in the US in 2002," Economic Review LIII(5): 941-73. Diabetes Care 26: 917-32. Ashton, S. J., G. M. Mackay, and S. Camm. 1983. AHA (American Heart Association). 2003. "Heart and "Seat Belt Use in Britain under Voluntary and Stroke Statistics-2003 Update." AHA, Dallas, TX. Mandatory Conditions." In Proceedings of the 27th Alberti, K. G. M. M. 1991. "Role of Diabetes," British Conference of the American Association for Medical Journal 303: 769-72 Automotive Medicine. Chicago, IL: 65-75. Allin, S., E. Mossialos, M. McKee, and W. Holland. Australia Commonwealth Department of Health and 2004. "Making Decisions on Public Health: A Aged Care. 2003. Returns on Investment in Public Review of Eight Countries." European Observatory Health. ISBN: 0 6428219 1 7. on Health Systems and Policies, World Health Babor, T. F., R. Caetano, S. Casswell, G. Edwards, N. Organization, Geneva. Giesbrecht, K. Graham, J. W. Grube, P. J. Alsan, M., D. E. Bloom, and D. Canning. 2004. "The Gruenewald, L. Hill, H. D. Holder, R. Homel, E. Effect of Population Health on Foreign Direct Österberg, J. Rehm, R. Room, and I. Rossow. 2003. Investment." NBER Working Paper 10596, National Alcohol: No Ordinary Commodity. Research and Bureau of Economic Research, Cambridge, MA. Public Policy, Oxford Medical Publication, Oxford University Press, Oxford. Amemiya, T., and T. E. MaCurdy. 1986. "Instrumental- Variable Estimation of an Error Components Babor, T. F., and M. Grant (eds). 1992. Project on Model," Econometrica 54: 869-81. Identification and Management of Alcohol-Related Problems. Report on Phase II: A Randomised Anderson, P. 1991. "Alcohol as a Key Area" in "The Clinical Trial of Brief Interventions in Primary Health of the Nation: Responses," British Medical Health Care. Geneva: World Health Organization. Journal 303: 766-69. Babor, T. F., A. Wilson, C. Campillo, F. K. Del Boca, et Anderson, P., and G. Lehto. 1994. "Prevention al. 1996. "A Cross-National Trial of Brief Policies," British Medical Bulletin 50(1): 171-85. Interventions with Heavy Drinkers," American Andreev, E. M. 2005. "Demographic Consequences of Journal of Public Health 86(7): 948-55. Mortality Reversal in Russia." Paper for the XXV IUSSP International Population Conference, Section Bakilana, A.M. 2005. "The Demographic Situation in 36: "Demographic and Socio-Economic Russia." A Background paper prepared for this study. Consequences of Adverse Mortality and Health World Bank. Washington, D.C. Trends," Tours, France, July18. Baldwin, M., L. Zeager, and P. Flacco. 1994. "Gender Andreev, E. M., M. McKee, and V. M. Shkolnikov. Differences in Wage Losses from Impairments," 2003. "Health Expectancy in the Russian Federation: Journal of Human Resources 29: 865-87. A New Perspective on the Health Divide in Europe," Bulletin of the World Health Organization 81(11): 778-87. Epub 2004 Jan 20. 131 Barr, N. 1994. "The Role of Government in a Market Bliss, T. 2004 "Lessons Learned from Road Safety Economy." In Labor Markets and Social Policy in Management in New Zealand and Their Application Central and Eastern Europe. The Transition and to Vietnam." Presentation at World Bank Institute, Beyond, ed. N. Barr. New York: Oxford University The World Bank, Washington, DC, November 18. Press. Published for the World Bank and the London --­. 2004. "Implementing the Recommendations of School of Economics and Political Science. the World Report on Road Traffic Injury --­. 1998. The Economics of the Welfare State. Prevention," Transport Notes TN-1 Oxford: Oxford University Press. Bloom, D.E., D. Canning, and B. Graham. 2003. Barro, R. 1991. "Economic Growth in a Cross-Section "Longevity and Life-Cycle Savings," Scandinavian of Countries," Quarterly Journal of Economics, Journal of Economics 105: 319-38. CVI(425): 407-43. Bloom, D. E., D. Canning, and D. T. Jamison. 2004. --­. 1996. "Health and Economic Growth." PAHO "Health, Wealth, and Welfare; Finance and (Pan American Health Organization) Program on Development." International Monetary Fund. Public Policy and Health. Washington, D.C. http://www.imf.org/external/pubs/ft/fandd/2004/03/ --­. 1997. Determinants of Economic Growth: A pdf/bloom.pdf. Cross-Country Empirical Study. Cambridge, MA: Bloom, D., D. Canning, and J Sevilla. 2001. "The MIT Press. Effect of Health on Economic Growth: Theory and Bartel, A., and P. Taubman. 1979. "Health and Labor Evidence." NBER Working Paper 8587, National Market Success: The Role of Various Diseases," The Bureau of Economic Research, Cambridge, MA. Review of Economics and Statistics 61(1): 1-8. www.nber.org/papers/w8587. Baskakov, V., and E. Yanenko. 2005. "The Evaluation of --­. 2002. "Health, Worker Productivity, and Insurance Risks and Disability Pension Insurance," Economic Growth."School of Public Policy and Pension Funds and Investments No. 1. Management, Carnegie Mellon University, Pittsburgh. Bazzoli, G. J. 1985. "The Early Retirement Decision: New Empirical Evidence on the Influence of Bobadilla, J. L., J. Frenk, T. Frejka, R. Lozano, and C. Health," Journal of Human Resources 20: 214-34. Stern. 1993. "The Epidemiological Transition and Health Priorities." In Disease Control Priorities in Beaglehole, L. 2004. "Diabetes Action Now : an initia- Developing Countries, eds. D. T. Jamison, W. H. tive of the World Health Organization and Mosley, A. R. Measham, and J. L. Bobadilla, eds. International Diabetes Federation" World Health New York: Oxford University Press. Organization and International Diabetes Federation. http://www.who.int/diabetes/actionnow/en/DANboo Bobak, M., M. Kristenson, H. Pikhart, and M. klet.pdf. Marmot. 2004. "Life Span and Disability: A Cross Sectional Comparison of Russian and Swedish Beaglehole, R., and D. Yach. 2003. "Globalisation and Community Based Data," British Medical Journal the Prevention and Control of Non-communicable 329(7469): 767. Disease: The Neglected Chronic Disease of Adults," The Lancet 362: 903-08. Bobak, M., H. Pikhart, C. Hertzman, R. Rose, and M. Marmot. 1998. "Socioeconomic Factors, Perceived Bennett, J. 2003. "Investment in Population Health in Control and Self-reported Health in Russia. A Cross- Five OECD Countries," OECD Health Working sectional Survey," Social Science Medicine, 47(2): Paper 2, Organisation for Economic Co-operation 269-79. and Development, Paris. Bonita, R., M. de Courten, T. Dwyer, K. Jamrozik, and Berkovec, J., and S. Stern. 1991. "Job Exit Behavior of R. Winkelmann. 2001. "Surveillance of Risk Factors Older Men," Econometrica 59: 189-210. for NCDs:" The WHO STEPwise Approach. Bhargava, A., D. T. Jamison, and C. Murray. 2001. Summary. World Health Organization, Geneva. "Modelling the Effects of Health on Economic Bound, J. 1991. "Self-reported versus Objective Growth," Journal of Health Economics 20: 423-40. Measures of Health in Retirement Models," Journal of Human Resources. 26: 106-38. 132 Bound, J., T. Stinebrickner, and T. Waidmann. 2003. Chenet, L., M. McKee, D. Leon, V. Shkolnikov, and S. "Health, Economic Resources and the Work Vassin, 1998. "Alcohol and Cardiovascular Mortality Decisions of Older Men," National Institute in Moscow; New Evidence of a Causal Association." on Aging. Journal of Epidemioogy and Community Health 52: http://socserv.socsci.mcmaster.ca/cesg2003/ 772-74. stinepaper.pdf. Cherkesov, V. 2005. "On Investigation Operations of Bozicevic, I., B. Lokrantz, M. McKee, and M. Suhrcke, the Russian Federation Federal Services for Control 2004 "Evidence Base on How to Reduce the Burden Over Drug Turnover in 2004 and Objectives for of Disease due to Injuries in Low-Resource Settings 2005." Report at the full-scale meeting of the Board with Special Attention to Eastern Europe and Central of the Russian Federation Federal Services for Asia." Draft Report, World Health Organization, Control Over Drugs Turnover: Press release. February Geneva. 18. Brainerd, E., and D. M. Cutler. 2004. "Autopsy on Chervyakov, V.V., V. M. Shkolnikov, W. A. Pridemore, an Empire: Understanding Mortality in Russia and and M. McKee. 2002. "The Changing Nature of the Former Soviet Union." NBER Working Paper Murder in Russia," Social Science & Medicine 10868, National Bureau of Economic Research, 55(10): 1713-24. Cambridge, MA. Chopra, M., and I. Darnton-Hill. 2004. "Tobacco and http://www.nber.org/papers/w10868. Obesity Epidemics: Not So Different After All?" Brooks, D. "Mourning Mother Russia." Washington British Medical Journal 328(7455): 1558-60. (DC) Post, April 28, 2005. CINDI (Countrywide Integrated Noncommunicable Broughton, J.; Allsop, R. E.; Lynam, D. A. 2000. "The Diseases Intervention) Program. 2004. "Behavioral Numerical Context for Setting National Casualty Risk Factor Surveillance System Development in Reduction Targets." TRL Report 382, Transport Russia in 2004." Technical Report. Research Laboratory Ltd, Crowthorne, Berkshire, http://www.cindi.ru. United Kingdom. Cirillo, J. A., and F. M. Council. 1986. "Highway Caballero, B. 2005. "A Nutrition Paradox-Underweight Safety: Twenty Years Later," Transportation Research and Obesity in Developing Countries," New Record, 1068: 90-95. England Journal of Medicine 352(15): 1514-16. Clark, G. L. 2004. The Macro-Economic Context. Califano, J. A. 1994. Radical Surgery. What's Next for University of Oxford. Oxford, United Kingdom. America's Health Care. New York: Times Books, CMH (Commission on Macroeconomics and Health). Random House. 2001. "Macroeconomics and Health: Investing in Casey, B., H. Oxley, E. Whitehouse, P. Antolin, R. Health for Economic Development." Report of the Duval, and W. Leibfritz. 2003. "Policies for an Aging Commission on Macroeconomics and Health, Society: Recent Measures and Areas for Further chaired by Jeffrey Sachs, World Health Organization, Reform." OECD Economics Department Working Geneva. Paper 369, Organization for Economic Co-operation Cohen, L., and S. Swift. 1997. "Beyond Brochures: and Development, Paris. Preventing Alcohol-Related Violence and Injuries. Cercone, J. A. 1994. "Alcohol-Related Problems as an Draft Report, Prevention Institute, Berkeley, CA. Obstacle to the Development of Human Capital." Coile, C. 2003. "Health Shocks and Couples' Labor World Bank Technical Paper 219, The World Bank, Supply Decisions." CRR Working Paper 08, Center Washington, DC. for Retirement Research, Boston College, Boston, Cha A. E. 2005. "Firms Make It Their Business to Push MA. Health: Incentives, Monitoring Aimed at Cutting Commission on the Future of Health Care in Canada. Costs." Washington (DC) Post. February 20. 2002. "Building on Values: The Future of Health Chaloupka, F. J., M. Grossman, and H. Saffer. 2002. Care in Canada." Final Report. Commission on "The Effects of Price on Alcohol Consumption and Future of Health Care in Canada, Saskatoon, Alcohol-Related Problems," National Institute on Canada. Alcohol Abuse and Alcoholism. http://www.hc-sc.gc.ca/english/pdf/romanow/pdfs/ http://www.niaaa.nih.gov/publications/arh26-1/ HCC_Final_Report.pdf 22-34.htm. 133 Cook, I. G., and T. J. B. Dummer. 2004. "Changing Davis, C. 2005. "Economic Consequences of Changes Health in China: Re-evaluating the Epidemiological in the Health Status of the Population and Economic Transition Model," Health Policy 67(3): 329-43. Benefits of Medical Programmes in the USSR during Coote, A. 2004. Prevention Rather Than Cure: Making 1950-1991." Background Paper prepared for the the Case for Choosing Health. ISBN 1 85717 485 2, Macroeconomics and Health Report for Eastern Kings Fund, London. Europe and Central Asia. European Office for Investment for Health and Development, World Coote, A., J. Allen, and D. Woodhead. 2004. Finding Health Organization, Venice. Out What Works. Building Knowledge about Complex, Community-Based Initiatives. King's Daynard, R. A. 2003. "Commentary: Lessons from Fund, London. Tobacco Control for the Obesity Control Movement," Journal of Public Health Policy 24 Cornia, G. A. 2002. "The Forgotten Crisis: Transition, (3-4): 291-95. Psychosocial Stress and Mortality over the 1990s in the Former Soviet Block." In Investment for Health: Demin, A. K. 1993. "Premature Population Mortality A Discussion of the Role of Economic and Social Patterns on the Territories of Russia: Implications Determinants. Geneva: World Health Organization. from Action." IIKTL Symposium, Publications of the National Public Health Institute B4/93, Helsinki. Costa, D. L., and M. E. Kahn. 2004. "Changes in the Value of Life, 1940-1980," Journal of Risk and Dinh-Zarr, T.B., D.A. Sleet, R.A. Shults, et al. 2001. Uncertainty 29(2). "Reviews of Evidence Regarding Interventions to Increase the Use of Safety Belts," American Journal Cotoyannis, P., and N. Rice. 2001. "The Impact of of Preventive Medicine 21(4S): 48-65. Health on Wages: Evidence from the British Household Panel Survey," Empirical Economics 26: Drucker, P. F. 1999. Management Challenges for the 599-622. 21st Century. Burlington, MA: Butterworth- Heinemann. Crafts, N. 2003. "The Contribution of Increased Life Expectancy to Growth of Living Standards in the Druss, B. G., S. C. Marcus, M. Olfson, and H. A. UK, 1870-2001." Unpublished manuscript, London Pincus. 2002. "The Most Expensive Medical School of Economics, London. Conditions in America." Health Affairs 21(4): 105- 11. Currie, Janet M., and Brigitte C. Madrian. 1999. "Health, Health Insurance, and the Labor Market." Eberstadt, N. 2005. "Health and Demography: The In Handbook of Labor Economics, vol. 3, edited by Achilles' Heel of Post-Socialist Development in Orley Ashenfelter and David Card (3309-3416). Europe." Presentation at "Advancing Economic Amsterdam: Elsevier Science. Growth: Investing in Health," Chatham House, London, June 22-23. Cutler, D.M. 2004. Your Money or Your Life. Strong Medicine for America's Health Care System. New The Economist. 2004. "Death Wish." September 30. York: Oxford University Press. http://www.uwec.edu/geography/Ivogeler/w111/ articles/russia-death-rates.htm. Cutler, D. M., and S. Kadiyala. 1999. "The Economics of Better Health: The Case of Cardiovascular Edlin, A. S. 2002. "Per-Mile Premiums for Auto Disease." Technical Report, National Bureau of Insurance." Working Paper E02-318, Department of Economic Research, Cambridge, MA, November. Economics, University of California, Berkeley, CA. http://repositories.cdlib.org/iber/econ/E02-318, Cutler, D., and E. Richardson. 1997. "Measuring the accessed 5 December 2003. Health of the U.S. Population" in Brookings Papers on Economic Activity, Microeconomics 29: 519-39. Elvik, R., and T. Vaa. In press. Handbook of Road Safety Measures. Amsterdam: Elsevier. Danishevski, K., and M. McKee. 2002. "Campaigners Fear That Russia's New Tobacco Law Won't Work," Eroshina, K., K. Danishevski, P. Wilkinson, and M. British Medical Journal 324(7334): 382. McKee. 2004. "Environmental and Social Factors as Determinants of Respiratory Dysfunction in Junior --­. 2005. "Reforming the Russian Health-Care Schoolchildren in Moscow," Journal of Public System," Lancet 365(9464): 1012-14. Health 26(2): 197-204. 134 EU (European Union). 2001. "Transport Policy for Frenk, J., J. L. Bobadilla, and M. Lopez Cervantes. 2010: Time to Decide." White Paper, Luxembourg. 1989. "Health Transition in Middle-Income http://europa.eu.int/comm/energy_transport/library/ Countries: New Challenges for Health Care," Health lb_texte_complet_en.pdf Policy and Planning 4(1): 29-39. European Foundation for the Improvement of Living Frid, E. 2005. "Health Care Costs in the Russian and Working Conditions. 1997. "Preventing Federation." Background assessment prepared for the Absenteeism at the Workplace." Luxembourg. World Bank, Moscow, March. http://www.eurofound.eu.int/publications/files/EF971 Gerasimenko, N. F., and A. K. Demine. 2001. 5EN.pdf. "Tobacco Policy and Politics in Russia." Russian European Health for All Database (HFA-DB). World Public Health Association, Moscow. Health Organization. Gilmore, A. B., and M. McKee. 2004a. "Tobacco and http://www.who.dk/hfadb. Transition: An Overview of Industry Investments, European Transport Safety Council. 1993. "Reducing Impact and Influence in the Former Soviet Union," Traffic Injuries through Vehicle Safety Improvements: Tobacco Control 13(2): 136-42. The Role of Car Design." Review, European --­.2004b. "Moving East: How the Transnational Transport Safety Council, Brussels. Tobacco Industry Gained Entry to the Emerging --­. 1995. "Reducing Injuries from Excess and Markets of the Former Soviet Union-Part II: An Inappropriate Speed." Report, Working Party on Overview of Priorities and Tactics Used to Establish a Road Infrastructure, European Transport Safety Manufacturing Presence," Tobacco Control 13(2): Council, Brussels. 151-60. --­. 1996. "Seat-Belts and Child Restraints: Increasing Gilmore, A., J. Pomerleau, M. McKee, R. Rose, C. W. Use and Optimizing Performance."European Haerpfer, D. Rotman, and S. Tumanov. 2004. Transport Safety Council, Brussels. "Prevalence of Smoking in 8 Countries of the Former --­. 1998. "Forgiving Roadsides." Briefing, European Soviet Union: Results from the Living Conditions, Transport Safety Council, Brussels. Lifestyles and Health Study," American Journal of http://www.etsc.be/documents/bri_road5.pdf Public Health 94(12): 2177-87. (checked August 12, 2005). Glasunov, I. S., V. Grabauskas , W. W. Holland, and F. Ezzati, M., A. D. Lopez, A. Rodgers, S. Vander Hoorn, H. Epstein. 1983. "An Integrated Programme for the C. J. Murray; and the Comparative Risk Assessment Prevention and Control of NCDs: A Kaunas Report Collaborating Group. 2002. "Selected Major Risk of a Meeting, Nov. 16-20, 1981." Journal of Factors and Global and Regional Burden of Disease," Chronic Disease 36(5): 419-26. Lancet 360(9343): 1347-60. Goskomstat (Russian State Statistical Bureau). 2000. Feachem, R. G. A., T. Kjellstrom, C. J. L. Murray, M. Regions of Russia. Report. Moscow. Over, and M. A. Phillips. 1992. The Health of --­. 2002. Demographic Yearbook of Russia: Adults in the Developing World. New York: Oxford Collection of Articles. Report. Moscow. University Press for the World Bank. --­. 2003. Russia in Figures. Report. Moscow. Finch, D. J., et al. 1994. "Speed, Speed Limits and Gribble, J. N., and S. H. Preston, eds. 1993. The Accidents." Project Report 58, Transport Research Epidemiological Transition: Policy and Planning Laboratory, Ltd., Crowthorne, Berkshire, United Implications for Developing Countries. Committee Kingdom. on Population, National Research Council, National Fontaine, K. R., D. T. Redden, C. Wang, A. O. Academy Press, Washington, DC. Westfall, and D. B. Allison. 2003. "Years of Life Lost http://www.nap.edu/books/0309048397/html/. Due to Obesity," Journal of the American Medical Guindon, G.E., S. Tobin, and D. Yach. 2002. "Trends Association 289: 187-93. and Affordability of Cigarette Prices," Tobacco Foxcroft, D. R., D. Ireland, G. Lowe, and R. Breen. Control 11: 25-43. www.tobaccocontrol.com. 2002. "Primary Prevention for Alcohol Misuse in Gyarfas, I. 1992. "Review of Community Intervention Young People," The Cochrane Database of Studies on Cardiovascular Risk Factors," Clin Exp Systematic Reviews. Issue 3. Art. No. CD003024. Hypertens A. 14(1-2): 223-37. DOI:10.1002/14651858.CD003024. 135 Hardy GE Jr. The burden of chronic disease: the future ILO (International Labor Organization). 2005. "Global is prevention. Introduction to Dr. James Marks' pres- Estimates of Fatal Work Related Diseases and entation, The Burden of Chronic Disease and the Occupational Accidents, World Bank Regions." Future of Public Health. Prev Chronic Dis [serial Geneva. online] 2004 April [date cited]. Available from URL: http://www.ilo.org/public/english/protection/ http://www.cdc.gov/pcd/issues/2004/apr/ safework/accidis/globesti.pdf 04_0006.htm. IMF (International Monetary Fund). 2004. World Hausman, J. A. 1978. "Specification Tests in Economic Outlook: The Global Demographic Econometrics," Econometrica 46: 1251-71. Transition. Washington, DC. Hausman, J. A., and W. E. Taylor. 1981. "Panel Data Institute for Alternative Futures. 2003. "Preliminary and Unobservable Individual Effects," Econometrica Summary Survey on Key Levers of Change to 49: 1377-98. Prevent and Control Chronic Disease." Preliminary Haveman, R, M. Stone, and B. Wolfe. 1994. "Market survey prepared for WHO Noncommunicable Work, Wages and Men's Health," Journal of Health Disease Strategy Development and Oxford Vision Economics 13: 163-82. 2020. http://www.oxfordvision2020.org/getMedia.asp?mb_ Haveman, R., B. Wolfe, and F. M. Huang. 1989. GUID=731BFF11-958E-4283-9D3C- "Disability Status as an Unobservable: Estimates from 9D5BF2FC05D3.pdf (accessed July 27, 2005). a Structural Model." Working Paper 2831, National Bureau of Economic Research, Cambridge, MA. Islam, N. 1995. "Growth Empirics: A Panel Data Approach," Quarterly Journal of Economic 110(4), Hawkes, C. 2004. "Food: A Determinant of Nutrition 1127-70. and Health in the ECA Region." Background paper prepared for WHO European Office, Geneva. Ivaschenko, O. 2003. "Mortality in Russian Regions: Do Poverty and Low Public Health Spending Kill? Heckman, J., H. Ichimura, and P. Todd. 1997. Pre-publication draft, Washington, DC, The World "Matching as an Econometric Evaluation Estimator: Bank. Evidence from Evaluating a Job Training Programme," Review of Economic Studies 64: 605- Jamison, D., L. Lau, and J. Wang. 2004. "Health's 54. Contribution to Economic Growth in an Environment of Partially Endogenous Technical Heleniak, T. 2005. Russia's Demographic Decline Progress." Disease Control Priorities Project Working Continues. Population Reference Bureau. Paper 10, Fogarty International Centre, National http://www.prb.org (accessed June 06, 2005). Institutes of Health. Bethesda, MD. Hellermann, J. P., T. Y. Goraya, S. J. Jacobsen, S. A. www.fic.nih.gov/dcpp. Weston, G. S. Reeder, B. J. Gersh, M. M. Redfield, Jiménez-Martín, S., J. M. Labeaga, and M. Martínez. R. J. Rodeheffer, B. P. Yawn, and V. L. Roger. 1997. 1999. "Health Status and Retirement Decisions for "Incidence of Heart Failure after Myocardial Older European Couples." Paper, TMR Programme, Infarction: Is It Changing over Time?" Archives of European Commission, Brussels. Internal Medicine 157: 2413-46. http://www.ceps.lu/iriss/documents/irisswp1.pdf. Holder, H. 1994. "Mass Communication as an Kalemli-Ozcan, S., H. E. Ryder, and D. N. Weil. 2000. Essential Aspect of Community Prevention to "Mortality Decline, Human Capital Investment, and Reduce Alcohol in Traffic Crashes," Alcohol, Drugs Economic Growth," Journal of Development and Driving 10(3-4): 295-307. Economics 62: 1-23. House of Commons Health Committee. 2004. Obesity. Kearney, P. M., M. Whelton, K. Reynolds, P. Muntner, Third Report of Session 2003-04, Volume 1, Report P. K. Whelton, and J. He. 2005. "Global Burden of and Formal Minutes, London, Great Britain, May 10. Hypertension: Analysis of Worldwide Data," Lancet http://www.publications.parliament.uk/pa/cm200304 365: 217-23. /cmselect/cmhealth/23/23.pdf 136 Kelley E, E. Moy, B. Kosiak, D. McNeill, C. Zhan, D. Ladnaia, N., V. Pokrovsky, and C. Rühl. 2003. "The Stryer, and C. Clancy. 2004. "Prevention Health Economic Consequences of HIV in Russia: An Care Quality in America: Findings from the First Interactive Simulation Approach." Moscow, The National Healthcare Quality and Disparities World Bank. Reports," Preventing Chronic Disease: Public Lang, T., and M. Heasman. 2004. "Diet and Nutrition Health Research, Practice, and Policy 1(3): Epub Policy: A Clash of Ideas or Investment?" 2004 Jun 15. Development, 47(2): 64-74. www.cdc.gov/pcd/issues/204/jul/pdf/04_0031.pdf. http://www.oxfordvision2020.org/getMedia.asp?mb_ Kingkade, W. W., and E. E. Arriagada. 1997. "Mortality GUID=11845FE1-D5E8-4B01-B5C3- in the New Independent States: Patterns and 5FC02F8A661D.pdf accessed July 27, 2005. Impacts." In Premature Death in the New Lechner, M., and R. Vasquez-Alvarez. 2004. "The Effect Independent States, eds. J. L. Bobadilla, C. A. of Disability on Labor Market Outcomes in Costello, and F. Mitchell, 156-83. Washington, DC: Germany: Evidence from Matching," Center for National Academy Press. Economic Policy Research (CEPR) Discussion Paper Kirchberger, M., and M. Fiorin. 2004. "Is There an Series 4223. Economic Rationale for Public Policy against Alcohol Leeder, S., S. Raymond, H. Greenberg, H. Liu, and K. and Tobacco Consumption in Central and Eastern Esson. 2004. A Race against Time: The Challenge Europe?" Degree thesis, Free University of Bolzano. of Cardiovascular Disease in Developing Economies. Klein, R. 2004. "Britain's National Health Service New York, NY: Columbia University. Revisited," New England Journal of Medicine Leon, D. 2005. Personal communication. London 350(9): 937-42. School of Tropical Medicine, April. Knai, C., M. McKee, and M. Bobak. 2004. "Evidence Leon, D., Andreev, E., Kiryanov, N., McKee, M., Base on How to Improve Health in Low-Resource Suburova, V., Shkolnikov, V., and Tomkins, S. 2005. Settings with Special Attention to ECA: CVD "Izhevsk Family Study." Interim Report. London Prevention, Management, Rehabilitation." Draft School of Hygiene and Tropical Medicine. report, World Health Organization, Geneva. Leon, D., L. Chenet, V. M. Shkolnikov, S. Zakharov, J. Laatikainen, T. 2000. "Cardiovascular Risk in the Shapiro, G. Rakhmanova, S. Vassin, and M. McKee. Republic of Karelia, Russia: Comparison of Major 1997. "Huge Variation in Russian Mortality Rates Risk Factors with North Karelia, Finland." Thesis, 1984-1994: Artefact, Alcohol, or What?" Lancet National Public Health Institute, Helsinki. 350: 383-88. Laatikainen, T., H. Alho, E. Vartianinen, P. Jousilahti, P. Leon, D., and M. Mckee. 2005. Personal communica- Sillanaukee, and P. Puska. 2002a. "Self-reported tion. London School of Hygiene and Tropical Alcohol Consumption and Association to Medicine, April. Carbohydrate-Deficient Transferring and Gamma- Glutamyltransferase in a Random Sample of the Levine, R., and D. Renelt. 1992. "A Sensitivity Analysis General Population in the Republic of Karelia, Russia of Cross-Country Growth Regressions," American and in North Karelia, Finland." Alcohol and Economic Review 82: 942-63. Alcoholism 37(3): 282-88. Levintova, M., and T. Novotny. 2004. Laatikainen, T., L. Delong, S. Pokusajeva, M. Uhanov, "Noncommunicable Disease Mortality in the Russian E. Vartiainen, and P. Puska. 2002b. "Changes in Federation: From Legislation to Policy," Bulletin of Cardiovascular Risk Factors and Health Behaviours the World Health Organization 82: 875-80. from 1992 to 1997 in the Republic of Karelia, Litman, T. 2000. "Distance-Based Vehicle Insurance: Russia," European Journal of Public Health 12(1): Feasibility, Costs and Benefits." Comprehensive 37-43. Technical Report, Victoria Transport Policy Laatikainen, T., V. Korpelainen, T. Vlasoff, M. Uhanov, Institute, Victoria, BC. and P. Puska. 2005. The Pitkäranta Project: A http://www.vtpi.org/dbvi_com.pdf, accessed 5 Successful Pilot Programme in Non-communicable December 2003. Disease Prevention in the Republic of Karelia, Russia. Draft Report by the National Public Health Institute, Finland. 137 Lock, K., E. Andreev, V. M. Shkolnikov, and M. McMichael, A.J., M. McKee, V. Shkolnikov, and V. McKee. 2002. "What Targets for International Valkonen. 2004. "Mortality Trends and Setbacks: Development Policies Are Appropriate for Improving Global Convergence or Divergence?" Lancet 363: Health in Russia?" Health Policy & Planning 17: 1155-59. 257-63. Mesle, F. 2002. "Mortality in Central and Eastern Macaulay, J., and R. McInerney. 2002. "Evaluation of Europe: Long-Term Trends and Recent Upturn." the Proposed Actions Emanating from Road Safety Demographic Research Special Collection 2 Article Audits." Austroads Publication AP-R209/02, Sydney. 3. Rostock, Germany: Max Planck Institute for Magnus, P., and R. Beaglehole. 2001. "The Real Demographic Research. Contribution of the Major Risk Factors to the Mesle, F., V. Shkolnikov, V. Hertrich, and J. Vallin. Coronary Epidemics," Archives of Internal Medicine 1996. Tedances recentes de la mortalite par cause en 161: 2657-60. Russie, 1965-1994. Paris, Institut National d'Études Marmot, M. 2002. "The Influence of Income on Demographiques. Health: Views of an Epidemiologist." Health Affairs Miller, T. R. 2000. "Variations between Countries in 21(2): 31-46. Values of Statistical Life," Journal of Transport Marquez, P., and M. Suhrcke. 2005. "Combating non- Economics and Policy 34(2): 169-88. communicable diseases." British Medical Journal Mintz, P. H. 1999 "Managing Acute, Chronic, and 331: p. 174-174. Catastrophic Health Care Costs. Experience and Maslennikova, G. Ya, S. A. Martynchick, S. A. Policy Issues in the U.S. Context of Managed Care Shalnova, A. D. Deyev, and R. G. Oganov. 2005. and Comparative Analysis of the Chilean Regulatory Unpublished paper. National Research Centre for Framework."A report prepared for the World Bank, Preventive Medicine, MOHSD, Moscow. Washington, DC, June. McKee, M. 2004. "Learning from Tobacco to Address Mitchell, J., and R. Burkhauser. 1990. "Disentangling Diet and Nutrition more effectively." PPL. Report of the Effect of Arthritis on Earnings: A Simultaneous the Oxford Vision 2020 Tobacco Working Group. Estimate of Wage Rates and Hours Worked," http://www.oxfordvision2020.org/getMedia.asp?mb_ Applied Economics Letters 22: 1291-310. GUID=4B5356F7-1E61-4380-AACB- MOH (Ministry of Health, Russia). 1994. Towards a 0EC1B3CE3E31.ppt Healthy Russia: Policy for Health Promotion and --­. 2005. Personal communication. London School of Disease Prevention; Focus on Major Tropical Medicine, March. Noncommunicable Diseases. State Research Center for Preventive Medicine. Moscow. McKee, M, M. Bobak, R. Rose, V. Shkolnikov, L. Chenet, and D. Leon. 1998. "Patterns of Smoking in MOPH (Ministry of Public Health, Russia). 1997. Russia, Tobacco Control 7: 22-26. "Towards a Healthy Russia: Policies and Strategies for the Prevention of Cardivascular and Other McKee, M, D. A. Leon, S. Tomkins, V. M. Shkolnikov, Noncommunicable Diseases within the Context of and E. Andreev. 2005. "INTERHEALTH. Public Health Reform in Russia." Moscow. Correspondence," The Lancet 365: 117-18. --­. 1999. "Monitoring of Risk Factors for Non-com- McKee, M., V, Shkolnikov, and D. A. Leon. 2001. municable Diseases, Mortality and Other Indicators "Alcohol Is Implicated in the Fluctuations in of the CINDI Program Development." CINDI- Cardiovascular Disease in Russia since the 1980s," Russia Final Report. Moscow. Annals of Epidemiology 11(1): 1-6. Mossey, J. M., and E. Shapiro. 1982. "Self-rated Health: McKeown, T. 1985. "Looking at Disease in the Light of A Predictor of Mortality among the Elderly," Human Development," World Health Forum 6: 70- American Journal of Public Health.72: 800-08. 75. Mullahy, J. 1991. "Gender Differences in Labor Market McKinlay, J. 1993. "The Promotion of Health through Effects of Alcoholism," American Economic Review: Planned Socio-Political Change: Challenges for Papers and Proceedings 81: 161-65. Research and Policy," Social Science and Medicine 36: 109-17. 138 Murray, C. 1993. "Quantifying the Burden of Disease: North Karelia Project. 2005. Power Point Presentation. The Technical Basis for Disability Adjusted Life National Public Health Institute, Finland. Years." Health Transition Working Paper Series Notzon, F. C., Y. M. Komarov, S. P. Ermakov, C. T. 93.03. Cambridge, MA: Harvard University. Sempos, J. S. Marks, E. V. Sempos. 1998. "Causes of Murray, C.J.L., and J. L. Bobadilla. 1997. Declining Life Expectancy in Russia." Journal of the "Epidemiological Transitions in the Formerly American Medical Association 279: 793-800. Socialist Economies: Divergent Patterns of Mortality OECD (Organisation for Economic Co-operation and and Cuases of Death." In Premature Death in the Development). 2003. Health Data 2003. New Independent States, eds. J. L. Bobadilla, C. A. http://caliban.sourceoecd.com/vl=11322400/cl=42/ Costello, and F. Mitchell, 184-219. Washington, DC: nw=1/rpsv/~9294/v1n1/s1/p1. National Academy Press. Oganov, R. G. 2005. "CVD Mortality at the End of the Murray, C., and A. Lopez. 1993. "Quantifying the 20th Century: Russian Paradox." Presentation at the Burden of Disease: Data, Methods, and Results." National Research Center for Preventive Medicine. Health Transition Working Paper Series 93.05. Moscow. Cambridge, MA: Harvard University. Oganov, R. G., and G. Y. Maslennikova. 1999. Musgrove, P. 1996. "Public and Private Roles in Health: "Cardiovascular Disease Mortality in the Russian Theory and Financing Patterns." World Bank Federation during the Second Half of the 20th Discussion Paper WDP 339, Washington, DC: The Century." CVD Prevention 2: 37-43. World Bank. --­. 2005a. "Key Role of Caridovascular and other Naik, G. 2003. "Finns Use Group Effort to Come Up Non-Communicble Disease Prevention in Improving with Fix for Heart Disease." Wall Street Journal the Demographic Sitaution in Russia." Unpublished (New York). January 14. paper. National Research Centre for Preventive NHS (National Health Service). 2004. Choosing Medicine, MOHSD. Moscow. Health: Making Healthy Choices Easier. Executive --­. 2005b. "Smoking Health Effects: A Place of Russia Summary. Department of Health, Her Majesty's in Europe." Presentation at the State Research Center Government, United Kingdom. for Preventive Medicine, Moscow. Nickel, S. 1981. "Biases in Dynamic Models with Fixed Olshansky, S. J., D. J. Passaro, R. C. Hershow, J. Effects," Econometrica 49: 1117-26. Layden, B. A. Carnes, J. Brody, L. Hayflick, et al. NOBUS (National Survey of Household Welfare and 2005. "A Potential Decline in Life Expectancy in the Program Participation). 2003. Federal State Statistics United States in the 21st Century," New England Service. Moscow. Journal of Medicine 352(11): 1138-45. Nolte, E., McKee, M, and Gilmore, A. 2004. Omran, A. R. 1971. "The Epidemiologic Transition: "Morbidity and Mortality in Transition Countries in A Theory of the Epidemiology of Population the European Context. Background Paper for the Change," The Milbank Memorial Fund Quarterly Thematic Session on Morbidity, Mortality and 49: 509-38. Reproductive Health at the Euroepan Population --­. 1999. The Epidemiologic Transition Theory Forum." Geneva, January 12-14. Revisited Thirty Years Later. Washington, DC: --­. 2005. "Morbidity and Mortality in Transition George Washington University. Countries in the European Context." In The New Owen, L. 2000. "Impact of a Telephone Helpline for Demographic Regime: Population Challenges and Smokers Who Called during a Mass Media Policy Responses, eds. M. Macura, A. MacDonald, Campaign," Tobacco Control 9: 148-54. and W. Haug. New York and Geneva: United Nations. PAHO (Pan American Health Organization). 2003. "Report from 37th Session of the Subcommittee on Nordhaus, W. 2003. "The Health of Nations: The Planning and Programming of the Executive Contribution of Improved Health to Living Committee." Washington, DC, March 26-28. Standards." In The Measurement of Economic and Social Performance, ed. M. Moss. New York: Palosuo, H. 2003. "Health and Well-Being in Moscow Columbia University Press for the National Bureau of and Helsinki." STAKES Research Report 129 Economic Research. Helsinki, Finland. 139 Paoli, P., and Merllié, D. 2003. "Third European Survey Posten, W. S. C., and J. P. Foreyt. 1999. "Obesity Is an on Working Conditions 2000." European Environmental Issue," Atherosclerosis 146: 201-09. Foundation for the Improvement of Living and Pramming, S., and N. Nordisk. 2004. "The Journey So Working Conditions. Far: A Personal Perspective." Presented at Oxford www.eurofound.eu.int. Vision 2020 Summit, Saïd Business School, Oxford, Pauly, M., S. Nicholson, J. Xu, D. Polsky, P. M. September. Danzon, J. F. Murray, and M. Berger. 2002. "A http://www.oxfordvision2020.org/getMedia.asp?mb_ General Model of the Impact of Absenteeism on GUID=1CEE172D-DDB3-4C2C-B29A- Employers and Employees," Health Economics 11: 10E9BF9DD9A5.ppt. 221-31. Puska, P. 1995. "Health Promotion Challenges for Peden, M, R. Scurfield, D. Sleet, D. Mohan, A. A. Countries of the Former Soviet Union: Results from Hyder, E. Jarawan, and C. Mathers. 2004. World Collaboration between Estonia, Russian Karelia, and Report on Road Traffic Injury Prevention. ISBN: 92 Finland," Health Promotion International 10(3). 4 156260 9. Geneva: The World Health --­. 1996. "Development of Public Policy on the Organization. Prevention and Control of Elevated Blood Pekka, P., P. Pirjo, and U. Ulla. 2002. "Influencing Cholesterol," Cardiovascular Risk Factors 6(4): 203- Public Nutrition for Noncommunicable Disease 10. Prevention: From Community Intervention to --­. 2000. "Nutrition and Mortality: The Finnish National Programme-Experiences from Finland," Experience," Acta Cardiol 55(4): 213-20. Public Health Nutrition 5(1A): 245-51. --­. 2002. "Successful Prevention of NCDs: 25 Year Pelkowski, J. M., and M. C. Berger. 2004. "The Impact Experience with North Karelia Project in Finland," of Health on Employment, Wages, and Hours Public Health Medicine 4(1): 5-7. Worked over the Life Cycle," Quarterly Review of Economics and Finance 44: 102-21. Puska, P, J. Tuomilehto, A. Nissinen, and E. Vartiainen (eds.). 1995. The North Karelia Project: 20 Year Peto, R., A. D. Lopez, J. Boreham, M. Thun, and C. Results and Experiences. The National Public Health Heath. 1994. Mortality from Smoking in Developed Institute. Helsinki, Finland. Countries 1950-2000. Oxford: Oxford University Press. Puska, P., E. Vartiainen, J. Tuomilehto, V. Salomaa, and A. Nissinen. 1998. "Changes in Premature Deaths in Pietinen, P., M. Lahti-Koski, E. Vartiainen, and P. Finland: Successful Long-Term Prevention of Puska. 2001. "Nutrition and Cardiovascular Disease Cardiovascular Diseases," Bulletin of the World in Finland since the Early 1970s: A Success Story," Health Organization 76(4): 419-25. J Nutr Health Aging 5(3): 150-54. Putin, V. 2004. "Annual Address to the Federal Plavinski, S. L., S. I. Plavinskaya, and A. N. Klimov. Assembly of the Russian Federation." Federal News 2003. "Social Factors and Increases in Mortality in Service. May 26. Russia in the 1990s: Prospective Cohort Study." British Medical Journal 326(7401): 1240-42. Raistrick, D., R. Hodgson, and B. Ritson (eds.). 1999. Tackling Alcohol Together: The Evidence Base for a Plavinski, S. L., S. I. Plavinskaya, V. Richter, et al. UK Alcohol Policy. Society for the Study of 1999. "The Total and HDL-Cholesterol Levels in Addiction. London: Free Association Books. Populations of St. Petersburg (Russia) and Leipzig (Germany)," Nutr Metb Cardiovasc Dis 9: 184-91. Raw, M., A. McNeill, and R. West. 1999. "Smoking Cessation: Evidence Based Recommendations for the Poikolainen K. 1999. "Effectiveness of Brief Healthcare System," British Medical Journal Interventions to Reduce Alcohol Intake in Primary 318(7177): 182-85. Health Care Populations: A Meta Analysis," Preventive Medicine 28: 503-09. Rechel B., L. Shapo, and M. McKee. 2004. Millennium Development Goals for health in Pomerleau, J., A. Gilmore, M. McKee, R. Rose, and C. Europe and Central Asia: relevance and policy W. Haerpfer. 2004. "Determinants of Smoking in implications. Washington, D.C.: World Bank. Eight Countries of the Former Soviet Union: Results from the Living Conditions, Lifestyles and Health Reddy, K. S. 2004. "Cardiovascular Disease in Non- Study," Addiction. 99(12): 1577-85. Western Countries," New England Journal of Medicine 350(24): 2438-40. 140 Rehm, J., and G. Gmel. 2002. "Average Volume of Sala-I-Martin, X., G. Doppelhofer, and R. I. Miller. Alcohol Consumption: Patterns of Drinking and 2004. "Determinants of Long-Term Growth: A Mortality among Young Europeans in 1999," Bayesian Averaging of Classical Estimates (BACE) Addiction, Letters to the Editor 97(1): 105-09. Approach," American Economic Review 94(4): Rivera, B., and L. Currais. 1999. "Economic Growth 813-35. and Health: Direct Impact or Reverse Causation?" Sammartino, F. J. 1987. "The Effect of Health on Applied Economics Letters 6: 761-64. Retirement," Social Security Bulletin 50(2). RLMS (Russian Longitudinal Monitoring Survey). Schultz, T. P., and A. Tansel. 1995. "Measurement of 1998, 2004, 2005. Returns to Adult Health: Morbidity Effects on Wage Rodríguez, D. Y., F. J. Fernandez, and H. A. Velasquez. Rates in Cote d'Ivoire and Ghana," Living Standards 2003. "Road Traffic Injuries in Colombia," Injury Measurement Study Working Paper, N. 95, The Control and Safety Promotion 10: 29-35. World Bank, Washington, DC. Roland, M. 2004. "Linking Physicians' Pay to the --­. 1996. "Wage and Labor Supply Effects of Illness in Quality of Care: A Major Experiment in the United Côte d'Ivoire and Ghana: Instrumental Variable Kingdom," New England Journal of Medicine 351: Estimates for Days Disabled," Journal of 14. Development Economics 53: 251-86. Rose, G.. 1992. The Strategy of Preventive Medicine. Shafey, O., S. Dolwick, and G. E. Guidon, eds. 2003. Oxford, England: Oxford University Press. Tobacco Control Country Profiles, 2nd ed. 12th World Conference on Tobacco and Health, Helsinki, Rosembaum, P., and D. Rubin. 1983. "The Central Finland. Atlanta, GA: WHO Tobacco Free Initiative, Role of the Propensity Score in Observational Studies Americana Cancer Society, and International Union for Causal Effects," Biometrika 70: 41-55. against Cancer. Ross, H. 2004. "Russia (Moscow) 1999 Global Youth Shishkina, S., G. Besstremyannaya, M. Krasilnikova, Tobacco Survey: Economic Aspects." HNP et.al. 2004. "Russian Health Care: Payments in Discussion Paper: Economics of Tobacco Control Cash." Independent Institute of Social Policy, Paper 23, The World Bank, Washington, DC. Moscow. Sachs, J. 2001. Macroeconomics and Health: Investing Shkolnikov, V., E. M. Andreev, D. A. Leon, M. McKee, in Health for Economic Development. Geneva: F. Meslé, and J. Vallin. 2004. "Mortality Reversal in World Health Organization. Russia: The Story So Far," Hygeia Internationalis 4: Sachs, J., and A. Warner. 1995. "Economic Reform and 29-80. the Process of Global Integration," Brookings Papers Shkolnikov V., V. V. Chervyakov, M. McKee, and D.A. on Economic Activity. 1-118. Leon. 2004. "Russian Mortality beyond Vital Saffer, H. 1995a. "Alcohol Advertising and Alcohol Statistics: Effects of Social Status and Behaviors on Consumption: Econometric Studies." In The Effects Deaths from Circulatory Disease and External of the Mass Media on the Use and Abuse of Alcohol, Causes-A Case-Control Study of Men Aged 20-55 ed. S. E. Martin. Bethesda, MD: National Institute Years in Udmurtia, 1998-99." Max-Planck- on Alcohol Abuse and Alcoholism. Gesellschaft, Rostock, Germany. --­. 1995b. The Control of Tobacco Advertising and Shkolnikov, V., and D. A. Leon. 2005. "World Promotion. Background paper. Publisher information Mortality 1950-2000: Divergence Replaces unavailable. Convergence from the Late 1980s." Bulletin of the Sakevich, V. 2005. "Maternal Mortality Worldwide." World Health Organization 83: 202-09. Demoskop Weekly/Population and Society Bulletin, Shkolnikov, V., M. McKee, and D. A. Leon. 2001. Electronic Version. The Center for Demography and "Changes in Life Expectancy in Russia in the Mid- Human Ecology, Institute of Economic Forecasting, 1990s," Lancet 357: 917-21. Russian Academy of Sciences. N.199-200, 18 April-1 Shkolnikov,V., M. McKee, D.A. Leon, and L. Chenet. May 2005. 1999. "Why Is the Death Rate from Lung Cancer http://www.demoscope.ru/weekly/2005/0199/ Falling in the Russian Federation?" European reprod01.php Journal of Epidemiology 15: 203-06. 141 Shkolnikov V., F. Mesle, and J. Vallin. 1997. "Recent Suslina, Z.A. 2005. "Urgent Measures to Reduce Trends in Life Expectancy and Causes of Death in Cerebrovascular Disease and Coronary Heart Disease Russia, 1970-1993." In Premature Death in the Morbidity and Mortality among Russians of Working New Independent States, eds. J. L. Bobadilla, C. A. Age." Unpublished paper. RMAS Neurology Costello, and F. Mitchell, 34-65. Washington, DC: Research Institute, MOHSD. Moscow. National Academy Press. Thomas, D. 2001. "Health, Nutrition and Economic Shkolnikov, V., and A. Nemtsov. 1997. "The Anti-alco- Prosperity: A Microeconomic Perspective." CMH hol Campaign and Variations in Russian Mortality." Working Paper WG1:7, Commission on In Premature Death in the New Independent States, Macroeconomics and Health, World Health ed. J. L. Bobadilla, C. A. Costello, and F. Mitchell, Organization, Geneva. 239-61. Washington, DC: National Academy Press. Tobacco Control Country Profiles. Siddiqui, S. 1997. "The Impact of Health on http://www.globalink.org/tccp/. Retirement Behaviour: Empirical Evidence from Tragakes, E., and S. Lessof. 2003. Health Care Systems West Germany," Econometrics and Health in Transition: Russian Federation. European Economics 6: 425-38. Observatory on Health Care Systems, Copenhagen. Specter, M. 2004. "The Devastation." The New Yorker, Transit New Zealand. 1992. "Accident October 11, 58-69. Countermeasures: Literature Review." Research --­. 1996. "Measuring Child Work and Residence Report Number 10, Wellington, New Zealand. Adjustments to Parent's Long-Term Care Needs," Available at http://www.Itsa.govt.nz/about/con- Gerontologist 36: 76-87. tracts.html. State Federal Statistics Service (Rosstat). 2004. Russian Trans-Milenio, S. A. 2001. "A high capacity/low cost Yearbook of Statistics. Moscow. bus rapid transit system developed for Bogotá, Stern, S. 1989. "Measuring the Effect of Disability on Colombia." Bogotá. See also: E. El Sandoval and D. Labor Force Participation," Journal of Human Hildalgo. Not dated. "TransMilenio: A High Resources 24(3): 361-95. Capacity - Low Cost Bus Rapid Transit System Developed for Bogotá, Colombia." Conference --­. 1996. "Semiparametric Estimates of the Supply Proceeding Paper, Urban Public Transportation and Demand Effects of Disability on Labor Force System: Ensuring Sustainability Through Mass Participation," Journal of Econometrics 71: 49*70. Transit, American Society of Civil Engineers, Strauss, J., P. Gertler, O. Rahman, and K. Fox. 1992. Reston, VA. "Gender and Life-Cycle Differentials in the Patterns Treml, V. G. 1997. "Soviet and Russian Statistics on and Determinants of Adult Health," Journal of Alcohol Consumption and Abuse." In Premature Human Resources 28(4). Also available at Death in the New Independent States, eds. J. L. http://www.rand.org/cgi-bin/Abstracts/ordi/getabby- Bobadilla, C. A. Costello, and F. Mitchell, 220-38. doc.pl?doc=RP-301. Washington, DC: National Academy Press. Strauss, J., and D. Thomas. 1998. "Health, Nutrition Trognon, A. 1978. "Miscellaneous Asymptotic and Economic Development," Journal of Economic Properties of Ordinary Least Squares and Maximum Literature 36: 766-77. Likelihood Estimators in Dynamic Error Suhrcke, M., M. McKee, R. Sauto Arce, S. Tsolova, and Components Models," Annales d l'INSEE 30/31: J. Mortensen. 2005a. "The Contribution of Health 631-57. to the Economy in the European Union." Report Tunstall-Pedoe, H., K. Kuulasmaa, M. Mahonen, H. commissioned by the European Commission. Tolonen, and E. Ruokokoski. 1999. "Contribution of Brussels, Belgium. Trends in Survival and Coronary Event Rates to Suhrcke M, Rocco L, McKee M, Urban D, Mazzucco S, Changes in Coronary Heart Disease Mortality: 10- and A. Steinherr. 2005b. "Economic Consequences Year Results from 37 WHO MONICA Project of Non-communicable Diseases and Injuries in the Populations," Lancet 353: 1547-57. Russian Federation." WHO European Office for Investment for Health and Development, Venice. Background paper prepared for this study. 142 Twigg, J. 2004. "National Security Implications of Vallin, J., and F. Mesle. 2001. Trends in Mortality in Russia's Health and Demographic Crisis," PONARS Europe since 1950: Age, Sex and Cause-Specific Policy Memo 360: 1-5. Mortality, in Trends in Mortality and Differential UN (United Nations). Population Database. World Mortality. Strasbourg: Council of Europe Publishing. Population Prospects. U. N. Population Division. Vartiainen, E., P. Jousilahti, G. Alfthan, J. Sundvall, P. http://esa.un.org.unpp. Pietinen, and P. Puska. 2000. "Cardiovascular Risk --­. 2002. "Aging Index 1990-2025." World Factor Changes in Finland 1972-1997," International Population Prospects: The 2002 Revision. United Journal of Epidemiology 29: 49-56. Nations, New York, NY. Vartiainen, E., P. Puska, J. Pekkanen, J. Toumilehto, and --­. 2003. World Population Prospects: The 2002 P. Jousilahti. 1994. "Changes in Risk Factors Explain Revision. United Nations, New York, NY. Changes in Mortality from Ischemic Heart Disease in Finland," British Medical Journal 309: 23-27. --­. 2004. World Population Prospects: The 2003 Revision. Population Database. UN Population Vergnano, M, 2004. "Achieving Behavior Modification: Division. Potential Contributions from Communications http://esa.un.org.unpp. Practitioners." Oxford Vision 2020. http://www.oxha.org/archive. United Nations Development Program (UNDP). Human Development Report (various years). New Viscusi, W. K., and J. E. Aldy. 2003. "The Value of York: Oxford University Press for the UNDP. Statistical Life: A Critical Review of Market Estimates throughout the World." NBER Working UNICEF (United Nations Children's Fund). 2004. Paper 9487, National Bureau of Economic Research, "Economic Growth and Child Poverty in the Cambridge, MA. CEE/CIS and the Baltic States," Social Monitor 2004. Florence: UNICEF Innocenti Research Wagenaar, A. C. 1993. "Research Affects Public Policy: Centre. The Case of the Legal Drinking Age in the United States," Addiction 88 (supplement): 75-81. USA Today. 2005. "Social Security and You. As Lives Get Longer, Benefits Should Come Later," May 20. Walberg, P., M. McKee, V. Shkolnikov, L. Chenet, and D.A. Leon. 1998. "Economic Change, Crime, and U.S. CDC (Centers for Disease Control and Mortality Crisis in Russia: Regional Analysis," Prevention). Not dated. British Medical Journal 317: 312-18. http://www.cdc.gov/nccdphp/dnpa/obesity/faq.htm#ad ults, citing the National Health and Nutrition Wanless, D. 2004. "Securing Good Health for the Examination Survey 1999-2000 (cite visited June Whole Population." Final Report, Her Majesty's 2005), Atlanta, GA. Treasury, London. --­. 2004. Physical Activity Fact Sheet. Division of West, R., A. McNeill, and M. Raw. 2000. "Smoking Nutrition and Physical Activity. Atlanta, GA: Cessation Guidelines for Health Professionals: An http://www.cdc.gov/nccdphp/dnpa. Update," Thorax 55: 987-99. Usher, D. 1973. "An Imputation to the Measure of Whelton, P. K., J. He, L. J. Appel, J. A. Cutler, S. Economic Growth for Changes in Life Expectancy." Havas, T. A. Kotchen, E. J. Roccella, R. Stout, C. In The Measurement of Economic and Social Vallbona, M. C. Winston, and J. Karimbakas for the Performance, ed. M. Moss, 193-226. New York: National High Blood Pressure Education Program Columbia University Press for the National Bureau of Coordinating Committee. 2002. "Primary Prevention Economic Research. of Hypertension: Clinical and Public Health Advisory from the National High Blood Pressure U.S. HHS (U.S. Department of Health and Human Education Program," Journal of the American Services). 2000. "Reducing Tobacco Use: A Report of Medical Association 288: 1882-88. the Surgeon General." Office on Smoking and Health, Centers for Disease Control and Prevention, WHO (World Health Organization). European Health National Centers for Chronic Disease Prevention and for All Database (HFA-DB). Health Promotion, Rockville, MD. http://data.euro.who.int/hfadb/. 143 --­. WHO Mortality Database. --­. 2004a. "Diabetes Program Fact Sheet." World http://www3.who.int/whosis/menu.cfm?path=who- Health Organization, Geneva. sis,mort&language=english. Copenhagen: World http://www.who.int/diabetes/en/index.html. Health Organization Regional Office for Europe. --­. 2004b. "The CINDI Health Monitor: A Study of --­. ICD (International Statistical Classification of Feasibility of a Health Behavior Monitoring Survey Diseases and Related Health Problems). Online edi- across CINDI Countries." Data Book, WHO tion available at Regional office for Europe, Copenhagen. http://www.who.int/classifications/icd/en/. --­. 2004c. Atlas of Heart Disease and Stroke. Geneva: --­. 1996. "The Tobacco Epidemic: A Global Public WHO. Health Emergency." Tobacco Alert, Tobacco or --­. 2004d. Global Status Report: Alcohol Policy. Health Programme, Geneva. Department of Mental Health and Substance Abuse. --­. 2001a. "Assessment of National Capacity for Geneva: World Health Organization. Noncommunicable Disease Prevention and Control." --­. 2004e. "Diabetes Action Now: An Initiative of the The Report of a Global Survey, The World Health World Health Organization and the International Organization, Geneva. Diabetes Federation." World Health Organization, --­. 2001b. "Macroeconomics and Health: Investing in Geneva. health for Economic Development." Commission on http://www.who.int/diabetes/actionnow/en/DANbookl Macroeconomic and Health, World Health et.pdf (checked July 27, 2005). Organization, Geneva. --­-. 2005a. European Health Report 2005. --­. 2001c. The World Health Report 2001. Geneva. Copenhagen: World Health Organization Regional See annex: Healthy Life Expectancy, available at Office for Europe. http://www.who.int/whr/2002/en/whr2002_ http://www.euro.who.int/eprise/main/who/progs/ annex4.pdf. ehros/home. --­. 2002a. The European Health Report 2002. --­-. 2005b. Preventing Chronic Diseases. A Vital Copenhagen: World Health Organization Regional Investment. World Health Organization, Geneva. Office for Europe. Wild, S., G. Roglic, A. Green, R. Sicree, and H. King. --­. 2002b. The World Health Report 2002: Reducing 2004. "Global Prevalence of Diabetes Estimates for Risks to Health, Promoting Healthy Life. Geneva: the Year 2000 and Projections for 2030," Diabetes World Health Organization. Care 27: 1047-53. --­. 2002c. CINDI Highlights 2002, Number 8. Wines, M. 2000. "An Ailing Russia Lives a Tough Life Geneva: World Health Organization That's Getting Shorter." The New York (NY) Times, --­. 2002d. Investment for Health: A Discussion of the December 3. Role of Economic and Social Determinants. Geneva: World Bank. World Development Indicators. The World Health Organization --­. 1993. World Development Report 1993 Investing --­. 2003a The World Health Report 2003: Shaping in Health. New York: Oxford University Press for the Future. Geneva: World Health Organization. the World Bank. --­. 2003b. Atlas of Health in Europe. Geneva: World --­. 1999. Curbing the Epidemic: Governments and Health Organization. the Economics of Tobacco Control. Development in --­. 20003c. "MDG Proposal Report for Improving Practice. Washington, DC: The World Bank. Baseline Information." Moscow, World Health --­. 2003a. "Road Safety," At a Glance series published Organization. by Health Nutrition Population Unit. September, --­. 2003. "Consultation Document to Guide Washington, DC: The World Bank. Development of a WHO Global Strategy for Diet, ______. 2003b. Averting AIDS Crises in Eastern Physical Activity and Health." The World Health Europe and Central Asia. A Regional Support Organization, Geneva. Available at Strategy. Washington, DC: The World Bank. http://www.who.int//hpr/gs.consultation.document. ______. 2003c. Country Assistance Strategy (CAS) for shtml#Consultation%20Document. the Russian Federation. 144 --­. 2004a. Health in Europe and Central Asia: Yach, D., and C. Hawkes. 2004. "The WHO Long- Transition Retrospective and Business Plan. Term Strategy for Prevention and Control of Leading Washington DC: The World Bank. Chronic Diseases." Draft Report, The World Health --­. 2004b. "Millennium Development Goals for Organization, Geneva. Health in Europe and Central Asia, Relevance and Yach, D., C. Hawkes, C. L. Gould, and K. J. Hofman. Policy Implications." Working Paper 33, The World 2004. "The Global Burden of Chronic Diseases, Bank, Washington, DC. Overcoming Impediments to Prevention and --­. 2004c. "Road Safety Taskforce Operational Control." Journal of the American Medical Guidance for World Bank Staff," Transport Notes Association 291: 2616-22. TN-1, The World Bank, Washington DC. Yusuf, S., S. Hawken, S. Ounpuu, T. Dans, A. --­. 2004d. "Implementing the Recommendations of Avenzum, F. Lanas, M. McQueen, A. Budaj, P. Pais, the World Report on Road Traffic Injury J. Varigos, L. Lisheng, and other INTERHEART Prevention." Transport Notes TN-1. The World Study Investigators. 2004. "Effect of Potentially Bank., Washington, DC. Modifiable Risk Factors Associated with Myocardial Infarction in 52 Countries (the INTERHEART --­. 2005a. CAS Progress Report for Russian Study): Case-Control Study," Lancet 364: 937-52. Federation. Washington DC: The World Bank Zeidner R. 2004. "Fitness On the Job. Special to --­. 2005b. Global Estimates of Fatal Work Related Washington (DC) Post, August 17. Diseases and Occupational Accidents. Washington, DC: The World Bank. Zhirova, I.A., O. G. Froilova, T. M. Astakhova, and E. Ketting, E. 2004. "Abortion-Related Maternal --­. 2005c. Russian Federation Reducing Poverty Mortality in the Russian Federation," Studies in through Growth and Social Policy Reform. Family Planning 35(3). Washington, DC: The World Bank. Zohoori, N. D. Blanchette, and B. Popkin. 2005. Yach, D. 2004. "Cardiovascular Disease, Cancer, and "Monitoring Health Conditions in the Russian Diabetes" in Social Injustice and Public Health. Part Federation. The Russian Longitudinal Monitoring III: How Social Injustice Affects Health. Oxford Survey 1992-2004." University of North Carolina at Vision 2020: Oxford, England. Available at Chapel Hill. http://www.oxfordvision2020.org/view.asp?ID=95. 145 Copyright © 2005 Europe and Central Asia Human Development Department/The World Bank 1818 H Street, NW Washington, DC 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing: November 2005 Library of Congress Cataloging-in-Publication Data has been requested.