Report No. 10289-VN Viet Nam: Population, Health and Nutrition Sector Review September 21, 1992 MICROFICHE COPY Country Department I Population and Human Resources Operation Division Report No.:10289-VN Type: (SEC) East Asia and Pacific Regional Office Title: HEALTH, POPULATION, & NUTRITIOF Author: CHAMBERLAIN, C. Ext.:81409 Room:E8 033 Dept.:EA1PH FOR OFFICIAL USE ONLY U Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. K) CURRENCY EQUIVALENTS The Vietnamese currency is the Dong (D). A currency reform in 1985 replaced ten oLd dong with one new dong. December 1990 US$1.00 = D 6,835 D 1,000 = US$0.15 Fiscal Year January 1 - December 31 Weights and Measures Metric System ACRONYMS ARI - Acute Respiratory Infections CIDSE - International Cooperation for Development and Solidarity CBR - Crude Birth Rate CDD - Control of Diarrheal Diseases CPCC - Committee for Protection and Care of Children CDR - Crude Death Rate CPR - Contraceptive PrevaLence Rate DRC - Domestic Resource Cost DRV - Democratic Republic of Viet Nam EPI - Expanded Program of Immunization GDP - Gross Domestic Product GNS - General Nutrition Survey GNP - Gross National Product GSO - General Statistical Office IEC Information, Education and Communication IMR - Infant Mortality Rate IUD - Intrauterine Device MCC - Menonites Central Committee MCH/FP - Maternal and Child Health/Family Planning MERUFA - Medical Rubber Factory MIS - Management Information System MOH - Ministry of Health MOP - Manpower Development Plan NCPFP - National Committee for Population and Family Planning NGO - Non-Governmental Organization NIHE - National Institute of Hygiene and Epidemiology NIN - National Institute Nutrtion PATH - Program for Appropriate Technology in Health PHN - Population, Health and Nutrition Sector PIACT - Program for the Introduction and Adaptation of Contraceptive Technology SCERFC State Committee for Economic Relations with Foreign Countries SCF - Save the Children Fund SIDA - Swedish International Development Agency SRV - Socislist Republic of Viet Nam TFR - Total Fertility Rate UNICEF - United Nations UNIDO - United Nations International Development Organization UNFPA - United Nations Population Fund USNCHS - United States National Center for Health Statistics VAC - Vaon (garden), Ao (Pond) and Chan nuoi (animal husbandry) VACVINA - Organization of Retired Horticulturalists VIMEDIMEX - Viet Nan Medical Import Export VINAPHA - Viet Nam Pharmaceutical United Enterprise VNDHS - Viet Nam Demographic Health Survey VINAFPA - Viet Nam Family Planning Association WFP - World Food Program WHO - World Health Organization FOR OMCIL USE ONLY VIET NAM POPULATION. HEALTH AND NUTRITION SECTOR REVIEW Table of Contents Paae No. SUMARY AND CONCLUSIONS ...... ..... . ..... .. i-xiv I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 1 A. Importance of Investments in Population, Health and Nutrition . 1 B. The Country Setting . . . . . . . . . . . . . . . . . 2 C. Recent Trends... . . . . . 2 D. Objectives of the Sector Review. . . . . . . . . . . . . 2 S. Recent Macroeconomic Policy Reforms and Implications for the Health Sector... 3 F. Unique Features of Viet Nam that have a Bearing on the Health Sector. .. . 4 II. POPULATION AND FAMILY PLANNING . . . . . . . . . . . . . . . 5 A. Population Outcomes. . . . . . . . . . . . . . . . . . . 5 B. Family Planning Policy . . . . . . . . . . . . . . . . . 11 C. Demand for Family Planning Services . . . . . . . . . . 14 D. Supply of Family Planning Services . . . . . . . . . . . 26 E. Family Planning Expenditure . . . . . . . . . . . . . . . 29 F. External Assistance ..31 G. Issues and Recommendations . . . . .. . . . . . . . . . 32 XII. NUTRITION . . .. . . . . . . . . . . . . . 37 A. Child and Maternal Nutritional Outcomes . . . . . . . . . 37 S. Nutritional Inputs . . . . . . . . . . . . . . . . . . . 45 C. Food Production and Security . . . . . . . . . . . . . . . 54 D. Nutritional Policy and Strategy . . . . . . . . . . . I . 60 E. Current Nutrition Interventions . . . . . . . . . . . . . 61 F. Institutional Arrangements . . . . . . . . . . . . . . . 64 G. Issues and Recommendations . . . . . . . . . . . . . . . 64 This report was prepared by Anil Deotalikar (Consultant) in conjunction with Oscar Echeverri and Shaikh Hossain (EA1PH). Important contributions were also made by Neera Shekar (Consultant), Jafes Greene (ASTPH), Chris Chawberlin (EAIPH), Xiyan Zhao (ASTPH), and James Altman (Consultant). Peer review was provided by S. Cochrane (PHRHN). B. Liese (HSDDR), S. Lieberman (EA3PH), and J. McGuire (PHNHN). The report is based on the findings of two missions to Viet Nam in January and August 1991. The World Bank would like to express its gratitude to the Ministry of Health, the National Comittee for Population and Family Planning, the National Institute of Nutrition, the General Statistics Office, and the provincial health services of Hai Hung, Quang-Nam Danang, TP Ho Chi Minh, and Cuu Long for their kind assistance and support, and to the Swedish International Development Agency for financing the 1991 Health Service Utilization Survey. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Paua Ng IV. H&ALTH . . . . . . . . . . . . . 69 A. Health Status: Achievements and Current Status . . . . . 69 B. Government Health Policy *.. . - * . 74 C. Structure of the Health Sector System . . . . .77 D. Health Care Inputs . .1.. ... . ...... . 78 B. Utilization of Health Services .......... 97 F. Health Inputs and Health Outcomes ............ 102 G. Health Expenditure . . . e . ........... -'*. . . 103 H. Health Financing ......... 108 I. External Assistance and Non-governmental organizations 113 V. MAJOR ISSUES AND STRATEGIC DIRECTIONS FOR THE HEALTH SECTOR 117 A. Vietnam's Health Sector in Historical Perspective . . . . . 117 B. Issues and Strategic Directions . . . . . ... . 119 ANNEX As Determinants of Interprovincial Fertility Differentials 131 ANNEX B: Determinants of Child Nutrition Status and Nutrient Intakes 134 ANNEX Cs Health Programs Included in the National Plan of the Council of Ministers ..... . . . 139 ANNEX Ds The 1990-2005 Manpower Development Plan.. o. . 143 ANNEX 3: Determinants of Utilization cf Health Services . .146 ANNEX F: Estimation of Choice of Provider with Data from the Survey of Health Facility Users and Providers . . . 148 ANNEX G: Determinants of Interprovincial Variations in Infant Mortality ...... . . . .. * . .. . . 152 STATISTICAL ANNEX . . ....* . . . 154-171 VIETNM POPULATION. HEALTH AND NUTRITION SECTOR REVIEW SUMMARY AND CONCLUSIONS A. Introduction $. Viet Nam's record on most social indicators La far better than that of most other countries at its income level, but a few other indicators range from average to below average. For example, the 1989 Census placed its infant mortality at 45 deaths per 1,000 live births -- a rate comparable to the average for middle income developing countries. The Census estimated average life expectancy at birth to be 65 years -- 63 years for males and 67'h years for females. Viet Nam' population growth rate of 2.1% per annum places it close to the average population growth rate of the group of low-income countries. The Census estimate of a total fertility rate of 3.8 is also close to what would be expected at Viet Nam's level of per capita income. But levels of child malnutrition in Viet Nam are high compared to other Asian countries. Thus, Viet Nam's performance on health indicators is far better than that of countries at its income level, while its performance on population and fertility is about average and on nutrition, below average. These and other incongruencies in social indicators demonstrate the exceptional gains of the PHN sector as well as the effects of poverty and an uneven performance in service delivery. ii. Nonetheless, Viet Wam's record on fertility decline and health Improvement during the last three decades has been impressive and matched by few low-income countries. For instance, the infant mortality rate was roughly halved from about 156 per 1,000 live births in 1960 to around 83 in 1979, and was then approximately halved again during the next ten years. The total fertility rate, which was 5.1 children in 1979, fell by over 25% between 1979 and 1989. A number of factors jointly contributed to the health gains achieved by Viet Nam: establishment of a vast network of primary health facilities throughout tne country; implementation of a number of very effective categorical health programs to deal with preventable health problems, such as malaria, diarrheal diseases, and immunizable diseases; and high rates of adult literacy, especialy among women. A strong political commitment to health resulted in a generous allocation of resources to the health sector in the past. iii. However, beginning with reunification in 1975, a number of problems arose. Among these problems were the burden of improving the poor health infrastructure of the South, the emigration of skilled health workers from the South, sharply diminished external assistance to the population, health and nutrition (PHN) sector sinco 1979, acute macroeconomic instability and a deepening fiscal crisis in the mid-1980s. The political imperative traditionally accorded the health sector probably also began to ebb around the same time. Under these circumstances, inputs to the health sector -- drugs, equipment, medical supplies, maintenance -- began declining. The quality of care offered by the primary health facilities slipped as did utilization rates. Thus, a well-functioning health service system has been breaking down as a result of a number of factors, including the difficult transition from a centrally-planned economy to a market economy. - Li - iv. An a result of these developments, the gains in health and fertility achieved over the last three decades are currently threatened. Indeed, although the evidence is sketchy and far from reliable, a few health and nutrition indicators are already beginning to show signs of reversal. There has been a major resurgence of malaria in the mountainous regions of the North and in areas bordering Cambodia, caused by scaling back of communicable disease control programs. Malnutrition rates in Viet Nam are significantly higher than those to be expected. Using conventional WHO standards, Viet Nam had a higher proportion of underweight (25%) and stunted children (56.5%) from 1987 to 1989 than almost any other low-income country in South and Southeast Asia, excepting Bangladesh and possibly Myanmar. For infants, there is evidence of recent increases in malnutrition rates. V. There is also evidence to suggest that the health gains in the past have been distributed unevenly. There are wide disparities in PHN indicators across regions and provinces and between urban and rural areas. For example, the total fertility rate has a large variance, ranging from 2.2 children in TP Ho Chi Minh to 6.8 in Lai Chau. The infant mortality rate ranges from a low of 26 in Thai Phong to a high of 78.5 in the province of Gia Lai-Kong Tum in the Central Highlands. The proportion of children under 5 years of age that are stunted varies from 28.4% in the cities of Ho Chi Minh and Hanoi to 64.6% in the Red River Delta, a chronic food deficit area. In general, the mountainous areas of the North and the Central Highlands have the highest levels of infant mortality and fertility, while the deltas of the Red River and the Mekong River have the lowest. B. Ponulation and Family Planning vi. Fertility rates have declined in Viet Nam, in part because of a sharp decline in total infant mortality and due to the density of public service providers (see Annex A for details of the statistical analysis). Strong governmental commitment to the family planning program should get a large share of the credit for lower fertility rates. The Government has recognized that rapid population growth is a serious constraint to economic development, and has consequently committed itself to a long-term family planning program. In spate of its low share in the national budget, the program now appears to be better funded than other government programs, and enjoys support at the highest levels of government. vii. The success of the Government's effort is apparent in the very high levels of contraceptive awareness in the country, and thus further reductions in fertility may be a problem of supply rather than demand. Available evidence suggests that there is already a large latent demand for fertility regulation and birth control among Vietnamese couples; actual usage of modern contraception is constrained, in some part, by the inadequate supply of contraceptives and the virtual absence of contraceptive method choice. viii. Population Indicators. The age pyramid in Viet Nam has a very wide base and a narrow top (especially starting with the age group 35 years), reflecting the influence of wars during the 35 years, 1945 to 1979. As a result, the proportion of children aged 14 years and lower is exceptionally high, at 39%, and the median age of the population is low (20.2 years in 1989). A consequence of the youth-heavy population structure is that birth rates, although falling, will remain high for some time. Given the population - Lii - growth momentum, the addition of some 30 million people to the Vietnamese population during the next 25 years is almost inevitable. This growth in population to 100 million people may well have serious repercussions on the economy, environment, and health care system. ix. Another unique characteristic of Viet Nam is its extraordinarily low max ratio (i.e., tho number of males per 100 females), reflecting the loss of males due to war and the large-scale migration that followed. In 1989, the sex ratio was 92, with lower ratios in most of the prime reproductive ages. It is possible that the birth rate fell sharply in recent decades because a large cohort of women were either separated from their husbands or widowed. As the sex ratio increases, it is plausible that the decline in birth rates may slow down or even stall. It is therefore imperative for the Government to sustain the progress achieved by the family planning program in the past. x. ContraceotLys Prevalence and Methods, With 53.8% of the married women aged 15-49 years reporting use of contraceptives in the 1988 Demographic and Health Survey, the contraceptive prevalence rate (CPR) is high. However, a large proportion of the women (28.5%) use traditional, less-effective methods like withdrawal and natural rhythm. A CPR of 54% for 1988 is quite similar to the CPRs in other Asian countries, such as Thailand (68%) and Indonesia (46%), that began family planning programs about the same time as Viet Nam and have roughly comparable age and nuptiality structures. xi. A peculiarity of contraceptive use in Viet Nam is the lop-sided distribution of modern contraceptive methods, with an overwhelming 87% of contracepting women relying on the intrauterine device (IUD), and methods such as sterilization, condoms, and pills being rarely used. The IUD has been the main method of choice because of its low cost, the provision of free supplies from the Eastern European countries, the non-availability of imported supplies of pills and condoms due to lack of foreign exchange, and its appeal among health workers because of its durability of protection and easier monitoring. In addition, the low usage of sterilization in Viet Nam is puzzling, beuause sterilization is a physician-intensive contraceptive method and there is no dearth of such health workers in the country. Perhaps, the long exposure to war and the consequent death of many young men has made Vietnamese women less enthusiastic about a permanent method like sterilization. Another possible reason for the lack of popularity of sterilization is that, the training, surgical equipment and facilities for such operations are simply lacking at most health centers. xiL. Another outcome of the shortage of contraceptives is the practice of menstrual regulation and induced abortion. Owing to the sheer lack of alternatives and the Government's laissez-faire attitude, menstrual regulation and induced abortion have become popular according to survey results, and their use has increased sharply in recent years. However, the Government, in cooperation with the UNFPA, is planning for a 50% reduction in the rate of induced abortions by the year 2000. xiii. Unmet Need for Family Plannino. According to the 1988 Demographic and Health Survey, the unmet need for contraception -- chat is, the proportion of couples who reportedly wish to regulate their fertility but are not using any contraceptives -- is very large (almost 41% of the women surveyed). The same survey showed that the total wanted fertility for women aged 15-44 was - iv - 2.5 childron, but that their actual total fertility rate was 4.5 children. In rural areas, total wanted fertility was 2.6 children, whereas actual fertility was 5 children. The sheer magnitude of the difference between wanted and actual fertility suggests a very large unmet need for family planning in Viet Nam. A third of the surveyed women cited possible health risks and cultural factore as reasons for not using contraception. It is not clear what proportion of the unmet need for family planning methnds is due to the lack of availability of contraceptlves, but casual empirLcism suggests that it is llkely to be large. xiv. SupplY of FamilXy Plannina Services. Since the provision of family planning information and services is integrated with tho provision of basic health care services Ln Vlet Nam, many of the issues relating to the reform of the health care system are relevant to the family planning program as well. For example, a more ratlonal management information system and retraining of health workers in newer family planning methods are needed to improve productivity in the family planning sector. In addition, this sector review suggests the need to expand the supply and distribution of contraceptives and to increase the variety of contraceptive methods available to couples. The evidence from other countries suggests that pills and condoms are the preferred birth control methods, especlally among newly-married and younger women interested in delaying their firat pregnancy or spacing their children. xv. In order to alleviate supply constraints and improve the mix of contraceptlve methods, the donor communlty may have to take on the responsibility of providing Viet Nam with oral contraceptives, condoms, IUDs, injectables and implants as required, until such time as the Government is able to allocate the necessary fo_eign exchange to import these supplies. Additionally, private traders could be encouraged to purchase those supplies that are ln short supply or that cannot be provided by the official program. The evidence indicates that the incipient private sector is already playing an important role in the provision of oral contraceptives and condoms, especially in the South. In order to promote more effective availability, assistance will also be needed to improve the procurement, storage and distribution system. xv. Information, Education and Communicat$on (ICI. Although populatlon education has made significant strides in the country, the fact that more than one-third of the married women in their reproductive ages not using contraceptives cited potential health risks for not using contraceptives indicates the need for greater information and education. The present coverage of IEC campaigns ia not wide because of lack of materials, main- tenance of media equipment and retraining of IEC staff. The Government is fully aware of this situation and the National Committee on Population and Family Planning allocated 25% of its budget to IEC activitles in 1991. However, greater technical assistance in developing messages, IEC training, and procuring materials and equipment for media is needed. xvlL. Famil Planning Ex2enditure. Although real family planning expenditure appears to have been maintained during the period of macroeconomic instabLilty and economic reform, Viet Nam currently spends a much smaller share of total government expenditure on family planning than other Asian countries at its income level (such as Bangladesh and Nepal) I/ A disturbing trend is the evolving composition of family planning expenditures. The share of contraceptive supplies in the total family planning budget has been falling, as has the share of total family planning expenditure spent on salaries of service delivery personnel, while that spent on administrator salaries has been either flat or increasing. Medical equipment has shown the sharpest increase in expenditure share. Insofar as service delivery personnel have the greateot influence on the quality of services dispensed to clients, the trend of its declining expenditure share needs to be addressed in order to improve the quality of family planning services. C. Nutrition xviii. Malnutrition rates among children are very high in Viet Nam. Most recent estimates put the proportion of children malnourished at 45% by weight- for-age and 56.5% by height-for-age -- rates that are comparable to those observed in Bangladesh. Further, although the evidence is sketchy and far from completely reliable, it appears that malnutrition, especially among infants under one year of age, may have increased between 1986-87 and 1987-89. Vist Nam's major child nutrition problem appears to be stunting from longer- term, chronic undernutrition rather than wasting from short-term, acute food deficits. Malnourishment for a significant proportion of children begins in the first year of life. Reasons for this may be low-birth weights (20% of the infants born in 1990 were estimated to weigh less than 2,500 grams at birth), compounded by inadequate breast-feeding and complementary feeding practices. xix. Food Production and Distribution. The cause for high rates of malnutrition cannot be found in inadequate food production in the country. Indeed, there has been an impressive increase in agricultural and food produc- tion during the last decade. However, despite the improvement in overall food availability, the poor system of food procurement, transportation, storage and distribution, combined with the existence of major agricultural productivity differences across regions, has meant that the population in some parts of the country continues to be prone to chronic undernutrition. The Government's past encouragement of rice self-sufficiency in each region retarded development of the food distrLbution system. In addition, this preoccupation with rice self-sufficiency resulted in the neglect of other subsidiary food crops, which are important in increasing dietary balance and variety. As a result, mLeronutrient deficiencies, especially of Vitamin A, iron and iodine, are pervasive in the country. xx. The problem of food availability in the food-deficit areas will not disappear overnight, since consumers in these areas do not have the purchasing power to bid up the price paid for foodgrains from the surplus regions. In fact, at present, it is financially more rewarding to export rice outside Viet Nam than to transfer it to the deficit regions within the country. Indeed, as private-sector grain trade expands, the availability of food in the deficit regions may initially decline before it improves, pointing to the need for public policy interventions. Major improvements in the food 1/ Of course, these countries have significantly higher levels of total fertility. Also, the effectiveness of funds spent is often more important than the amount spent. - vi - procurement, storage, transportation and distribution system will need to be undertaken, so that food can move from surplus to deficit areas efficiently and quickly. xxi. In the long run, however, the only sustainable means of improving food security in the deflcit regions will be by letting them exploit their comparative advantage in crops other than rice. There is substantial scope for increasing agricultural production in the deficit areas. Better supply of fertilizer and other inputs in the food deficit areas, combined with official financing of research and extension work targeted on their crops and farming systems, will allow the food-deficit areas to increase production of crops in which they can specialize. As personal incomes increase in Viet Nam, the demand for crops other than rice is likely to increase. xxii. Role of Economic Growth in Reducinc Malnutrition. Food insecurity is often more closely linked to personal incomes and consumer purchasing power than to food availability in a monetized economy. Survey data for 1987-89 indicate positive associations between indicators of socioeconomic status (such as presence of a bathroom or availability of running water in a house) and child nutritional status. Energy intake also appears to be responsive to household socioeconomic status. If these results are accurate, the economic growth that Viet Nam has experienced since 1989 should help alleviate some of its malnutrition problems. However, since food subsidies were also abolished after 1989, it is not clear whether the purchasing power of households (especially, employee households that were the major beneficiaries of food subsidies) has diminished or increased during the last 2-3 years. Further- more, evidence from other countries indicates that income growth alone can take an inordinately long time to eradicate undernutrition. Consequently, shorter-term measures are needed to avert or reverse growth faltering in children below 3 years of age through a combination of behavioral and health interventions, as well as targeted food supplementation programs. xxiii. Nutrition Education. Overall household food insecurity is a proximate cause of malnutrition, but its effects can be compounded by poor breastfeeding and weaning behavior and the impact of infection, diarrheal and other diseases. The journey into child stunting can begin with low birth weight which in turn may originate partly from the mother's malnutrition as an adolescent. There are two cultural practices in Viet Nam that may exacerbate child malnutrition. First, breastfeeding is generally initiated 3-4 days after birth, largely because of an incorrect perception that colostrum is an inferior food. Second, although the mean duration of breastfeeding is 12-14 months, exclusive breastfeeding is not common, and infants are introduced to supplementary foods as early as 2-3 months of age. Since infant formula is not available widely, breastfeeding is not supplemented with bottle-feeding but instead with solid foods. The premature introduction of supplemental foods greatly increases the risk of infection in small infants. xxiv. Another social-behavioral problem is that the typical Vietnamese diet -- dominated by rice -- is inadequate for older children and pregnant and lactating women. The caloric density of rice is very low, with the result that young children and pregnant/lactating women are often unable to obtain enough calories from rice to meet their special energy needs. In addition, rice does not contain several important nutrients, such as iron, thiamine, Vitamin A and fats (lipids). As a result, women and young children on an - vLL - exclualsv rlce dLet are vulnerable to dlmeases associated wlth theme deficienciesp such as anemia, beriberi and xerophthalmla (whLch can lead to blindness). The combination of low overall calorie intake and little variety ln the dLet create a fragLie nutrltlonal balance for the most vulnerable groups. xxv. There appears to be an lmportant role for providlng nutrltion educatlon, especially on lnfant nutrltlon, breastfeeding and dletary varlety, to pregnant women and mothars of young lnfants. Whlle the Natlonal Instituto of NutritLon (NIN) has been developing educatlonal materlal 'o help mothers lmprove the nutritional status and health of thelr chlldr -here is no instLtutional mechanLem or program for wldely dieseminating such materials. The Government urgently needs to designate an agency to take the lead ln LmprovLng famlly nutrition behavlor and to promote development of a strategy and program to do so. xxvL. A National Nutrltion Program. Indeed, glven the high rates of chlld malnutritlon, a strong case mlght be made for a national child nutrition program that would include growth monltoring, selective short-term supple- mentatlon, and nutritlon education. Such a program could be lmplemented through the primary health care system. In order to contain costs, the nutrLtLon interventlon program could be area-targeted (to reglons having the highest child malnutritlon rates), age-targeted (concentrating exclusLvely on chLidren 6-36 months of age and pregnant and lactatlng women), and need- targeted. Targetlng by need could be achieved by monitoring the weights of all chlldren 6-36 months old in the project co,imunes, and enrolling only those children whose weight gain over a certain period falls below standard. Those chlldren would be singled out for special health monltoring, food supplementa- tion and intensive nutrltlon educatlon for their families. In designing such a program, Viet Nam might be able to derlve lessons from the experience of other developLng countries that have experimented with such integrated nutrltion programs. xxviL. Whether the NIN can take on this broader role given its current staffing and more limited responsiblilties ls unclear but needs to be determined. The absence of a national nutrition strategy and program, training programs, and an effectlve mechaniem for intersectoral coordination on nutrition underscores the need for establishing a nodal point within government for nutrltlon cov -ination and leadership. D. Hmalth xxvllL. Morbidity and Mortality Patterns. Although Viet Nam has a low lnfant mortallty rate and high average life expectancy in relation to other developing countries, it resembles a typical low-income country in its disease proflle, wlth preventlble communicable diseasee being the leading causes of mortalLty and morbidity among adults and children. Malaria, followed by dlarrhea and resplratory lnfectlons, account for the majorLty of reported illnesses, while tuberculosLs, malaria and dlarrhea are the leading causes of mortallty. xxix. Recent Policv Reforms. Since 1989, the Government has implemented a number of bold measures designed to liberalize the health sector and - viii - mobilize new resources for the sector. These include introduction of user fees for health care in all facilities except commune health centers, legalization of private practice, male of drugs and medicines in the open market, and liberalization of the pharmaceutlcal industry. Theme measures place Viet Nam in the forefront of socialist economies attempting to restructure their health care systems. xxx. Pogulation Coverace of Health Services. The socialist policLes of the past were successful in establishing a network of basic health services that reached out to the majority of the population. In fact, official statistics show health servioes coverage to be 100% in urban areas and 75% in rural areas. Viet Nam ham more primary health care centers and hospital beds per 1,000 persons than most low- and middle-income countries in Asia. There is thus little need for expanding the total number of health facilities in the country. Of course, access to health facilities is not uniformly good across the country; because of low population densities in the mountainous provinces of the North and the Center, average distances to health facilities are still quite large in these provinces. But coverage in terms of existing facilities cannot be equated to coverage with services (see below). xxxi. gualitv and Utilization of Health Services. Quality of public health facilities and services in Viet Nam is low. Many commune health centers are simply housed in the homes of health workers, in the quarters of the People's Committee, or in very inadequate structures. Many district hospitals have very inadequate or no kitchens, laundry units, boilers, linen, or furniture. Medical and surgical equipment is sparse, antiquated, and barely functional in many cases. A significant proportion of commune health centers have virtually no drugs or injections to dispense. A survey of health facilities in three provinces found that only 49.3% of rural commune health centers had a functioning sterilizer and only 58.4% had a usable weighing scale for infants. Low salaries of public health personnel also contribute to poor quality of services. The problem of inadequate quality is worse in remote mountainous areas, where many commune health centers are essentially nonfunctional. Not surprisingly, utilization rates of basic health services are very low in the country. Bed occupancy rates for hospitals average less than 50%. Annual per capita contact rates with the health services average between 0.3 and 0.5 for the overall population, with wide variations across provinces and regions. Therefore, an important goal of restructuring the primary health care system should be to improve quality and thereby raise utilization rates significantly. xxxii. Salaries of Government Health Workers: Wages of health workers in the public health facilities are not only extremely low, but they have been falling precipitously in real terms during the last few years. On the other hand, physicians in the incipient private sector typically earn monthly incomes that are 3-4 times the public-sector salaries, and government health workers themselves can often double or treble their monthly salaries by practicing privately on the side. The low public-sector salaries, therefore, create severe incentive problems, and engender absenteeism, low productivity, and low morale among government health workers. The problem is most acute in rural areas, where a large number of health workers also engage in farming to augment their earnings. The productivity of public-sector health workers is also much lower than that of private-sector workers. For example, the survey of health users and providers showed a typical commune health center with a - ix - staff of about 5 health workers seeing an average of only 6 patients per day. In contrast, a single private physician or traditional healer was observ6d to examine the same number of patients in a day. xxxiii. Public Health Exoenditures. There are three observations to be made regarding government health expenditure. First, per capita government expenditure on health, which was 5,664 dong (US$0.83) in 1990, is considerably lower than that of any other country in Asia, including Nepal and Bangladesh. However, its health spending relative to ita per capita GNP and total govern- ment spending is not unusually lowl for instance, Viet Nam spends roughly the same proportions of its per capita GNP (0.8%) and total government expenditure (4.4%) on health as Indonesia and Thailand. Second, the available evidence appears to indicate that real health expenditure over time remained fairly stable between 1984 and 1990. This is impressive in itself, considering that Viet Nam was experiencing hyperinflation, acute macroeconomic instability, and a sharp drop in total external aid during this period. Third, there are enormous disparities in government health expenditures across provinces. These patterns indicate that Viet Nam will need to raise public revenues to finance public expenditures generally, including health, while taking steps to reduce regional disparities in health expenditure. S. Maior Issues and Strateaic Directions xxxiv. The picture that comes across from this review is of a once- vibrant health sector that has been in a process of decline -- for lack of funds, morale and political imperative -- for number of years. Clearly, rebuilding the entire public health sector is out of question. The Government has neither the resources nor the political imperative to restore the public health sector to its past level. Further, it is not at all clear that this would represent an efficient use of scarce resources. But, at the same time, it is important to arrest the erosion of past achievements in health that appears to have already begun. Maintaining the health indicators at their generally good levels is much less difficult than restoring them from significantly deteriorated levels. The strategy for the Government in the future should be to focus on a few key interventions that it can do well, and permit the emerging private sector to increasingly shoulder a larger share of the responsibility for curative health care. xxxv. Risk Reduction at the Community Level. The highest priority should be interventions at the level of the community that significantly reduce the risk of infection among individuals. Such interventions typically have a high impact on health outcomes at low cost. In this regard, the supply of safe water and sanitation is one of the most cost-effective community-based health interventions. Although the mortality and morbidity profile in Viet Nam is characterized heavily b-7 diseases that are linked to water supply and sanitation, such as gastroenteritis, dysentery, typhoid, cholera, viral hepatitis, and malaria, only about one-half of the urban population and one- third of the rural population in Viet Nam have access to safe drinking water. xxxvi. The focus should be on assisting communities with implementing simple water and sanitation tschnologies. The UNICEF Rural Water Supply Project and the UNICEF Sanitation Project are useful examples of low-cost water and sanitation projects that rely on substantial community involvement - x - water and sanitation projects that rely on substantlal communlty involvement and maLntenance. An approprLate level of water coverage for Viet Nam would be one well or tap for 100-150 people withln a maximum dlstance of 250 metres, whlch would requLre nearly 400,000 addltlonal wells or taps to be constructed by the year 2000. External resources would be needed to meet thli target, in *pite of substantlal communlty involvement ln the constructlon of the wells. Likewise, external donors could support pilot programs that seek to provide lncentlves and subsidies to famllies to invest ln appropriate sanitatlon technologies, such as the sulabh latrine. xxxvll. Public Health Programs in Selected Reaions. At the same time, the Government should selectively rehabilitate and better coordinate the exlsting categorlcal programs bullt around specific and well-defLned lnterventions. The national chlld nutrltion program proposed in para xxvl above is an example of a well-focused, categorical program. In addltion, categorlcal programs addresslng diarrhea, malarLa, and acute respiratory lnfectlons (ARI) have been among the most extenslve of health Lnterventions in Viet Nam. These programs should be contlnued and substantially upgraded. However, thelr management and organLsation n-ed to be changed. Currently, many of the categorlcal programs operate lndependently of each other. Some of them rely too much on a hierarchlcal (vertical) top-down management system. For long-run sustain- ability, greater communlty support and participation will be essential in all such programs. In addltlon, substantial economies of scale could be reaped by integratLng these programs lnto a strong basic health care system. The cost- effectiveness of the categorical programs could be further enhanced by targetlng them to those regLons having generally poor health indlcators and high prevalence of specific diseases, such as malaria and AR!. For effectlve targetLng, however, a comprehensive geographical mapping of the country's. disease profile ls needed. xxxviii. PrLvatizatLon of Health Services. By all indicatLons, private health servLces have grown slgnlficantly as a result of recent reforms, so that prLvate expendlture for health now accounts for about two-thirds of total health expendLture. In thls rapidly evolvlng system, lt li lnevltable i;,at the publlc sector health servlce structure will need to adapt to a more limited role. Some shlft of resources out of the public sector would be deslreable for two reasons: first it would significantly reduce the financial burden on public resources while freeing up the health budget to address salary shortfalls, maintenance needs, and other recurrent operatlonal improvementsl second, lt would allow a more manageable public service to improve quality and compete with the growing prlvate sector, whlch by all evldence is operatlng at a hlgher level of productivity and quallty. xxx$x. The public sector role in Vlet Nam wlll be dlfferent from the one it li playing at present. The Government would have a regulatory function to assure service standards in private and public facilities, and it could provide the private sector with trained manpower. In addition, the Government would still be involved in the provision of basLc health servlces, targetted to women and chlldren, vulnerable and underserved groups, the poor and directed for the most part to control of communicable diseases. This entalls a strong servLce delivery function for the public sector with some inevitable overlaps with the private sector. - xi - xl. Reformino the Primarv Health Care Sector. What is needed to revitalize the prlmary health sector is a package of options, some of whlch would increase costs and others that would offset these cost increasee. It is unlikely that such a package would be expenditure-neutral, but the budgetary implications of thle package would be relatively modest. Among some of the elements of the package would be a retrenchment of government health workers, an Lncrease in the wages of remaining government health-sector employees, selective refurblshlng and reequipping of primary health facillties, and greater cost recovery ln government health services. Government health workers could be allowed to move to the private sector. Public facLilties that are not needed to dellver priority health services could be leased or sold to the private sector, thus easlng the financlal burden on government budgets. xl. Cost Recovery in Health Services. Even if the Government downaLzes the public health sector, it will have to raise user fees ln order to finance salary lncreases for government health workers and refurbishing of government health facilities. The prospects for cost recovery appear good for a number of reasons. Flrst, survey data show that individuals are already paying considerable more than officially-established user fees at government health faclilties. By increasing officLal user fees, the Government health sector could tap into thie revenue stream. Second, the same survey data show individuals paying nearly two times as much for private health care as for health care from commune health centers and lntercommunal polyclinics. Hence, at least the better-off individuals in the country have the capacity to pay significantly higher fees for public health care. Finally, the limited empirLcal analy$si undertaken here with facility-based survey data suggests that the demand for government health facilities Lo not responsive to prlce (with a price elaeticity of -0.23). The fact that the estimated prlce elastLcity is less than one suggests that an increase in user fees will raise total revenues. xliL. However, there are two qualifications to the proposal to increase user fees. First, user fees should be increased only for curative servlces for which there is typlcally private willingness to pay. Preventive services, typically provlded through the categorlcal programs dLicussed earller, have a strong public-goods character, and should continue to be provided free of charge by the Government. Second, there should be a mechanism for protecting the poor from user fee increases. Although there is already such a mechanism ln place Ln Viet Nam, the system of exempting the indigent from user fees is not working in practlce. xl$ii. Before instituting full-cost user fees for high-cost items, such as hospital inpatient care, the Government will have to cover a large segment of the population in risk-sharing, insurance schemes. A number of developing countries have successfully experimented wlth health Lnsurance and other risk coverage schemes. Survey results ln Viet Nam indicate consLdera,le receptivity among the rural populatlon to health insurance and village- chemist schemes. It would be important for the Ministry of Health to launch piot schemeo to explore the viability of communlty risk-sharing arrangements, such as health insurance and drug-revolving funds. xl$v. Greater reliance on the private sector for curative health care also needs to be accompanled by greater cost recovery in medical education and - xii - training. While it may still be efficient for the Government to supply trained manpower to a heavily-privatized health sector in Viet Nam, it would be imprudent to continue subsidizing the medical education and training of private-sector health workers. Tuition fees would need to be set and collected, so that public subsidies for medical training could be significantly reduced or eliminated. xlv. Institutional Caoacitv Buildina in the Health Sector. The PHN sector in Viet Nam currently lacks an institutional capacity to monitor, manage and evaluate programs and interventions. This problem is most visible at the level of the district where primary health services should be managed. The district should have the managerial, administrative and technical capa- bilities to make medium-term health plans and efficiently allocate resources, including staff, equipment, and drug supplies, for their implementation; to identify projects, evaluate their feasibility, and execute them; and to monitor and evaluate national health programs operating in the district. Planning and budgeting for efficient resource use in the sector, especially at the lower administrative levels, should be strengthened. xlvi. Limited information that would be useful for monitoring, evaluation and management purposes does exist, but it is poorly organized and difficult to retrieve. In place of the separate registers maintained by the categorical disease control programs, it would make sense to move to a simpli- fied and integrated system of record-keeping that is individual- or child- based. In addition, there is no system of regularly collected information on the time spent by health workers in various activities and on worker perfor- mance and productivity. Such information is essential not only for better overall management but also in formulating wide-ranging policies on attrition and redeployment of the country's health manpower. Finally, to make the record-keeping at commune health centers useful, there should be a mechanism for the data to be transmitted to the higher levels of decision-making for evaluating the impact of alternative health interventions. xlvii. Internal Efficiencv of Government Health Expenditure. Efforts for improvement in internal efficiency through compositional shifts in recurrent budget should be continued. For example, since preventive services are public goods, it would be more efficient for the Government to subsidize preventive services and either leave the provision of some curative services to the private sector or sharply increase user fees for these services. Another area in which the internal efficiency of government expenditure could be improved is in the manpower mix. The ratio of 3.5 nurses to each physician is com- parable to ratios observed in developed countries, where the heavily curative- based systems of health care require relatively intense use of physician services. Since the leading causes of morbidity and mortality in Viet Nam are preventible diseases (primarily, infectious and parasitic), which typically do not need physician-intensity, there may be scope for substantial cost saving by changing the personnel mix toward community health workers, nurses, midwives and assistant doctors. Another example of internal inefficiency is the wide disparity in provincial government health expenditures. A strategy of redistributing government health expenditure from richer to poorer provinces will not only promote equity goals but will also bring about a larger aggregate decline in the infant mortality rate. - xiii - xlviii. External Resource for the Health Sector. Without enhanced foreign aid flows, the Government's intended health sector development goals probably will not be feasible. Increased domestic resource mobillzation may be impaired by inadequate private savings. However, equally important is the effective utilization of external aid flows in the long run. Improvements in ln the choice of core investments to be financed within capital budget, pro- curement, the use of technical assistance, and staffing of project entities need immediate attention. Critical problems like salary, equipment, medical and contraceptive supplies, manpower training, information/education/ communication (IEC) should be tackled first in externally supported projects. lor instance, although the salary iesue cannot be resolved except at the macro level, the provieion of better equipment, improved supplies, and manpower retraining within an individual project will have a positive impact on raising the morale and productivity of health and family planning workers. xlix. Imolieations for PHN of Economic Growth. Income Distribution and Povertv An important issue is the effect of economic growth and liberaliza- tion on health and nutritional status in Viet Nam. There are several indica- tion. that economic liberalization and the consequent emergence and growth of the private sector have widened income disparities in the country. At the same time as entrepreneurs and individuals working in the incipient private sector have experienced rapidly growing incomes, public-sector employees and other salaried persons have found their real purchasing power eroded, as their salaries have failed to keep up with inflation and the food subsidies that benefited them have ended. The widening disparity in income is a relatively new phenomenon for Viet Nam (particularly, the North), and has important implications for the population, health and nutrition of the poor. In other countrLes the combination of worsening income distribution and price decontrol (resulting in higher prices for food, health care and contraceptives) has reduced food consumption of the poor and their utilization of health and family plannlng services. Unfortunately, little information is available on the growth of income dieparities, poverty, and consumption in Viet Nam. This is an area in which further research would have high payoffs, especially in helping the Government find ways of cushioning the poor and other disadvantaged groups from the sharp price increases associated with increased liberalization and greater cost recovery in the government health sector. 1. Emeroina Environmental Problems. Another issue that will become inereasingly important in the future is the impact of worsening environmental and ecological conditions on health. Crowding, pollution, stress and occupational hazards are already beginning to adversely affect the environment and the quality of life ln Viet Nam. Hazardous solid wastes from industry and agrieulture are usually collected jointly with other common wastes, and the most common methods for waste disposal are open dumping an open burning, both of which produce health hazards, air pollution, and sanitary discomfort. Pesticides and other chemical insecticides are becoming increasingly common in agriculture, resulting in contamination of rain, surface and underground water and contamination of the food chain. In addition, increasing population pressure, combined with a housing stock that is expanding very slowly, is resulting iu overcrowding and unsanitary living conditions. With an average of 3.1 persons for every room, Ho Chi Minh City already ranks as a city with one of the scantiest living spaces per capita in the world. The addition of another 30 LilLion or so people to the Vietnamese population over the next 2-3 decades may put a severe strain on environmental health. - xiv - li. DeveloDment of the Pharmaceutical Industry. The Government has listed the development of the domestic pharmaceutical industry as a high priority. A detailed examination of whether Viet Nam has a comparative advantage in the production of pharmaceuticals is beyond the scope of this study. However, it is clear that, the country is severely deprived of financial resources for investment in new plant and equipment, licensing of internatlonally-available technologies, R&D activities, and imports of raw materials and spare parts. Indeed, owing to these problems, local factories are producing at only 40% of capacity. Unless the factors affecting capacity underutilisation are resolved, it may be premature for the Government to make major investments in new pharmaceutical enterprises. li. While the Government should continue to seek technical and financial support as well as joint ventures for upgradLng existing factories to facllitate increased production, there is a need for an in-depth evaluatlon of the pharmaceutical sector that would address such issues as future trends in the demand for drugs induced by the changing age structure of the popu- lation, the pattern of disease vector and morbidlty rates, alternative procurement of drugs and supplies, and the costs, institutional and manpower requiremnts of domestic pharmaceutical production. Such feasibillty studies might formn the bases for possible donor involvement in assisting selected pharmaceutical enterprises in expansion and upgrading. F. Concludina Remarks liii. It is important to preserve the impressive gains in health that have been made during the last three decades in Viet Nam. While the problems facing the health sector are daunting, the country is fortunate in having an abundant supply of human resources and an extensive network of health facilities. The population is highly literate, which means that responsible family planning, good preventive and curative health care practices, and better nutrition can easily be stimulated. Much of thLs potential demand for services is currently constrained by supply and quality problems, but when the supply constraints eventually ease in the future, further improvements in health and nutritional status and a decline in fertility could be swift. I. INTRODUCTION A. Importance of Investments in Population, Health and Nutrition 1.1 The health and nutritional status of a population are important indliators of overall well-being in a society. In addition, they represent a significant source of a country's human capital. A number of recent studies in such varied settings as Sierra Leone, Sri Lanka, India and the Philippines, have convincingly demonstrated the large agricultural productivity gains from health and nutritional improvements among rural households.l/ 1.2 The effects of chronic malnutrition and infection on children are even worse, since they are more lasting. Intestinal infections severely reduce the absorption of nutrients in the body, thereby causing malnutrition. Malnutrition adversely affects cognitive development and schooling performance, both of which in turn depress future economic productivity and well being. Chronic conditions account for a great deal of absentisism, productivity loss and high cost medical care. All of these effects have been well documented in the literature. 1.3 Health interventions in the form of maternal and child health and infectious disease control can reduce significantly the risks of dying early and greatly increase life expectancy. In turn, the impact of increased longevity and reduced morbidity on productivity can be substantial. One pioneering study for India concluded that declines in mortality during the 19605 -- achieved largely via an impressive malaria eradication program -- accounted for almost one-third of the increase in aggregate productivity in Indian agriculture.2/ Thus, even from an economic perspective, health and nutritional improvements deserve to be high on the list of policy goals. 1/ See John Strauss, 1986, "Does Better Nutrition Raise Farm Productivity?" Journal of Political Economy 94 (2): 297-320; Anil Deolalikar, 1988, "Nutrition and Labor Productivity in Agriculture: Wage Equation and Farm Production Function Estimates for Rural India," The Review of Economics and Statistics 70(3), August; and David Sahn and Harold Alderman, 1988, "The Effect of Human Capital on Wages, and the Determinants of Labor Supply in a Developing Country," Journal of Development Economics. V Another non-nutritional consequence of the overemphasis on local self- sufficiency is that the country has been unable to exploit regional comparative advantage. Hence, although agricultural production has been increasing impressively, it remains well below the country's full potential. -2- B. The Country Settina 1.4 The major beneflts of the socialist system in Viet Nam, established in the North after the departure of the French in 1954 and extended to the South after reunification in 1975, have been in the social sectors. There was a rapid expansion of educational opportunities and a dramatic decline in illiteracy. The 1989 Census indicated that 81.8% of the women and 88.5% of the men over 10 years of age are literate. The political imperative to health resulted in a generous allocation of resources to the health sector. The generous allocation of resources was apparent in the establishment of a vast network of primary health facilities throughout the country after 1954, but particularly after 1968. The expansion greatly increased people's access to primary health facilities, except in a few provinces (mostly in the remote mountainous regions of the country). In addition to this network of health facilities, the country invested considerable resources in developing a number of very effective categorical health programs to deal with priority health problems, such as malaria, diarrheal diseases, and immunizable diseases. An important consequence of the basic health infrastructure and the categorial health interventions was a fall in infant mortality rates and an increase in average life expectancy to levels that are unusual for a country at such a low level of per capita income. The socialist policies of the past have also prompted the socioeconomic status of women, so that women participate equally in all economic and political activities. C. Recent Trends 1.5 The impressive gains of the last three decades, however, are currently under threat in part because of a severe fiscal crisis facing the health sector in the last 6 years and in part because of structural changes. The shortage of funds to the health sector is so acute that it is unclear where the grass-roots facilities are going to find the inputs to continue functioning in the future. While the government has shifted the burden of financing health to the lower administrative levels (viz., provinces, dis- tricts and communes), the revenues collected by these administrative units are inadequate to even maintain current levels of spending, let alone increase them. As a result, there are severe shortages of drugs, medicines and contraceptives in the health facilities. In addition, with health worker salaries falling in real terms -- and health personnel receiving their monthly wages several months late in many cases -- the morale of health workers is low. There has been a major resurgence of malaria in the mountainous regions of the North and in areas bordering Cambodia, as communicable disease control programs have been neglected. Finally, the sketchy evidence that exists suggests that malnutrition, especially among infants, and the proportion of low birth weight babies may have increased in recent years. D. Obiectives of the Sector Review 1.6 There is an urgent need, therefore, to take a comprehenslve look at the health sector in Viet Nam with a view to identifying the cause of the current problems in the sector and the policy priorities that will revitalize the sector. This sector review, which is the first comprehensive evaluation of the population, health and nutrition sector in Viet Nam, attempts to fulfill this need. The objectives of the review are to (a) take stock of the country's past achievements and historical trends in health, population and -3- nutrition; (b) define the major sector issues; (a) catalog actlvities and areas in population, health and nutrition having the greatest need for external assistance, and (d) identify pollcy reforms and investment strategies that will make the Vietnamese health-care system and the family planning program more responsive to the needs of the population, more cost-effective, and more financially independent. E. Recent Macroeconomic Policy Reforms and implications for the Health 8-otor 1.7 The Government of Viet Nam initiated a wide-ranging economic reform program, known as doi moi, in 1986. The initial reforms included a return to household-based farming in agriculture, removal of restrictions on private- sector activities in commerce and industry, and decentralization of decision- making to managers of state-owned enterprlses. Initial progress on the reforms was slow, with many of the comprehensive macroeconomic reforms being adopted forcefully only in 1989. The 1989 reforms lncluded a devaluation of the official exchange rate to the parallel market rate, decontrol of prices, and an increase in real interest rates to positive levels. 1.8 Many of these reforms applied directly to the health sector. Price decontrol included the introduction of user fees for health care, especially at the level of hospitals. The removal of restrictions on private-sector activities included legalization of private medical practice and the commer- cial sale of medicines, drugs and contraceptives. The decree on decentralized decision-making for state-owned enterprises extended to the enterprises manufacturing pharmaceutlcal and condoms. 1.9 Several of these reforms have the potential of influencing population, health and nutrition trends in Viet Nam, although the full scope of their effects may not be known for several years (because of the typically long time lags involved in the relationship between health outcomes and health inputs). The policy reforms in agriculture resulted in a dramatic increase in agricultural production, which in turn expanded per capita food and energy availability in the country. However, because the food procurement, storage, and distribution systems in the country are still rudimentary, the increase in food availability has not yet manifested itself in the form of reduced levels of malnutrition in the food-deficLt areas of the country. The introduction of user fees has dramatically raised the cost of health care for most people. Even though commune health centers were excluded from cost recovery measures, a recent facility-based survey undertaken by the Mlnistry of Health found more than 80% of the patients visiting commune health centers paying for health care (presumably, drugs). It is wldely alleged that the user fees have reduced utilization of health services and bed occupancy rates in state hospitals. Finally, the legalizatLon of private medical practice has resulted in a mushrooming of private health services. Prlvate physicians and tradi- tional healers have been doing brisk business, even ln the rural areas, in the last two years. Many observers have claimed, often without any firm evidence, that policy reforms permitting physicians ln government health facilities to practice privately after hours have resulted in deterloration of services in commune health centers and government hospitals. -4- F. Uniaue Pe tures of Viet Nam that have a Bearina on the Hoalth Sector 1.10 There are several historicul, demographic and other characteristic. of Viet Nam that have a bearing on the health sector. First, it is a country that has been at war over a long period of time. The war of independence with the French, the war with the United States, and finally the war with China all came on each other's heels. The long period of conflict has left a demo- graphic scar; Viet Nam has one of the lowest sex ratios (that is, the number of males per 100 females) of any country in the world. The sex ratio is particularly low in the age groups above 35, reflecting the loss of males due to the war and the large-scale migration that followed. A falling sex ratio has probably contributed in some measure to fertility decline, although it is difficult to know exactly what proportion of the fertility decline can be attributed to this factor. 1.11 The second feature of Viet Nam that distinguishes it from other countries at roughly its income level is its abundant supply of human resources. Viet Nam has a highly literate population. There are more physicians per capita in Viet Nam than in most other low- or middle-income countries. The abundant supply of health professionals probably has something to do with the favorable health status (relative to income) achieved in the country. But, more recently, these health professionals represent an underutilized resource, because they have few medicines, medical supplies, and equipment to work with and because they are vastly underpaid. The high levels of literacy have favored the demand for birth control, preventive health care, prompt curative health care, and good nutrition. Many of these demands are currently constrained by supply shortages, but, when the supply constraints eventually ease in the future, further improvements in health and nutritional status and decline in fertility could be rapid. 1.12 Third, Viet Nam has developed an extensive network of health services that provide good access to basic health care for the vast majority of its population. There are over 10,000 commune health centers in the country, each serving approximately 2,-10,000 persons. Except in the remote mountainous regions of the country, the health centers are within easy reach of most people. While the quality of care available at the commune health centers has been deteriorating in recent years, largely because of the lack of medical supplies and equipment, the availability of these facilities in every community in the country represents a potentially important resource. 1.13 The fourth feature of Viet Nam that distinguishes it from many other low-incomq countries and that also has a bearing on the health sector is that it is currently in the midst of a critical transition from being a centrally- planned socialist economy to becoming a market-based system with decentralized decision-making. Indeed, Viet Nam's bold and far-reaching economic reforms put it in the forefront of socialist economies attempting to revitalize their economic systems. The problems of transition have been compounded by the external shock of a large decline in commodity aid from the Soviet Union. Since the external shock and the policy reforms have affected all sectors of the economy in major ways, many of the trends and problems observed currently in the health sector may be problems that are common to countries undergoing rapid structural changes. However, the changes that Viet Nam is going through are so fundamental that many of the institutions and management styles that worked well in the past may not perform as adeptly in the future. I I POPULATION AND FAMILY PLANNING A. Pogulation Outcomeg Pooulation Count and Growth Rates 2.1 Currgnt Poulation. On April 1, 1989, the Government of the Socialist Republic of Viet Nam undertook a population census. A manual count war completed on June 15, 1989, which showed a total of 64,412,000 persons residing in Viet Nam, making it the second most populous nation in Southeast Asia (after Indonesia) and thirteenth most populous country in the world. About 80% of the entire population is rural. The distribution of population between the North and the South is almost identical; 32.21 million reside in the North, while the South has 31.16 million inhabitants. There is a heavy concentration of the population in the two fertile, rice-growing deltas of the Red River (North) and Mekong River (South); each of these areas accounts for about 21-22% of the national population. Table 2.1: Population and Growth Rates, 1921-1989 Year Population (in million) Growth Rate (%) 1921 15.584 1926 17.100 1.86 1931 17.702 0.69 1936 18.972 1.39 1939 19.600 1.09 1943 22.150 3.06 1951 23.061 0.50 1954 23.835 1.10 1960 30.172 3.93 1965 34.929 2.93 1970 41.063 3.24 1976 49.160 3.00 1979 52.742 2.16 1989 64.412 2.10 Source: General Statistical Office, Viet Nam Pop- ulation Census 1989: Detailed Analysis of Sample Resulto, Hanoi, 1991 2.2 Historical Trends in Pooulation Growth. The last seventy years of Viet Nam's population growth can be divided into five distinct growth periods. The first period (1921-51) was characterized by large fluctuations in the rate of population growth, largely induced by catastrophic events, such as the global economlc depression, the suppression of peasant uprisings by the French, and the famine of 1945 (which was responsible for two million deaths) - 6 - (Table 2.1). The second perlod, covering the pro-war years of political normalcy and economic prosperity (1954-60), stands out for lts high population growth of 3.9% per year. The thlrd perlod (1960-76), covering the years of political turbulence and the war wLth the United States, witnessed a somewhat slower (but still hlgh by absolute standards) annual growth rate of 3.1%. The fourth period, covering the years immediately after the war and reunification, saw a sharp decline ln the populatlon growth rate to 2.16%, induced in part by the economlc devastation and large-scale outmigration related to the war but also by falling fertility. The flnal period is the decade of the 19809 (1979- 89) -- a period of polltlcal normalcy -- which was characterized by a continu- ing drop in fertility and an annual population growth rate of 2.1%. 2.3 Flgure 2.1, whlch plots the crude bLrth rates (CBR) and the crude death rates (CDR) for the last four decades, shows that the source of natural populatlon increase in Vlet Nam has been a sharply falling death rate -- not an iucreasing birth rate. Indeed, the crude blrth rate has been falling secularly since 1950. There li llttle doubt, therefore, that Viet Nam is well past the point of de mographic transition 4/ 2.4 ComRarison with VIETNAM Other Countries. The most POPULATION TRENDS recent growth estimate of 2.1% per year places Viot Nam close to the average population growth rate of the group of low-income countries.c/ Viet Nam's growth rate is also similar o to that of many other Asian countries. For example, the _ _ populations of Thailand, Indonesia, Myanmar, and the I to"s 1070 179 L9 Philippines grew at annual year rates of 1.9, 2.1, 2.1, and _ 2.5, respectively, over the period 1980-88. However, Vlet Nam's growth rate iL Figure 2.1 much hLgher than that of Chlna (1.3%) and Sri Lanka (1.5%) -- countries that have roughly comparable levels of female literacy. 3/ A country is regarded as having undergone a demographic transition when its blrth rate, which often lags behlnd the death rate, starts falling. Once the demographic transltion has occurred, the country's populatlon growth, although still pocltlve, continues to diminish over time. 9/ World Bank, World D-v looment Renort 1990, Oxford University Press, New York, 1990. - 7 - the North as well, growth rates were quite high (an average 2.9%). The provinces having very high growth rates (e.g., Dac Lac, Lam Dong, and Gia Lai- Ron Tum) have been attracting migrants because of the establishment of new economic zones and major construction sites. Many of these migrants have come from provinces in the Central Coast, which experienced some of the lowest population growth rates in the country during 1979-89. Table 2.2: Demographic Indicators By Province,1989 Total PopIlation Pop.growth Area Populatlon IMR In I rate (km.. Desty TFR (Per 1000) lyo n 9m, 00081989 1989 1989 1989 Entire country 64,412 52,742 2.1 329,841 195.2 4.00 45.00 Urban 20.1X 19.2X 2.50 34.00 Rural 79.9° 80.8X 4.40 47.00 U~ainJ'~JitiaI ~ 10 068 7,700 2.9 -cu 1,026 774 3.0 13,684 74.98 4.90 52.80 Cao 0mv ~~565 471 1.9 8,447 66.89 5.00 61.60 L:angon 611 478 2.6 8,140 75.06 4.90 56.50 al Chau 438 316 3.5 16,480 26.58 6.80 66.10 Nw LinSon 1,032 771 3.1 14,746 69.99 5.20 56.30 See Lai 1.030 809 J:6 6*495 158:58 3:90 45:00 Son La 6°32° 482 3.7 14 216 143.597 6.00 5445°° Quang Ninh 813 661 2.2 5,943 136.80 3.40 35.00 Vlrh Phu 1 807 1 376 2.9 4 573 395.15 3.50 33.10 He sac 2:064 1:562 3.0 4:611 447.63 3.50 36.30 &R.-_ ft-I13,576 10,968 2.3 NOed Rlr r Delta 3.056 2.456 2.3 2,142 1426.70 2.80 40.00 H:al Phong 1,447 1,150 2.4 1,503 962.74 2.90 26.00 Ha Son Slit 1,839 ~~1,426 2.7 5,787 317.78 4.00 47.90 Hal Hwtg 2,445 ~~~~1,959 2.4 2,55552 958.07 2.90 38.00 Thai Blnfro 1,632 1 382 1.8 553 1050.87 2.60 31.60 Hea Mar Ninh 3,157 2,595 2.1 3,798 831.23 3.30 34.90 CntalCoost Northla 8,573 6,977 2.2 fbann Hoa 2,993 2,348 2.6 11,162 268.14 4.10 36.20 Nghe Tfnh 3:583 2:870 2.4 22:525 159.07 4.60 53.30 Quang Birth 647 530 2.1 7,788 83.08 4.80 49.90 Quns Trn 459 433 0.6 4 867 94.31 4.80 49.00 Thus ThSen Hue 891 796 1.2 4,948 180.07 4.80 50.00 Centat Cast outhad 6,655 5,537 2.0 CentrNam Conas uhng 1,738 1,458 1.9 11,994 144.91 3.80 47.10 Qu ng Ngal 1,042 900 1.6 5,852 178.06 4.60 51.50 Birth Dirth 1,245 1,085 1.5 6,050 205.79 4.25 51.50 Phu Yen 642 517 2.3 5 070 126.63 4.20 44.80 Khanh Hoa 818 660 2.3 4,706 173.82 4.70 44.80 Thuan Hal 1,170 917 2.6 11,470 102.01 5.10 43.90 Centrot Hlahtand 2,491 1,455 5.8 %P614-151 lum 876 585 4.3 25 670 34.13 6.50 78.50 Dac Lac 976 482 7 7 19:875 49.11 6.20 44.90 Lam Dong 639 388 5.4 10,146 62.98 5.00 43.70 Nortb0ELA ouhth 7,797 6,004 2.8 - TV~IC1iFNII 3,924 3,293 1.9 2,089 1878.41 2.20 30.10 Song Be 938 651 3.9 9,582 97.89 4.40 45.70 Tay Ninrh g93 676 1 7 4,019 197.31 4.40 39.10 Dong Nal 2,007 1,292 4.7 7,572 265.06 4.80 33.80 Vung Tau-Con Dao 135 92 4.1 237 569.62 2.80 32.40 Fek DlI lta 14,171 11,811 1.9 Long An ~~1,120 949 1.8 4,344 257.83 4.20 42.30 Thop '~133 1173 1.4 327 408.00 4.00 48.90 tam 174 ,475 2.0 342 518.26 3.90 50.52 Tin Glang 1,4834 1258 1.7 2,338 634.30 4.00 30.00 B n Tre 1,214 1,034 1.7 2,248 540.04 3.90 41.10 Cuu Lron1,0 1,497 20 3,85 669.26 4.00 40.50 Hau GGang 2,681 2,226 2.0 6,165 434.87 4.20 45.40 Klaem Gfan 1,9 986 21 6,243 191.89 5.40 53.80 N nh Hai 1,555 1,213 2.6 7,656 203.11 4.60 44.40 Sources: Populatfon Census, 1979 and 1989. P02ulation Structure 2.6 Ace and Sex Comoosition. The age pyramid in Viet Nam is typical of that observed in most les-developed countries, with a very wide base and a narrow top (Table 2.1 in the Statistical Annex). In Viet Nam, the top, especially starting with the age group 35, is especially narrow, reflecting in large part the influence of wars during the 35 years, 1945-79. As a result, the proportion of children aged 14 years and lower is exceptionally high, at 39%, and the median age of the population is low (20.2 years in 1989). With the likely exception of Cambodia, these statistics put Viet Nam at the top of the list of Southeast Asian countries (after Laos and the Philippines) in terms of the proportion of children aged 0-14 in the total population. 2.7 However, there was a noticeable aging of the Vietnamese population between 1979 and 1989, caused by declining fertility rates. The median age increased by almost two years (from a level of 18.3 in 1979). The aging of the population will continue in the future as the fertility decline continues -- a fact that should be taken into account in planning for the health care system. 2.8 A consequence of the youth-heavy population structure is that birth rates, although falling, will remain high for some time. With mortality declining rapidly, the size of the surviving parental cohort will be larger than the preceding ones, and will be the main determinant of the size of the birth cohort. Until this parental cohort completes its childbearing years, overall birth rates will remain high even if actual family size is small. The present age structure also implies a continuing increase in the absolute number of young people entering the labor force over the next twenty years. Thus, even if fertility were to fall to replacement levels -- an unlikely proposition for the near future -- growth momentum will make the stationary population 1.5 times the current size. 2.9 The fact that the addition of some 30 million people to the Vietnamese population during the next 25 years is unavoidable given the present age structure is cause for concern. Unless managed adeptly, this growth in population may have serious repercussions on the economy, the environment, and the health care system. 2.10 eg Ratio. A unique demographic characteristic of Viet Nam is its sex ratio, which is extraordinarily low in comparison to that of other developing countries. The sex ratio (or the number of males per 100 females) in Viet Nam has been below 100 since 1931, but fell continuously between 1931 and 1979. It further decreased from a level of 94.2 males per 100 females in 1979 to a level of 92 in 1989 (Table 2.1 in the Statistical Annex). At 92, Viet Nam's sex ratio is considerably below that of many Asian countries, particularly China and the countries of South Asia, where the sex ratio is above 100 and often reflects excess female infant mortality caused by parental neglect of daughters in the provision of health care and nutrition. Viet Nam's low sex ratio can be ascribed to three factors: (i) excess male mortality (particularly, in the middle age groups) attributable to war, (ii) large-scale emigration of males, and (iii) little or no discrimination againot women in access to health care and nutrition. 2.11 Further increases in the sex ratio are likely in the future. The effects of an improving sex ratio on birth rates have not been fully 9- recognized in Viet Nam. It is possible that the birth rate fell sharply in recent decades because the risk of conception was unusually low for women who were separated from their husbands or widowed. It is not unusual to flnd over 10% of the female populatLon widowed (and therefore at lower risk of conception) in some Southern provinces. The Demographic and Health Survey found that 5.1% of all ever-married women in the South were widowed.§/ Such high rates of widowhood are unlikely to be sustained as the sex ratio im- proves. Therefore, it is possible that the decline in birth rates may slow down (or even stall) in the future in the absence of continued government intervention in family planning. Fetility 2.12 Trends. The 1989 Population Census obtained an estimate of total fertility rate (TFR) in Viet Nam of 3.8 chlldren per woman -- considerably lower than thw estimate of 5.1 provided by the 1979 Census. Corroborative evidence of declining fertility is provided by the Demographic and Health Survey of 1988, which estimated the TFR to be 4.5 for the period 1983-87 and 3.9 in 1987. Age-specific fertility schedules for two periods confirm the unambiguous decline in fertility for women of all ages (Figure 2.2). 2.13 Using reverse t 1 _ _ 1 ^9 survival methods, the 1989 Census suggests that fertil- ity rates were flat between 5-/ 1965 and 1974, but declined / steeply after that. The 2- reverse survival-estimated TFR fell from 5.9 in 1970-74 to 3.9 in 1985-89 -- a remarkable decline of over I 33% in 15 years. The decline in fertility between 1955 and 1989 can be divided into four periods. In the first period (1955-69), fertility S 2b A ab Sh 4b 4S 4h rates were high, but declin- Age-Specific Fertility Rates, 1979 and 1989 ing -- albeit at a slow pace I of about 1% per year. The Figure 2.2 second period (1969-79), which covers the war, reunification, and mass emigration, saw a rapid fertility decline (of about 2.1% per year). The fertility decline slowed down ln the third period (1980- 84). ordinarily, most prolonged wars are followed by a "baby boom," but the fact that this did not occur in Viet Nam indicates the success of the family planning program, which had been in effect in the North since the 19506 and was extended to the South after reunification. There was an acceleration in the rate of fertility decline in the fourth period (1985-89), again in large part due to the continued penetration and success of the family planning program. j XThe rate of widowhood was 8.7% and 12.4% for ever-married women aged 35-39 and 40-44 years, respectively. - 10 - 2.14 Comoarison wlth Other Countries. Unlike its infant mortality rate, which is very low relative to its income, Viot Nam does not have a much lower TFR than would be expected at its level of per capita income. This is observed in Figure 2.3, which plots the relationship between TFR and per- capita GNP for 14 Asian countries. Viet Nam is observed to be almost on the TFR regression line, unlike China, Sri Lanka and Thailand, which are clear outliers. 2.15 Another way to view Viet Nam's TFR is in relation to its infant mortality rate (IMR). Empirically, a strong relationship between fertility levels and infant mortality rates is observed across countries. Again, the basis of comparison could be the 14 countries of Asia. Figure 2.4, which plots the regression line between TFR and IMR for this group of countries, clearly shows that the TFR in Viet Nam is higher than would be expected given the normal relationship between TFR and IMR observed in Asia. 2.16 Reaional Differences. There are wide disparities in fertility across rural-urban areas and across provinces. The age-specific fertility schedules indicate that urban fertility is significantly lower than rural fertility for women of all ages. The 1989 Census estimate of TFR in urban areas (viz., 2.2) is almost one-half of the estimate of TFR in rural areas (4.3). The urban- rural disparity in fertility levels might be the result of differential income levels, literacy rates, and infant mortality rates between urban and rural areas. There may therefore be considerable scope for large reductions in fertility in the rural areas of the country. rot FultFty Rate and Per Capita GNP in 1.4 Stan Countries, is9 2.17 Fertility rates also 7L vary a great deal across provinces. Provinces in the Red River and the Mekong River .' 55 Delta areas have the lowest 5z 1* fertility rates in the country. On the other hand, highland -4.25 provinces, such as Gia Lai-Kon U. Tum, Dac Lac, and Lam Dong in a the Center and Son La, Lai 3 Chau, Ha Tuyen and Cao Bang in the North, have very high fer- tility rates (Table 2.2). The 1_75 -_ _ _ _ latter provinces are also do 7d0 15bo 3CUbS sodo characterized by some of the P__Capita_GNP_CUSS) highest infant morta- lity Figure 2.3 rates in the country. This positive association between fertility and infant mortality has been observed in most developing countries (see Annex A). - 11 - Fertility and Infant Mortality In B. Family Planninc 14 Asian Countres, 1989 Policy 7 _ Historical Review 2.18 Viet Nam has ,.pu had a longstanding 55- policy commitment to reduce population growth through a 4.25- national population Elo ar policy and family ifolz planning program. The 3- Government policy to control population growth was first 1.75 articulated in 1963, 175- with the Ministry of 20 4b 6b 80 ioo 1do Health of North Viet Inf ant Mortality Rate Nam as the main agency F 2.4 for administering and delivering family planning services .6/ 2.19 Since then, the program has evolved through a number of stages. During its first decade, despite repeated commitment to reducing the rate of population growth and achieving a more regionally balanced distribution of the population, the program vacillated in terms of intensity and manner of implementation. The main reasons were the war with the United States and subsequent hostilities with China. After the reunification of Viet Nam in 1976, overriding priority was given to post-war reconstruction, and conse- quently the population programs were not given enough attention. In 1981, however, the Council of Ministers at the Fourth Congress reiterated the importance of family planning as a part of national population policy and also recognised this as a social movement to improve further the quality of life of the population. To promote family planning more strongly throughout the country, the Council of Ministers listed three specific guidelines: (a) each couple should have only two children; (b) the two children should be spaced at least five years apart; and (c) a woman should be at least 22 years old before marrying and having a child. With increasing political and financial commitment, following the Sixth Party Congress in 1986, population and family planning program became the second highest priority, after food production. It was in this Congress that a range of quantitative targets on vital rates was set, including a reduction of the annual rate of natural growth to 1.7% by the end of 1990, which would allow a replacement-level fertility by the year 2005. 6/ The Government of South Viet Nam launched its family planning program in 1968, but the program could not succeed due to the exigency of a 1933 French colonial law which forbade the advocacy of birth control. - 12 - 2.20 The period 1988-90 witnessed essentially a progressive consolidation of the program in terms of renewed political support, budgets and infrastructure. The baby "bulge" that resulted from high levels of fertility combined with sharply declining mortality during the 1960e was entering child- bearing ages in large numbers, and caused a growing concern about rapid population growth in the future among policy makers. In anticipation of a heavy burden on social services and long-term economic development by this large cohort of post-war births, the Government, at the recommendation of the Council of Ministers on October 8, 1988, broadened the dimension of the family planning by (a) integrating the program with maternal and child health services -- an important component of improving women's status and family welfare, and (b) recognizing the program as a strategic measure for poverty alleviation and successful development. Current obiectives 2.21 The 1988 population policy paper begins by reiterating the goal of achieving an annual rate of population growth of 1.7% by the year 2000. To attain that target the paper emphasized the following norms: (a) Practicing family planning would be the responsibility of the whole society, male as well as female citizens. (b) Most Vietnamese families were expected to have no more than two children. Families of ethnic minorities could have a maximum of three children. Special exemptions would be granted to couples with children of previous marriages, couples having twins or triplets, and couples having handicapped children. (c) City dwellers were expected to abstain from having children until the wife was at least 22 years of age and the husband was at least 24 years of age. (d) The second child was expected to be spaced 3 to 5 years after the first, unless the first child was born to a mother past 30 years of age, in which case the second child could be spaced 2 to 3 years after the first. (a) Each couple of childbearing age would be required to register with their local administration, which was expected to keep abreast of the couple's specific situation concerning the implementation of family planning guidelines and new births. In turn, the local administration would be required to report regularly on the situation to the next higher level of government. (f) Efforts would be intensified to provide widespread information, education, and communication (IEC) on all aspects of family planning to the population at large. (g) Every agency was expected to put a cadre in charge of population and family planning activities. The meeting of family planning targets would be considered as a criterion of success for such agencies. - 13 - (i) All levels of government, from the Ministry of Finance to the commune-level People's Committees, would be required to provide appropriate funding to the population committee at their level, so that an adequate supply of equipment, implements, drugs, and specialized public health cadres would be available to meet the needs for gynecological examinations and family planning services. (j) A system of incentives and disincentives would be adopted to encourage compliance with these norms. (k) Contraceptive devices would be supplied free of charge to cadres, manual workers, civil servants, members of the armed forces, and the poor. The widespread sale of birth control devices would be permitted in order to facilitate their use by anyone needing them. 2.22 The policy announced in October 1988 was formally incorporated into the Law of People's Health Protection which was promulgated on July 11, 1989. 2.23 There is an incongruity between the family planning policy enunciated above and the actual fertility situation in the country. The official policy is typical of that observed in most low-income countries, and is based on the premise that Vietnamese couples are resistant to further fertility decline. This is certainly not the case, as there is a large unmet need for family planning in the country (see section C below). Indeed, the available evidence suggests that further improvements in contraceptive use, and thereby further fertility decline, are constrained not by high levels of fertility demand among Vietnamese couples but instead by the inadequate supply of contraceptives and the virtual absence of contraceptive method choice. official family planning policy should take account of the uniqueness of the Vietnamese situation, viz., the fact that the potential demand for small families is not only already present but possibly quite large in Viet Nam in relation to other countries at its per-capita income level. Proaram Oraanizat ion and Management 2.24 To improve the Government's capacity to manage and implement the population programs, the Council of Ministers created the National Committee for Population and Family Planning (NCPFP) in April 1984 as an advisory body. The status of the NCPFP was further elevated in 1989, when the Council of Ministers broadened its role as the sole body in advising the Government on policy formilation, program development and intersectoral coordination in the implementation of family planning programs of various Ministries, agencies, levels of government, and mass organizations. The NCPFP also undertakes field surveys and prepares policy studies and briefs to carry out these functions. 2.25 A vice-chairman of the Council of Ministers acts as Chairman of the NCPFP. He is aided by a standing vice-chairman, three coordinating vice- chairmen who are appointed by the Ministry of Health, Ministry of Finance and the State Planning Committee, and a Minister in charge of population and family planning. In addition, the NCPFP has institutional members drawn from various ministries and mass organizations, including the Ministry of Education and Training; Ministry of Labor, War Invalids and Social Welfare; Ministry of - 14 - Culture, Information, Sports and Tourism; General Statistics Office; the Confederation of Trade Union; the Youth Union's Central Committee; the Women's Union's Central Committee; and the Central Committee of the Peasants' Association The NCPFP's Secretariat has five subcommittees, responsible respectively for (a) Demography; (b) Planning and Implementation; (c) Mother and Child Health Care; (d) Provision of Technical Services for Contraceptives; and (e) Education and Information. Since its creation, the NCPFP has had a separate budget, as if it were a Ministry. 2.26 The national committee is paralleled by committees for population and family planning at each of lower levels of government (province, city, district, and commune). The NCPFP sets targets for birth rates for each province; each province then sets targets for the districts; and the districts set targets for the communes.7/ The targets are based on indicators such as the proportion of women in childbearing age, the number of acceptors of family planning, the number of births in the previous year, and the birth rate in the previous year. The vital statistics and other related important information are reviewed with local authorities before targets are officially established. C. Demand For Family Plannino Services 2.27 In this section, information from the 1988 Viet Nam Demographic and Health Survey (VNDHS) -- the first nationally representative demographic survey ever undertaken in Viet Nam 8/ -- will be utilized to analyze the levels and patterns of contraceptive use and the demand for family planning services. Current and Past Patterns 2.28 Contraceptive Awareness. The 1988 VNDHS suggests a high degree of contraceptive awareness in Viet Nam. An overwhelming 94% of the women surveyed knew about at least one method, with 91.8% knowing about the intrauterine device (IUD), which is the single most common contraception method used in the country. The surprising finding is that contraceptive awareness does not vary much across rural and urban areas, between the North and the South, and among women with different levels of education. The single exception is illiterate women (comprising only 6.4% of the total sample), among whom only 74% were aware of the IUD and 81% were aware of any method. Thus, knowledge about contraception appears to be very widely diffused in the Vietnamese population. 2.29 Contraceptive Use. The percentage of currently married women aged 15-49 years using contraceptives currently was estimated to be 53.2% by the 2/ Officials of each commune's Statistics Office, the Health Service, and the Legal Service meet once a month to compare data on vital statistics and to prepare a report submitted to the district authorities. Similar reports are sent at regular intervals from each district to the provincial authorities, and from each province to the national authorities. The VNDHS received substantial technical and financial support from the UNFPA. - 15 - VNDHS. Although the 1988 VNDHS was the first national family planning survey in Viet Nam, the commune health centers in the country maintain statistics on the number and type of contraceptives dispensed as well as on the number of eligible women residing in the commune who are not yet contracepting. Since these data are transmitted to successively higher levels of the family planning administration (viz., the district, province and Center), the MOH has available contraceptive use rates, including the contraceptive method mix, going back to 1975. Although it is not clear how complete and reliable these data are or how comparable they are to the estimates provided by the VNDHS, they are the only information available on the change in contraceptive use over time in Viet Nam. The MOH data indicate a dramatic increase in the contraceptive prevalence rate (CPR) from 8% in 1975 to 34% in 1985. A cPR of 53.2% reported by the VNDHS for 1988 is not strictly comparable to the earlier data, since the VNDHS included non-modern methods, such as withdrawal and calendar rhythm, as contraceptive choices. If only the modern methods are considered, a CPR of 38% is obtained for 1988. 2.30 A CPR of 53% is broadly similar to that observed in other Asian countries that began family planning programs about the same time as Viet Nam did and have similar age and nuptiality structures (Table 2.3). For example, the CPRs in Indonesia and Thailand were estimated to be 46% and 68%, respectively, in 1987. However, if the service statistics are to be believed, the annual percent growth in contraceptive use in Viet Nam has far exceeded that of most other Asian countries. of course, this phenomenal growth reflects the fact that Viet Nam started with an unusually low contraceptive prevalence rate relative to other Asian countries. 2.31 Contraceotiv Method Mix. The 1988 VNDHS shows a predominance of IUDs in the method mix in Viet Nam. Nearly 87% of married women aged 15-49 currently using modern contraceptives rely on IUDs. The next most common method is sterilization (7.8%), followed by condoms (3.1%) and pills (1.1%). 2.32 Table 2.4 shows the distribution of contraception users by method for the period 1975 to 1988. The method mix data from the 1988 VNDHS have been adjusted, so that the individual percentages in Table 2.4 add up to the number of married women 15-49 currently using modern (as opposed to all) contracep- tives. Table 2.4 shows a massive shift from pills and condoms to IUDs, particularly during the 1980s. While there also was a large increase (from extremely low initial levels) in sterilization as a method, the latter remained a relatively underutilized method even in 1988. In large part, the prevailing method mix reflects the supply constraints for other contraceptives and the preference of health and family planning workers. The IUD has been the main method of choice because of its low cost, the limited availability of imported supplies of pills and condoms, and its appeal among health workers because of its durability of protection and easier monitoring. Although the Government maintains that the family planning program is based on a free choice of contraceptive methods, in practice choice is restricted by the shortage of supplies and equipment. A "cafeteria,' approach that provides a variety of contraceptive methods in the family planning delivery system could - 16 - significantly improve both the coverag, and the effectLveness of family planning interventions. 2/ Table 2.3: Contraceptlve Prevalence Rates, Selected Asian Countries % of marrled women % annual Country Year 15-49 currently using change in CPR contraceptives (CPR) Chlna 1975 71 1985 74 1.4 Thallands 1970 19 1987 68 15.2 Indonesia 1976 18 1987 48 15.2 PhilippLnesb 1968 1S 1988 36 4.3 Sri Lanka 1975 34 1987 62 6.9 Viet Nam 1975 8 1988 53c 28.8 Notes * Beginning value of CPR based on Indonesia Family Planning Perspectives in the 1990s, World Bank, WashLngton D.C., 1990, p. 107, Table 2.5. b Based on 1968 & 1988 National Demographic Surveys. c 38% with modern contraceptLves only. Sources World Bank Asla Populatlon Dlscussion Note July 2, 1990, based upon (1) "Levels and Trends of Contraceptive Use as assessed in 1988," Population StudLes Series, No. 110, United Nations, N.Y., 1989; and (2) Family Planning and Chlld Survival: 100 Developing Countries, Center for Populatlon and Family Health, Columbia University, 1988. 8/ The Government has in fact accepted the cafeteria approach durlng the latest UNFPA cycle of assLstance. Whereas the IUD was almost the only method available before, access to both pLils and condoms has improved somewhat recently. However, there is considerable scope to further broaden the contraceptive method mix. - 17 - Table 2.4t Trends in Contraceptive Prevalence and Use of Contraceptive Methods, 1975-1988 Method 1975 1980 1981 1982 1983 1984 1985 1988 a/ Percentage of married women aged 15-49 using modern contraceptives (CPR b/)s 8 18 20 23 26 30 34 38 Percentage distribution of methods among users of modern methodas :ID 72 36 33 36 43 52 56 87 Condom 1 47 50 49 46 43 37 3 Pills 1 15 16 14 9 3 5 1 Sterilization 0.1 2 1 1 2 2 2 8 Others 25 NA NA NA NA NA NA 1 All methods 100 100 100 100 100 100 100 100 Notes: CPR by program methods was 45.2% in 1988. a/ Data from the 1988 Viet Nam Demographic and Health Survey have been adjusted to reflect use of modern contraceptives only. b/ Mission Estimates, except 1975 and 1981. Sources: MOH and VNDHS 1988 2.33 Another pattern that seems to be unique to Viet Nam is the wide practice of menstrual regulation and induced abortion, even though these are not considered as contraceptive methods by the Government. Induced abortion has been legal since the late 1960s. It used to be performed only in district - 18 - and provincial hospitals, although the service is now provided even in commune health centers. / Today, owing to the limited availability of alternatives, menstrual regulation and induced abortion have become popular and their use has increased sharply. According to the 1988 Demographic Survey, about 7% of the married women in the reproductive age groups had sought recourse to either abortion or menstrual regulation. Based on casual empiricism and anecdotal evidence, it appears that this rate itself may be an underestimate, and that menstrual regulation may be even more prevalent. 2.34 How does the method mix in Viet Nam compare with that observed in other developing countries? In Table 2.5, which reports the percentage distribution of contraceptive methods among currently contracepting married women in Viet Nam, Indonesia, and the Philippines, Viet Nam stands out for its high relative use of IUDs. While IUD use has been increasing over time in both Indonesia and the Philippines, it accounts for only 30.5% and 6.6%, respectively, of total contraceptive use in the two countries. In contrast, it accounts for 62.3% of total use in Viet Nam. Sterilization, which is very common in the Philippines and Thailand, does not appear to be a popular choice among Vietnamese (and Indonesian) couples. The low rate of sterilization use in Viet Nam is puzzling, since sterilization is a physician-intensive contraceptive method and there is no shortage of health workers in the country. While the lingerlng Catholic influence in the South might be expected to increase the resistance of couples there to use sterilization, the data in fact show greater reliance on sterilization in the South than in the North (Table 2.6). Perhaps, the long exposure to war and the consequent death of many young men have made Vietnamese women less enthusiastic about a permanent method like sterilization. Another possible reason for the lack of popularity of sterilization is that, although there are plenty of health workers in the commune health centers and intercommunal polyclinics who are competent in performing sterilizations, the surgical equipment and facilities and training for such operations are simply lacking at most health centers. At any rate, what is Important to note is that, although there is no single acceptable distribution of contraceptive methods from country to country, most couqtries typically have a range of methods that are used by contracepting couples. In Viet Nam, on the other hand, there is effectively no choice, as seventh-eighths of all married women currently using modern contraceptives use IUDs. }~Q/ Within 15 days from the time a woman has missed a menstrual period, she may seek help from the health post to "regulate menstruation." If more than 15 days, but fewer than twelve weeks, have elapsed, a woman can obtain an abortion if she makes a written request. Thereafter, abortion is normally not permitted, although exceptions are made when medically recommended, particularly for women with kidney or heart dise&se. - 19 - Table 2.5t Percentage Distributios of Contraceptive Use in Selected Asian Countries Indonesia Philippines Thailand Viet Method Nam 1976 1987 1978 1988 1978 1984 1988 Modern Pill 56.7 31.4 13.2 19.2 41.0 30.7 0.8 IUD 21.3 30.5 5.3 6.6 7.5 7.6 62.3 Female sterilization 1.1 6.9 13.2 30.5 24.3 36.4 5.0 Male sterilization 0.0 0.4 0.0 1.1 6.6 6.8 0.6 Other 0.8 21.8 0.0 0.6 12.9 14.6 0.6 % modern 79.9 91.0 31.6 57.9 92.3 96.0 71.5 Other traditional 20.1 9.0 68.4 42.1 7.7 4.0 28.5 Total 100 100 100 100 100 100 100 Sources: Indonesia Family Planning Perspectives in the 1990e, World Bank, Washington D.C., 1990, p. 90, Table 1.8; Philippines Family Planning Sector Report 1991, Annex 2, Table 2.3; Insti- tute for Population and Social Research, Mahidol University, National Family Planning Program, University, National Family Planning Program, Ministry of Public Health, Research Center of the National Institute of Development Administration, Thailand: Third Contracention Prevalence Survev, Survey Report, April 1985, Bangkok; Viet Nam Demographic and Health Survey 1988. Democraphic and Socioeconomic Differences in Contraceptive Use 2.35 Contraceltive Use By Age. The VNDHS data show an anticipated increase in the use of contraception with age. The use of contraception in the age group 20-24 is low, with only 31.7% of married women in this age group using contraception (and 19.7% using modern contraceptives) (Table 2.6). It is likely that contraceptive use in this age group would have been substantially higher if access to reversible and easily-usable methods, such as the pill and the condom, were readily available. 2.36 Contraceltive Use by Socioeconomic Status. Only two socio- economic variables, female education and urban/rural residence, were included in the 1988 VNDHS. Table 2.6 shows that more schooled women are more likely to use contraceptives. However, interestingly, the relative use of "modern" methods is lowest (viz., 66.7% of total) among users with the highest education (secondary schooling or more). Women with secondary schooling appear to rely much more on the natural rhythm method than women with less schooling, since education presumably makes it easier to keep a track of menstrual cycles. - 20 - Table 2.6: Contraceptive Use and Method Mix, by Age, Urban/Rural Residence, Regfon, and Education, 1988 Any Female Mole With- Not Methods IUD Pill Condom Steeri Sterlt Rhythm draw Other Usina Total An 15-19 5.26 3.51 0.00 1.75 0.00 0.00 1.75 O.00 0.00 94.74 100 20-24 31.72 19.09 0.16 0.49 0.00 0.00 4.85 6.96 0.00 68.28 100 25-29 52.17 34.34 0.10 1.21 0.50 0.20 7.75 7.65 0.40 47.83 100 30-34 59.84 37.81 0.34 1.90 2.35 0.11 9.62 7.49 0.22 40.16 100 35-39 68.85 41.42 0.88 0.88 6.02 0.71 10.09 7.96 0.35 31.15 100 40-44 65.43 38.30 0.53 1.60 5.59 0.80 11.17 6.65 0.80 34.57 100 45-49 47.07 29.01 1.02 0.25 5.85 0.51 5.60 4.33 0.51 52.93 100 ....... .................... ........................................................................ Urban 67.17 29.20 0.76 4.08 *6.81 0.76 17.85 7.72 0.00 32.68 100 Rural 50.25 33.96 0.37 0.56 1.82 0.22 6.03 6.86 0.43 49.72 100 ...................................................................................... North 58.68 47.17 0.19 0.72 0.72 0.00 4.45 5.12 0.29 41.37 100 South 46.83 17.29 0.71 1.59 4.87 0.66 12.14 9.14 0.44 53.01 100 ................... ........................................................................... Illiter 28.63 18.80 0.00 0.00 1.71 0.00 3.85 3.85 0.43 70.94 100 Read/ Write 45.55 26.34 0.64 0.64 3.44 0.25 7.51 5.98 0.76 54.33 100 Primary 55.42 36.21 0.48 0.88 2.47 0.35 6.92 7.80 0.31 44.54 100 Second. & above 63.76 35.91 0.33 3.13 2.80 0.33 14.66 6.59 0.00 36.41 100 Total 53.10 33.15 0.44 1.13 2.67 0.28 8.06 7.03 0.36 46.77 100 Source: Viet Nam Demographic and Health Survey, 1988. 2.37 Contraceptive use varies significantly by region. A much larger percentage of married women aged 15-49 use contraception in urban areas (67.2%) than in rural areas (50.3%). However, much, although not all, of this increase comes about because of the greater use of the natural rhythm method among urban women (who are generally more educated than rural women). Urban women also tend to rely more on female sterilization and condoms than rural women. Finaily, there are important differences between the North and the South in both contraceptive use and the method mix. The percentage of married women aged 15-49 using any contraceptive method is significantly higher in the North (67.2%) than in the South (50.3%). Furthermore, the method mix differs even more across the two regions; the predominance of IUDs in the method mix is observed to be mostly a Northern phenomenon. While 80.4% of contracepting women in the North rely on lUDs, only 36.9% of contracepting women in the South rely on this method. Almost one-half of the contracepting women in the South appear to rely on the natural rhythm and withdrawal methods. The source of the large North-South difference in the method mix is not clear. It is likely that IUDs are not as widely available in the South as in the North. In addition, since people have historically relied much less on the health system for their family planning needs in the South, they have had to depend to a greater extent on traditional contraceptive methods than couples in the North. Potential Demand for Family Planning 2.38 It is possible that the use of modern contraceptives is relatively low in Viet Nam because couples desire larger families. The desire for larger families may be driven by economic factors, such as the economic contribution of children to the household, or cultural considerations, such as a preference for sons. If this is the case, a family planning program is likely to have - 21 - limlted success unless it alters the structure of social and economLc incen- tlves associated with hlgh fertLlity. On the other hand, it iL also possible that, even when the social and economli conditLons in a country favor smaller familles, couples contlnue to have excess fertllity because of lack of contraceptlve awareness and poor access to contraceptlve methods. "Unmet need" is deflned ln family plannlng as the proportLon of couples who deslre to regulate their fertllity but are not using any contraceptives. Subject to the usual caveat of lnterpreting data on expressed fertility preferences,ll/ the analysis on unmet need ls of value in deslgnlng future family plannlng strategles, since it helps to ldentlfy both the future potential as well as the target groups of contraceptlve use.la/ 2.39 In the 1988 VNDHS, women who said they dld not want additional chlldren were assumed to want fertllity regulation. Unmet need ls then the percentage of women who did not want additlonal children, yet were not uslng any contraceptives. Table 2.7 shows the magnitude of unmet need for family plannlng to limit blrths. The unmet need ls substantial;l3/ almost 41% of the women surveyed who said thelr actual number of children was greater than their desired number and who did not want additional childrenlA/ were not usLng contraceptLves. To the extent that contraceptives are used not merely for limiting fertillty but also for spacing births, the estimate of unmet need is likely to be an underestimate. The VNDHS report does not provide informatlon on the use of contraception for spacing purposes.l5/ I/ Survey questions on fertility preferences are subject to controversy, because women in developing countries normally do not plan their family siLes or are not informed about how to affect the number of births they will eventually have. Socio-cultural factors and attitudes of husbands are important, but those are not accounted for. 12V In most countries, women respond fairly consistently to the first question, and their replies generally turn out to be good predictors of contraceptlve use and of future fertility norms. For further evidence, see Bryan Bouller, "Unmet Need for Contraception: Evaluation of Estimates for Thirty-six Developing Countries," World Bank Staff Working Paper, No. 678, Washington D.C. J2/ Since unmet need is not necessarily equal to demand, these measures are only indicative of potential users. }i/ In the VNDHS, the deslred number of children was not broken down by the desired sex distribution. Hence, a woman who already had the desired number of children but not the desired number of, say, sons could still want more children. flI As a result, the method of estimating total unmet need proposed by Bongaarts cannot be applied to the VNDHS data. See J. Bongaarts," The Measurement of Wanted FertLilty," Poculation and Development Review 16 (3), September 1990. - 22 - Table 2.7 Percentage of Currently Married Women Aged 15-49 Using Contraceptive Methods by Their Actual Number of Children and Desired Number of Children want More Want No More GrouD g2ing Not Usina Using Not Using A > D 29.41 70.59 59.06 40.94 A - D 73.33 26.67 71.66 28.34 A < D 39.47 60.53 37.45 62.55 Notes: A U Actual Number of Children D o Desired Number of Children Source: Viet Nam Demographic and Health Survey, 1988. 2.40 Table 2.8 reports informatlon on unmet need by age group. Over one-half of the married women in the 20-24 years age group not wanting additional children were not using any contraceptives. 2.41 Another way to assess the unmet need for contraception is to compare the difference between total wanted fertility and the actual fertility rate. Total wanted fertility is inferred from the information on whether a live birth was wanted or not. In principle, total wanted fertility measures desired fertility if all unwanted births are prevented. Two countervailing arguments concerning this measure are worth pointing out here. One argument argues that this measure has the advantage of being more realistic as it takes into account the fact that low fecundity prevents some women from having wanted births and from attaining their desired family size. The other argument maintains that the measure is vulnerable to transitory influences on the level of recent fertility.J&/ Subject to these qualifications, Table 2.9 shows that the total wanted fertility for women aged 15-44 was 2.5 children, but that their actual total fertility rate was 4.5 children. The difference between these two numbers is even greater in rural areas, where the total wanted fertility rate was 2.6 and the actual fertility was 5.0. Even if one is skeptical about the exact value of the wanted fertility rate, the sheer magnitude of the difference between wanted and actual fertility in the VNDHS sample suggests a very large unmet potential demand for family planning in Viet Nam. Interestingly, the average wanted fertility rate for the VNDHS sample of women (2.5) is close to the Government's stated two-child norm, which raises the possibility that survey respondents may have been influenced by government norms. ji/ See, for instance, J. Bongaarts and R. Potter, Fertility. Bioloqv and Behavior, 1983, Academic Press, New York, and J. Bongaarts," The Measurement of Wanted Fertility," Population and Development Review 16 (3), September 1990. - 23 - Table 2.8s Percentage of Currently Married Women Aged 15-49 Using Contraceptive Methods by Their Age and Their Future Birth Plan, 1988 Want More Want No More Age Group Using Not Using Using Not Using 15-19 6.12 93.88 0.00 100.00 20-24 28.23 71.77 48.15 51.85 25-29 44.92 55.08 59.91 40.09 30-34 49.39 50.61 64.87 35.13 35-39 40.66 59.34 74.67 25.33 40-44 30.77 69.23 69.59 30.41 45-49 17.65 82.35 48.66 51.34 Source: Viet Nam Demographic and Health Survey, 1988. 2.42 The question naturally arises as to why almost one-half of the married women in their reproductive ages who wish to have no more additional children are not contracepting? The 1988 VNDHS asked married (but currently not pregnant) women aged 15-49 who did not want any additional children, yet were not using contraceptives or abstaining from intercourse, why they were not using contraceptives. The most common reason (given by 37.8% of the women) for not using a contraceptive was that they were breastfeeding. Another 19.1% of the women cited "health reasons" for not using contraceptives, while 31.1% of the responses fall under the category "other reasons." Only 1.44% of the women cited unavailability of the desired method as a factor in not using any method. 2.43 The fact that a large proportion of women cited potential health risks in using contraceptives indicates an important role for contraceptive information and education. Although the proportion of women citing unavailability of the desired method as a reason for not using contraceptives is relatively small, the experience of other countries suggests that more couples desiring to regulate their fertility will use contraception once they have ready access to a variety of easy contraceptive methods, such as pills and condoms. - 24 - Table 2.9s Total Wanted Fertility and Total Fertility Rates for Women Aged 15-44, 1988 Total Wanted Total Actual Fertility Rate Fertility Rate Urban-rural Residence Urban 1.97 2.59 Rural 2.61 5.05 Reagon North 2.58 4.65 South 2.35 4.44 Total sample 2.47 4.52 Source: Viet Nam Demographic and Health Survey, 1988. Demand Proiections 2.44 The current rate of population growth and the total fertility rate in Viet Nam is appreciably higher than the Government's goal. It is also much higher than the fertility level desired by most couples. The demand for contraception will continue to expand in the coming decades, as the cohort of women entering reproductive ages increases. It is critical that such demand not be constrained by an inadequate supply of contraceptives. Otherwise, an important opportunity to significantly reduce population growth will have been squandered. 2.45 If the Government's goal of lowering the total fertility rate to 2.8 by the year 2000 (from about 3.7 in 1990) is to be met, the supply of contraceptives will have to almost double from current levels. The UNFPA has estimated that an additional 4.6 million users will be needed between 1991 and 2000 (implying an increase in the contraceptive prevalence rate from 55.9% to 68.8%) to reach the Government's fertility rate target (Table 2.10).17/ The UNFPA demand projections assume that, over the next 8 years, there will be a shift to a more balanced method mix -- one that will also include male contraceptive methods, and that the rate of induced abortions will fall by 50% by the year 2000 (Table 2.11). 22/ UWFPA, Prograame Review and Strategy Development Report: Viet Nam," New York, 1991. - 25 - Table 2.10, Projected Nwmber of Users (of all Contraceptive Methoda) and Contraceptive Prevalence Rate, 1989-2000 No. of Users (in Contraceptive Preva- Year '000) lence Rate (%) 1989 5,670.7 53.2 1990 5,973.9 54.5 1991 6,296.7 55.9 1992 6,640.4 57.3 1993 7,006.7 58.7 1994 7,397.0 60.2 1995 7,733.7 61.4 1996 8,088.7 62.6 1997 8,463.3 63.7 1998 8,858.7 64.9 1999 9,276.3 66.1 2000 10,906.2 68.8 Source: UNFPA, "Programme Review and Strategy Development Reports Viet Nam", Hanoi, 1991, Table 1, p. 78. Table 2.11: Proposed Shift in Contraceptive Method _ix, 1989 to 2000 Method 1989 2000 Male Sterilization 0.6 3.0 Female Sterilization 5.0 6.0 Injectables 0.0 10.0 IUD 62.4 30.0 Fills 0.8 20.0 Other 29.0 6.0 Condom 2.2 20.0 Implants 0.0 5.0 Total 100.0 100.0 Source: UNFPA, "Programme Review and Strategy Development Report: Viet Nam," Now York, 1991, Table 2, p. 79. - 26 - D. Supplv Of Family Plannina Services Delivery System 2.46 The provision of family planning information and services is integrated with the provision of basic health care services. Most of the family planning services in the country are provided through the network of 10,000 commune health centers, although not all of the commune health centers or intercommunal polyclinics provide family planning services. Currently, there are a total of 2,678 government facilities that dispense family planning services in the country. Few of thee.- facilities, however, offer a range of contraceptive methods to potential users. For instance, it is estimated that only about 500 of the family planning facilities carry any supply of pills. Fewer than 50 facilities nationwide offer injectables, while implants are handled by only 3 facilities in the entire country. 2.47 These centers were originally developed to provide basic health services; with the introduction of family planning program, these centers were used as a convenient delivery facility. In addition to their regular health care functions, commune health center staff provide information and counseling, including premarital counseling, on various methods of family planning, and attempt to motivate the couples to accept and practice family planning. In addition, the health staff distribute condoms and contraceptive pills (when available), perform menstrual regulation, follow up regularly on those women who have agreed to practice family planning, and provide prenatal pregnancy care and delivery of babies. Abortions, sterilizations (tubal ligations or vasectomies) and IUD insertion are normally provided at the secondary or tertiary level of the health care system (at intercommunal polyclinics and the district or provincial hospitals). The higher-level establishments also offer the same basic family planning services provided at the commune level. Since the higher levels usually are better provided with facilities, equipment, supplies and trained staff than the lower levels, people are more likely to go first to a higher-level facility if one is available nearby. 2.48 Mobile units are sometimes used to deliver family planning services in rural areas, particularly in the slack agricultural season. Services provided by these units include gynecological examinations, IUD insertions, and the distribution of condoms and pills. The mobile units are capable of performing menstrual regulation, abortion and sterilizations as well. Equipped with audiovisual aids and generators (for areas where no electricity is available), mobile units are also engaged in the Information, Education and Communication (IEC) activities. The use of the mobile units was more common during the 1970s, when the health network was less well-developed. Their use has declined in the recent past because virtually every one of the district hospitals now has a complete family planing unit. They might, however, still have a role in disseminating knowledge about newly-introduced family planning methods, creating awareness among youth, and reaching remote villages, especially in the mountainous and highland regions of the North and the Center. - 27 - SuoDly. Pricinc, and Availability of Contraceotives 2.49 Sources of SupplV. The Government health sector is the major -- and, until recently, the exclusive -- provider of contraceptives in Viet Nam. The 1988 VNDHS indicated that 45% of the women who had ever used contraception had obtained their supplies (or received clinical services) from commune health centers. Another 37 and 12%, respectively, listed intercommunal polyclinics and district hospital. as their sources of supply (Table 2.12). Recently, the private sector has begun to play an important role in supplying certain types of contraceptives; 31% of the pill users and 17% of the condom users in the VNDHS survey reported obtaining their supplies from commercial sources. Of course, since these methods account for a very small fraction of total contraceptive usage, the role of the private sector in terms of the total extent of contraceptive coverage provided is insignificant. However, these results indicate that there may be substantial scope for expanding the use of pills and condoms by encouraging their distribution through commercial retail outlets. Tabte 2.12: Distribution of Sources of Domestic Supply of Contraceptives by Current Users (Women aged 15-49 years), 1988 ------Supply MethodsC --l- --------------------Ctinic Methods.----------------------- Sub IUD Femle Mate MR Abortion Sub Source Pill Condom Total Steri Steri Total Commune HC 14.7 22.0 18.8 54.0 5.4 14.5 3.6 44.9 District PC 23.9 24.8 24.4 33.1 22.1 22.3 57.7 74.1 36.8 Province H 20.0 12.5 15.7 8.3 57.8 33.1 18.9 18.3 12.3 Central H 2.4 1.8 2.0 2.4 18.4 20.0 4.4 4.1 3.6 Other H Inst 4.6 19.7 13.1 1.8 0.6 3.1 1.4 1.6 Private Mrk 31.8 16.8 23.3 0.1 0.9 0.1 other 0.8 0.4 0.1 0.9 0.1 Donut Know 1.3 1.3 1.3 1.1 5.4 0.1 Total 100 100 100 100 100 100 100 100 100 Source: 1988 VNDHS 2.50 Pricina. Although family planning services are free throughout Viet Nam, the sale of birth control devices is permitted. In some provinces, hospitals have begun charging token fees for certain services since May 1989. For instance, the provincial health services of Ho Chi Minh City has introduced a system of incentives and disincentives to users. In 1990, incentives included (a) a payment of 5,000 dong per woman for IUD insertion, and (b) a payment of 20,000 dong per acceptor for vasectomies and tubal ligations. Disincentives included: (a) no salary supplements for workers having more than two children, (b) a charge of 3,000 dong for each delivery beyond the second; (c) all families, regardless of size, now receive 35 square meters of living space. 2.51 Imports. At present, most contraceptives are imported by the UNFPA at the request of the Government. Table 2.13 shows the quantities of pills, condoms and IUDs imported by the UNFPA during the last seven years. Two points need to be made about these figures. First, contraceptive supplies by the UNFPA have fluctuated considerably from year to year. There were no - 28 - IUDs received ln 1988 and 1989 and no pills received in 1988.18/ Second, since the UNFPA assisted the government in setting up local condom manufacturing facilities in 1987 (see para 2.52 below), it discontinued importing condoms for the government from 1988 onwards. Table 2.13: UNFPA Contraceptive Supplies Year IUD Pill Condoms (pieces) (cycles) (gross) 1984 186 200 110 1985 1,305 1 1986 1,171 300 15 1987 339 1,168 55 1988 1989 500 1990 1,000 1,800 j 1991 550 3,500 Source: UNFPA, Hanoi 2.52 Progducton. The Government, with assistance from the UNFPA, started the domestic production of condoms in 1987. The Medical Rubber Factory (MBRUFA), located in Ho Chi Minh City, has a capacity of producing 90 million condoms a year. Initially, the imported raw materials, packaging materials and quality control equipment were supplied by the UNFPA with co- financing from Australia. The factory was to be supported by the Government after receiving operating support from the UNFPA for the first two years after its establishment. However, the Government's 1989 directive that state enterprises be financially independent put MERUFA in some difficulty, and the enterprise cut back on quality control and switched to domestic sources of raw materials and packaging. As a result, quality of condoms produced by MERUFA suffered. More recently, however, the enterprise has been able to improve the quality of locally-produced condoms substantially. Indeed, recent independent tests show that not only do MERUFA condoms meet international quality standards, but that they are as popular among users as imported condoms. 2.53 In the past, MERUFA has generally operated well below its potential capacity (in 1990, 60 million condoms were produced) because of difficulties ln materials, distribution and quality. While there has been a significant improvement in the quality of MERUFA condoms in recent years, there has also been a major change recently in the way condoms are procured by the government, and this may create new distribution problems for MERUFA. These interruptions in supply must surely have affected the population's access to contraceptives. However, it is not clear how the supply disruptions affected the contraceptive prevalence rate estimates obtained by the 1988 Demographic and Health Survey. - 29 - Until 1992, the MOH purchased condoms from MERUFA for distributLon in the dlfferent provLncee. Slnce 1992, the central government has delegated the condom procurement decisLon to the provlnces, whlch means that MERUFA wlll now have to deal with 53 separate cllents instead of one major client. 2.54 The Government was seeklng UNFPA assistance for settlng up a local IUD production facility. But thli project was eventually not undertaken, as it was declded that UNFPA would contlnue to import lUDs on behalf of the Government. The Government also has plans for the local productlon of contraceptive pills. However, these plans are as yet uncertain and contlngent upon the availability of external fLnancing. B. Famlly Plannina Exoenditure 2.55 Table 2.14 shows the trends in real government famlly plannlng expenditure between 1984 and 1990 (with Tables 2.2, 2.3 and 2.4 ln the StatLstical Annex contalnlng the details). Before analyzLng these data, it is important to note that budgetary data are perhaps the least reliable of the statistics collected and reported by the Government. In 1985, for example, Viet Nam was experiencing hyperinflatLon, and the Government attempted a dLsastrous ten-for-one currency swap, introduclng the "new dong." Reported government expendlture in all categories is much higher in 1985 than in neighboring years. Hence, the precipitous fall in real family planning expenditure from 1985 to 1987 is probably a statistical artifact. IPvt. Hesth 6 FSiy PInig E., 1981-90 Indeed, what is surprising 280a- is the sharp increase (of over 100%) ln real family 240- planning expenditure between 1987 and 1990 (Figure 2.5). 200 - Again, these numbers cannot / \ be taken at face value, 1 0- since the method of paying all government workers -2 - changed in 1989, wLth -22 monetary remuneration \17 increasing and payments in 0 kind (mostly rice) to decreasing. Since pre-1989 t984 2995 196 1987 lgh8 l989 19gO government expenditure data Yw did not include the implicit value of Ln-kLad payments to Figure 2.5 government employees, it is likely that post-1989 government expenditures are artlficially inflated. 2.56 The most llkely possibillty is that real famlly planning expendlture per capita (adjusted for the fact that rice subsidies for public employees were replaced by wage lncreases) dld not change much between 1985 and 1990. But, if true, this trend ls in itself laudatory, since Viet Nam was experiencing hyperinflation, acute macroeconomic instability, and a sharp drop in total external aid during this period. While there was a large decline in - 30 - total external aid after 1989, most of that aLd was from the Eastern bloc and included little for the family planning sector. The modest resumption of Western aid in recent years, however, has included a large component for the social sectors and for family planning in particular. Therefore, it is possible that the Government of Viet Nam has been able to maintain family planning expenditure largely because of external assistance from Western donors and the international agencies. Table 2.14: Composition and structure of Family Planning Expenditure, 1985-90 Variable 1985 1986 1987 1988 1989 1990 As % of GDP 0.03 0.05 0.02 0.03 0.04 0.04 As % of Total Govt. Expenditure 0.41 0.40 0.18 0.15 0.19 0.21 Real Per Capita exp. (1982 dong) 0.36 0.21 0.10 0.12 0.21 0.19 Souce: Statistical Annex Tables 2.2, 2.3 and 2.4 2.57 The overall structure and evolution of family planning expenditure are shown in Tables 2.1, 2.2 and 2.3 in the Statistical Annex. There are three observations that follow from these data: first, government expenditure on family planning in Viet Nam is modest as compared with other Asian countries at similar stages of development. For instance, in Bangladesh and Nepal -- countries that have roughly comparable income and contraceptive prevalence levels -- family planning accounted for nearly 2% of total govern- ment expenditure over the period 1985-88. In Viet Nam, it accounted for between 0.15 and 0.41% of total government expenditure over the same period.19/ Second, the share of total family planning expenditure spent on salaries of service delivery personnel has been falling over time, while that spent on administrator salaries has been either flat or increasing. Insofar as service delivery personnel have the greatest influence on the quality of services dispensed to clients, the declining allocation to delivery personnel salaries is likely to have adverse effects on the quality of family planning services. Third and finally, the share of contraceptive procurement and supplies in the family planning budget has been falling over time. Only 5.8% of the family planning budget was devoted to contraceptive supplies in 1990 -- down from a high of 15t in 1988. Reduced budgetary allocation to contra- ceptive procurement has coincided with the severe shortage of contraceptive supplies being experienced throughout the country. Because there is a large unmet demand for contraceptives in Viet Nam, the shortage of pills, condoms and other safe birth control methods has forced women to resort to induced abortion and menstrual regulation. Also, disruption in the supply of oral contraceptives and condoms has contributed to an excessive reliance on IUDs. Indeed, a deceleration in key demographic parameters, such as the contra- ceptive prevalence rate (CPR), the volume of new family planning acceptors, 12/ Of course, what is more important than the amount spent on family planning is the effectiveness of funds spent. It is difficult, if not impossible, to compare this variable across countries. - 31 - and the rate of continuation of use, in recent years provides additional evidenc- that reduced expenditure on contraceptive procurement and supplies has had undesirable consequences. F. External Assistance 2.58 UNFPA. UNFPA has been the main external donor agency assisting the family planning program in Viet Nam since its inception. From 1978 to 1991, its three country programs (1978-83, 1984-87, and 1988-91) have amounted to a total of more than US$50 million in aid. The level of assistance will be increased sharply in the Fourth Country Program (1992-95), for which UNFPA has already committed itself to providing US$25 million from its own resources and to raising another US$11 million from other bilateral sources. Already in 1992, more than 1.7 million IUDs and 4.5 million pill-cycles have been either supplied or ordered by UNFPA. These numbers represent a sizeable increase over UNFPA contraceptive supplies in previous years. In addition, UNFPA will be providing equipment and health-worker training for sterilizations. 2.59 UNFPA's assistance to Viet Nam has primarily been in three areas: (a) Maternal and Child Health care and Family Plannina 1MCH/FP) has accounted for more than one-half of the total resources spent by the UNFPA in Viet Nam. Supported projects include the training of health workers, improving and strengthening the delivery of MCH/FP services, financing the production of contraceptives (mainly, condoms), the importation and distribution of contraceptives and drugs, and research in human reproduction. (b) Information, Education and Communication (IECI Proiects have accounted for about 15% of the total UNFPA support in the third country program. Supported projects include population education, family life and sex education, education of the parents of kindergarten children, and programs for the Vietnamese Women's Union and the Vietnamese Youth Union. (C) Basic Data Collection accounted for about 19% of total UNFPA support. The two most successful data collection projects were the 1989 Census on Population and Housing and the 1988 nationally-representative Demographic and Health Sample Survey. The UNFPA is financing three projects related to the Census, various demographic research projects, and the establishment of a system to provide the Government with intercensal information on population. 2.60 Although some of the UNFPA-assisted activities are national in scope, those with a regional focus are being carried out in eight target - 32 - provinces which together account for about 25% of the population of Viet Nam.2Q/ The eight provinces are characterized by high birth rates and are mostly rural. 2.61 Other Bilateral and NOs. Viet Nam also receives flnanclal assLitance for family planning actLvities from other donors. The AustralLan government has provlded some support for the condom manufacturlng factory (joLntly wlth the UNFPA). Several NGOO are also actlve ln the family planning area. For example, the Internatlonal Planned Parenthood Federation contrlbutes US$200,000 annually ln equipment and contraceptlves to lts local afflilate, the Vlit Nam Family Planning AssocLatLon (VINAFPA). 2.62 UNFPA uses external NGOs for the executlon of several of its projects. The Program for the Introductlon and Adaptation of Contraceptive Technology (PIACT) has assisted in retraining 12,000 health workers ln IUD ±nsertLon, lncludlng the productlon of approprlate IZC materLals. The Program for Approprlate Technology ln Health (PATH) has provlded the technical and managerial support to some UNFPA-supported projects. The Australlan National Unlverslty is the executlng agency for a UNFPA project with the InstLtute of Sociology to strengthen social science and demographic research. JOICFP is executing a UNFPA-supported integrated family planning project. 2.63 In the next cycle of projects, beginnLng in 1992, UNFPA hopes to call on varlous NGOs to assist wlth social marketlng, IEC, community based programs, contraceptive supply, and health-worker traLnlng ln demography. It contlnues cooperatlng wlth the lncreasing number of NGOs (the Population CounclI, International Famlly Planning Association, the Population Crisis Committee, etc.) who have been sendlng missLons to Viet Nam since the end of 1988 to explore project development and collaboration. 0. Issues and Recommendations 2.64 Vlot Nam has experienced a remarkable decline in fertillty, in part because of a sharp fall in infant mortallty but also ln part to a strong governmental commitment to the family planning program. Interprovincial analysi of total fertility rates in 1979 and 1989 suggest that low fertllity rates are strongly associated with low infant mortality rates and high levels of per-caplta provinclal domestic product and female literacy (Annex A). These are the factors that drlve the demand for smaller familles. But the same analyaLs shows that provinces that provide better access to health facilities and that generally have hlgher levels of publlc spending on health and family plannlng tend to have significantly lower fertillty rates. These results are important, slnce they suggest that further declilnes in fertility can be achLved best by a combination of pollcLes that seek to Lnerease the demand for smaller families among marrled couples and, at the same time, improve the supply of famlly planning servLces. These Lnclude pollcLes that emphaLse and promote economlc growth, general Lmprovements Ln health (that result ln a further reductLon Ln Lnfant mortalLty rates) and education (viz., 20/ During the thlrd cycle of assiLtance, UNFPA provlded support to eight provinces. However, according to the new diviLson of provlnces Ln VLet Nam (whieh results Ln the country having 53 -- iLntead of 44 -- provinces), UNFPA has been providLng support to ten provinees. - 33 - female literacy), and specific improvement, i the access to and delivery of family planning services. 2.65 A great deal of the credit for falling fertility rates goes to the Government, which has greatly reduced illiteracy and has established an extensive and comprehensive family planning program. Although analysis of government expenditure in Viet Nam over time is fraught with methodological problems, it appears that in recent years the Government has maintained the level of real family planning expenditure despite a deepening macroeconomic crisis. The success of the Government's effort is apparent in the very high levels of contraceptive awareness in the country and in the large numbers of women who would like to regulate their fertility. In fact, the Government has been slow to realize its own accomplishments and the magnitude of the fertility transition that has taken place in Viet Nam in recent years. The pronouncements of the Government's family planning program continue to be based on the premise that Vietnamese couples can experience further fertility declines but will have to be goaded into birth control. The reality is otherwise. Available evidence suggests that there is already a large latent demand for fertility regulation and birth control among Vietnamese couples; this demand is currently constrained by the inadequate supply of contraceptives and the virtual absence of contraceptive method choice. The Viet Nam Demographic and Health Survey found a significant proportion (over one-half in some cases) of married women, especially in younger age groups and In rural areas, expressing a desire to control births were not using contraceptives. It is important that this extraordinarily high level of unmet need for family planning not be restricted much longer. 2.66 Since the provision of family planning information and services is integrated with the provision of basic health services in Viet Nam, many of the issues relating to the reform of the health care system -- discussed in Chapters IV and V -- are relevant to the family planning program as well. In addition, the discussion in this chapter suggests the need to: (a) rationalize government expenditure on family planning, (b) expand the supply and improve the distribution of contraceptives, (c) increase the variety of contraceptive methods available to couples, (d) enlarge information, education and communication activities related to family planning, (e) retrain health workers in newer family planning methods as these methods become available in the country, and (f) develop a management information system for collecting and monitoring the progress of the family planning program. 2.67 Famly Planning Expenditure. The share of family planning in total government expenditure appears to be significantly smaller in Viet Nam than in some other countries at its income level (such as Bangladesh and Nepal).ZV Of course, the effectiveness of funds spent is more important to program success than the amount spent, but it is unlikely that Vietnamese family planning expenditure is significantly more effective than that of, say, Bangladesh. An even more disturbing trend is the evolving composition of family planning expenditures. The share of contraceptive supplies in the total family planning budget has been falling, which in turn may be aggravating the shortage of contraceptive supplies in the country. In IV Of course, these countries have significantly higher levels of total fertility. - 34 - additlon, the share of total family plannlng expendlture spent on salarles of servlce dellvery personnel has been falling relative to administrator salaries. insofar as servlie delivery personnel have the greatest influence on the quallty of servLeso dlipensed to cllents, the latter trend does not bode well for lmprovlng the quallty of famlly plannlng servloes. 2.68 ContraceotLve Method Mix. Although Viet Nam subscrLbes to a cafeteria approach, supply constralnts for other contraceptlves and the preference of health and family plannlng workers for the WUD (because of lts durability of protectlon and easier monltorlng) have made IUDo the only method avallable to couples, with menstrual regulatLon as a backup ln the event of method fallure. A true cafeterla approach that provldes a varlety of contraceptLve methods ln the famlly plannlng dellvery system, lncludlng male contraceptlve methods, could signlfleantly lmprove both the coverage and the effectlveness of famLly plannlng LnterventLons. Evldence from other countrles suggests that the preferred birth control methods would be plls and condoms, especially for newly-marrled and younger women lnterested in delaying thelr flrst pregnancy or spailng their children. 2.69 In order to alleviate supply constraints and improve the mix of contraceptive methods, the donor communlty may have to take on the responsibility of provldlng Vlit Nam wlth oral contraceptives, condoms, IUDs, injectables and implants as requLred, until such time as the Government is able to allocate the necessary forelgn exchange to import these supplies. Currently, UNFPA is the only donor agency provldlng contraceptlves to VLet Nam. In addltlon, prlvate traders could be encouraged to purchase those methods that are ln short supply or that cannot be provided by the official program. The evldence Lndicates that the inc$pient private sector is already playing an lmportant role in the provlolon of oral contraceptives and condoms, especially ln the South. 2.70 Current practlces of contraceptive procurement, storage and dLstrLbutLon in the country are far from adequate. Technlcal assistance from donors, as well as substantlal investments for upgrading warehouse, equipment and transport, may be needed to deslgn better logLitLcs for delivering contraceptlves nationally. The participation of the private sector and NGOs could ease the financial burden of developlng the delivery system to a reasonable extent. In thls respect, the NCPFP needs to establish a strategic approach to social marketing polLcy that would provlde guidance on how contraceptLves would be promoted and sold in the private sector. Obviously, the number of couples able or willing to pay commerclal prlces will be limited, so pilot efforts to test the effects of subsidised and non-subsldized prices could provlde useful lnsights into prospects for expandlng the commercial sector. 2.71 Information. Education and CommunLcatLon (tIC). In Viet Nam, populatlon educatlon has made sLgnlficant strLdes. But the fact that more than one-thlrd of the married women in thelr reproductLve ages not using contraceptives elted potential health risks for not using contraceptives Lndlcates the need for greater Lnformation and educatlon. The present coverage of IZC campaigns li not wlde because of lack of materials, maLntenance of media equipment and retralnlng of IEC staff. The Government is fully aware of thli sLtuatlon and deoplte the fact that the NCPFP has allocated 25% of lts budget to IEC actlvitles in 1991, famlly planning - 35 - motivation still poses a serious challenge to the policy makers. Although the UNFPA provided assistance in the past, the funds were limited, with the result that the IEC program could not be expanded beyond eight provinces. Further technical assistance in developing messages, IEC training, procuring materials and equLpment for medla might be required. 2.72 Trainino. Budgetary support for training in MCH/FP activities dropped steeply during 1985-87 and its share with respect to the total budget remained low (Chapter V). The recently completed Report, "Health Manpower Development Plan -- Viet Nam, 1996-2005," prepared by the MOH, identifies "refresher training" needs for 25,000 workers in MCH/FP and outlines a training plan. At the commune level, the current training curriculum for midwives and nurses needs further reviews, so that they can have added skills in providing better ante-natal care, family planning services, child care and the like. Likewise, with each curriculum, appropriate teaching materials should be designed in tune with the working conditions of Viet Nam (diverse ethnicity, languages, dialects, etc). To conform with the national standard, these materials could be prepared centrally and distributed through secondary medical schools to all trainees. 2.73 Consistent with the Manpower Development Plan of the MOH, the external donors should aim at providing institutional support for the Training Department in the MOH in order to (a) improve planning for MCH/FP manpower development, (b) develop a retraining system, (c) upgrade pedagogical skills of trainers, (d) prepare curricula with changing demand, and (e) enhance printing and distributing training materials. The Swedish International Development Agency (SIDA) is planning to provide a senior long-term resident adviser on training to the MOH, who will help the training unit of the Ministry in preparing curricula, a training plan, and materials. UNFPA and UNICEF are also working with SIDA revising training materials and curriculum for MCH/PF training. To complement this, the following modes of support from other active and potential donors will be relevants provision of equipment, including transport to support MCH/FP training programs in the districts and communes, reprographic equipment, video facilities, short overseas training for main MCH/FP personnel, and financial support for training activities carried out by other training institutes. 2.74 Manacement Information System (MIS). The current system organizes the collection and analysis of service data from communes, districts and provinces needed by managers to make decisions. But feedbacks from managers to lower levels are slow and managers cannot respond better to local needs. Commune health centers maintain statistics on the number of users and type of contraceptives used, the characteristics (e.g., age, parity, education, etc) of the users, and also the eligible women in the commune who are not using any contraceptives. This information is transmitted to the district level and then onto the provincial and central levels. While such information and its transmittal is potentially useful, it is not employed effectively, as the higher levels often lack the capacity to analyze the data in a functional way. - 36 - 2.75 An important lacuna in the data collection system is the lack of a mechanism to regularly collect information on the time spent by MCH/FP workers in various activities and on worker performance and productivity. Such statistics, which could be obtained from time-and-motion studies, are essential for better overall management of the family planning program, since they would enable managers to swiftly reassign health workers and reallocate staff time and resources in response to changing demand conditions. They would also allow managers to make decisions on salary, incentives or compensation based on worker productivity. 2.76 Broadly, the management information system should be streamlined according to the following steps in order to make better use of information and have timely decisions from managers. First, there should be "standard" register books on (a) continuing users and their characterlstics, (b) eliglble couples not using contraceptives and their characteriatics, (c) maternal chlild health (e.g., BPI, oral rehydration, etc.), (d) births and deaths, and (e) inventory of contraceptives and equipment. Second, performance indicators should be developed jointly by workers and managers for measuring and report- ing performance in accordance with achievements of pre-determined objectives. Third, there should be a systematic mechanism for monthly transmiselon of such reports to the next hierarchy. Fourth, each supervisory level should have the capacity to analyze and interpret such reports, and make decisions on perfor- mance. Fifth, there should be a mechanism for feedback of comments, including support for resolving problems, from managers at each level to workers at the next lower level. 2.77 Strengthening the system nationwide will be a complex task, and its benefits in terms of raising operational efficiency take some time to materialize. Nevertheless, immediate action in this area is warranted, and the eventual payoff will justifies the initial investment. - 37 - III. NUTRITION A. Child and Maternal Nutritional Outcomes 3.1 Malnutrition Among Children. Child malnutrition rates in Viet Nam are high relative to its infant mortality rate. Results from the recently analyzed General Nutrition Survey (GNS) data show that, among children under six years of age, the percentages of moderately and severely malnourished children are 45% for weight-for-age, over 56.5% for height-for-age, and 9.4% for weight-for-height (Table 3.1).22/ Malnutrition rates are much lower in the first year of life (viz., 24.5% for weight, 41.3% for height, and 5.7% for weight-for-height), but increase sharply in the second year of life, after which they remain more or less constant.22/ Data also suggest that urban children are better nourished than rural counterparts. A strong seasonal pattern to malnourishment exists in rural areas, where substantial numbers of children are hospitalized with ;evere malnutrition from May to October. This is when the old crop has been consumed but the new crop not yet harvested. However, Vietnam's major child nutrition problem appears to be stunting from longer-term, chronic undernutrition rather than wasting from short-term, acute food deficits. Malnourishment for a significant proportion of children begins in the first year of life. Reasons for this may be low-birth weights (20% of the infants born in 1990 were estimated to weigh less than 2,500 grams at birth), sustained and nurtured by inadequate breast-feeding and complementary feeding practices. For many children malnutrition sets in during weaning when breast milk intakes decline sharply and adequate complementary feeding is crucial for growth. That problem may be further complicated by premature introduction of weaning foods. 3.2 International Comparisons. If the results from the 1987-89 GNS accu- rately represent the nutritional situation in the country, Viet Nam has a higher proportion of underweight and stunted children than almost any other country in South and Southeast Asia, excepting Bangladesh and possibly Myanmar (Table 3.1). Surprisingly, the magnitude of wasting is low -- comparable to levels observed in Thailand and the Philippines. 22J The reference that is used for malnutrition here (as in most other studies) is the United States National Center for Health Statistics (NCHS). The percentage of children whose anthropometric indicators are more than minus two standard deviations from the NCHS mean level are considered malnourished. 2/ These trends are similar to those found in malnourished children elsewhere (WHO, 1986) although height-for-age usually levels off after 3 years. - 38 - Table 3.1: Magnitude of Malnutrition in Selected Asian Countries Percentage of Children (1980-87) suffering from: moderate and se- moderate and se- moderate and se- vere underweight vere wasting (12- vere stunting Country (0-4 years)a 23 months)b (24-59 months)0 Myanmar 38 17 75 Indonesia 51 17 Bangladesh 60 17 59 Philippines 33 7 42 Sri Lanka 38 19 34 Thailand 26 10 28 Viet Namd 45.0 9.4 56.5 Notes: 8 Percentage of children with greater than minus two standard devia- tions from the 50th percentile of the weight-for-age reference population (US NCHS). b Percentage of children with greater than minus two standard devia- tions from the 50th percentile of the weight-for-height reference population (US NCHS). 0 c Percentage of children with greater than minus two standard devia- tions from the 50th percentile of the height-for-age reference population (US NCHS). d Data are from the General Survey of Nutrition and refer to the period 1987-89. Source: UNICEF, The State of the World's Children 1990, Oxford University Press, New York, 1990. 3.3 Time Trends. Various household surveys undertaken by the National Institute of Nutrition (NIN) allow us to trace the magnitude of malnutrition over time. Weight-for-age of young children appears to have improved between 1982-85 and 1986-87, with the improvement being more marked in the 12-35 month ages (Table 3.2). Between 1986-87 and 1987-89, however, both weight-for-age and height-for-age of younger children (0-11 months) deteriorated significant- ly, while the malnutrition rates of older children did not change appreciably. A parallel increase in low-birth weights is also indicated at this time (see discussion below). Reasons for the apparent retrogression in the nutritional status of younger children warrant further investigation. The downturn may reflect the fact that data from the third NIN survey are not strictly comparable to earlier data. In addition, 1987-88 was an exception year because of a poor crop and overall lower distribution of rice by the Government. But it is also possible that the deepening macroeconomic crisis during the 1984-88 period may have contributed to the deterioration in nutri- tional status. Without additional information, it is difficult to substantiate the downturn in nutritional status since 1986-87. - 39 - Table 3.2: Pereage of Chrm under S yeani of age maourised, 193249- 198245 198647 198789 % malnourished % malnourished % manorished AgP No. of No. of No. of (moP chil- Weight Height Weight dren Weoight Height Weight h Wi ight Height Weight (Months) dren -(or- -for- -for- dren -for- -for- -for- dren -for- -for- -for-, go AP height age age height age age hei S.7 0-11 1,955 21.7 21.5 3.9 2,978 19.6 28.2 7.3 1,153 24.5 41.3 12-23 2,737 59.0 61.3 11.7 2,782 45.3 50.0 10.1 1,297 43.8 S7.9 10.9 24-35 2,297 62.0 64.7 8.5 2,888 50.4 55.4 9.9 1,4S4 51.A 55.1 9.7 36-47 2,288 56.9 68.3 4.4 3,012 46.9 56.8 8.9 1,S57 49.8 60.0 9.5 48-59 2,S32 55.0 71.8 5.2 3,213 47.1 55.6 7.6 1,583 50.3 64.5 10.S 0-S9 11,80 52.2 59.7 7.0 14,873 41.8 49.1 8.7 7,044 45.0 56.5 9.4 La Figurem in the table are perentas of children with-for-age, height-for-age, and weight-for-height below two sandard deviations as compared to the NCHS reference cohoft. Source: NIN, various surveys. 3.4 Low Birth Weiaht. Despite the crucial role women play in agriculture, women had no real status in the village and the family prior to 1954. After the Communist party assumed power, women could, for the first time, hold land in their own names and were elected to leadership posts during the land reform period. However, the gains that women have made under socialism may be threatened by the new contracted system. In addition, a seriously unbalanced sex ratio from years of war and extensive male migration has increased the economic burden on women. The necessity to manage full time work and child care may have compromised women's health under the new system.24/ 3.5 A measure of maternal nutritional status is the prevalence of low birth weights, viz., the proportion of infants born with a weight under 2500 gme. The Government of Viet Nam reported in 1982 that about 8% of the babies ZA/ C. P. White, SocLalLst Transformation of Agriculture and Gender Relations: The Vietnamese Case," IDS 8ueti.n 13 (4): 44-51. - 40 - were born with low birth weight.2$/ In 1985 and 1987, the MOH estimated the proportion of low birth weight infants to be 18 and 14 percent, respec- tively. Birth reports a figure of 20% in 1990.2&/ If the proportion of low birth welght babies has indeed increased, as the various estimates suggest, it represents a serious decline in the health status of women. This may also explain the increase in malnutrition among infants between 1986-87 and 1987-89 (noted earlier). The GSO does report an attrition in health services directed to mothers and children during 1988-89.7Z/ 3.6 Other evidence also suggests that Viet Nam has a high ratio of low birth weight babies relative to its infant mortality and other indlcators. A review of birth weights in Hanoi from 1976 to 1986 showed that about 21.7% of the infants were born with low birth weight.28/ Estimates from two hospitals in Ho Chi Minh City indicate the proportion of low birth weight babies to be 7.8% and 21.6% of all birthe.29/ 3.7 Reaional Differences. Table 3.3 presents data on some of the key indicators associated with child nutritional status in 1987-89 for the eight ecological regions of Viet Nam. The data show substantial differences in average weights and heights of children under 5 years and the magnitude of malnutrition among children across ecoregions. Generally, the Red River Delta, Central Coast of Northlands, Central Coast of Southlands, and the Central Highlands show the highest rates of malnutrition, while the Mekong River Delta and the cities of Ho Chi Minh and Hanoi show the lowest rates. Surprisingly, the North Mountain and Midlands appear to be better off than many other regions in terms of malnutrition rates. The low malnutrition in the mountcinous areas of the North is counter intuitive, since this region has among the lowest agricultural productivity and per capita incomes and the highest infant mortality rates in the country. 3.8 There might be three explanations for the low malnutrition rates in the North Mountain and Midlands. First, that the NIN survey in the mountainous areas of the North selectively sampled the more accessible households close to main roads -- a fact acknowledged by the National Institute of Nutrition.30/ Since households residing in the accessible 25j T. Brun, "Food Consumption and Nutritional Status in the Socialist Republic of Viet Nam," Nutrition Consultant's Report Series 82, FAO, Rome, 1990 (mimeo). ZiO C. Birt, "Eutabliahment of Primary Health Care in Viet Nam," British Journal of General Practice 40: 341-344. f2/ General Statistical Office, Statistical Data of the Socialist Re_ublic of Viet Nam 1976-1990, Hanoi, 1991, p. 159. aft/ Brun, op. cit. 21J Brun, op. cit.-, p. 37. 30/ Other than agriculture, households in the mountainous regions are involved in trade, both legal and illegal, of wood and other products across the border with China. - 41 - areas of the mountain regions generally are better off than those in the remote areas, the estimates of food consumption and malnutrition recorded for this selected sub-sample may not be representative of the entire ecoregion. The second plausible reason for the discrepancy between malnutrition and production iu the existence of illegal rice trade on the mountain borders between Viet Nam and China. It is likely that actual food production is higher than that "registered" with the Government. Underregistration enables the mountainous provinces to obtain greater rice allocation from Government security stocks. In addition, some of the unregistered produce is smuggled across the borders to China where it fetches a higher cash value. This may increase cash income among the mountain populace, which may be yet another explanation for a relatively better diet and lower rates of malnutrition in the mountain areas. A third hypothesis is the existence of lower prices for food and other consumables in these areas than prevail elsewhere in rural Viet Nam, thus permitting poor familes to obtain more nutrition with their incomes. 3.9 Socioeconomic Differences. Data on income are not available. The following analyses use food expenditure as a proxy for socioeconomic status. An attempt was also made to construct a wealth/asset index as an alternate proxy for socioeconomic status. Information on the price of rice in each of the areas surveyed was also available, but the reliability of these data is suspect. Table 3.4 presents mean values of key indicators for children by food expenditure quartiles. The percentage of children above minus two standard deviations of weight-for-age (NCHS standards) (i.e., the proportion of well-nourished children) increases gradually as we move from the lowest to the highest food expenditure quartile. Differences between the third and the fourth quartile are much l-es than those between the second and the third quartile. Differences in height-for-age follow a similar pattern. If food expenditure is a reliable proxy for income, these data suggest that income growth has a greater impact on reducing malnutrition at lower income levels than at higher income levels. 3.10 Determinant. Analysis of individual-level data from the General Nutrition Survey indicates that the existence of a bathroom in the house -- reflecting the level of environmental hygiene and income in the household -- is positively associated with weight (Annex B). After controlling for age and sex, children residing in houses with bathrooms are 157 gms heavier than those without bathrooms. Another indicator of household socioeconomic status is a household's participation in the 5% land scheme -- a program aimed at improving household food security by making available 5% of the commune land to individual families for cultivation. The empirical results show an inverse association between a household's participation in the 5% land scheme and child weights. Although prima facie this evidence seems counter-intuitive, in fact it indicates a positive association between child weights and household socioeconomic status, since the 5% land scheme is targeted to the needy. But, more importantly, this evidence indicates that the gap between the nutritional need of children and the supply of nutrition is not being adequately bridged by the supplementation (of food security/cash income) from the 5 land. 3.11 Another important finding is that, even after controlling for household socioeconomic status and child age and sex, there are significant differences in child weights across ethnic groups and ecoregions. Children - 42 - belonging to the Viet ethnic group have lower weights than other children. Children in the Central Highlands, the Central coast of Southland, and those in the Red River Delta have lower weights than children in Hanoi and Ho Chi Minh cities. ChLidren in the Mekong River Delta are observed to have the highest weights, followed by chlldren in the North East of Southland, the Central Coast of Southland, and the North Mountaln Midlands, in that order. It is umeful to note here that this ordering of scoregions Ln terms of weight- for-age is somewhat different from the unadjusted order presented in Table 3 3. 1/ 3.12 Analysis of data on child heights shows broadly *imilar results (Annex B). The presence of a bathroom, ownership of a pig, and availability of running water in the house are all posLtively associated wlth heights .of young children. Participation in the 5% land schme is again negatively associated wlth child heights. Energy intakes and food expenditure do not show an association with height. The latter result is not surprising, and merely reflects the fact that height is influenced cumulatively by past food intakes. Another explanation is that energy intakes (and quite probably food expenditure as well) in a predominantly rural country like Viet Nam vary considerably during different times of the year, and that measurement of energy intakes at a single point in time is unlikely to be related to measures of long-term nutritional status. 3.13 As in the case of child weights, children belonging to the Viet ethnic group are likely to have lower heights than those belonging to non-Viet ethnic groups. Children in the North-east of Northlands tend to be the tallest, followed by those in the Central Highlands, North Mountain and Mid- lands, and the Mekong River Delta, respectively. Children in the Red River Delta have the shortest stature, followed by those residing in the Central Coast of Northland and the Central Coast of Southland. 3J1/ In the data presented in Table 4.3, there are no controls for age, gender, participation in the 5% land scheme, etc. Table 3.3: Key Indicators for Children 0-59 months old by Eco-Region ECO-REGIOUS INDICATORS 1 Z 3 4 5 6 7 8 1. Calorie Intake 727 806 792 874 ... 817 818 944 0-59 Months 0-11 Months 585 687 707 756 ... 786 743 791 12-24 Months 693 781 750 842 ... 783 755 888 25-59 Months 767 840 828 907 ... 833 849 985 2. Mean Food Expenses 292 68 398 278 ... 648 430 2117 3. Mean Price of Rice 618 206 611 403 ... 499 574 638 4. Mean weight (Kg) 10.54 10.24 10.20 10.69 10.17 11.06 11.33 11.24 5. Mean height (Cm) 81.9 80.8 80.4 80.9 80.96 84.6 82.6 85.8 6. Mean age (Months) 30.7 31.4 30.2 31.9 29.3 32.9 30.7 33.8 41 7. X children above -2SD weight-for-age 59.8 48.6 48.7 53.0 54.1 65.0 70.9 70.9 8. X children above -2SD height-for-age 49.0 35.4 38.6 46.8 45.9 56.4 52.1 71.6 9. X children above -2SD vt-for-height 91.1 90.2 86.6 88.2 91.8 93.6 94.5 94.5 10.X children non-Viet ethnic group 42.14 0.00 0.00 0.26 90.67 7.67 19.48 0.00 * Eco-Region Code: 1 = North Mountain & mid-lands 2 = Red river delta 3 = Central coast of northlands 4 = Central coast of southlands 5 = Central Highlanrds 6 = NorthEast of southlands 7 = Mekong river delta 8 = Hanoi & Ho Chi Ninh cities Source: NIN, General Nutrition Survey, 1987-89 Table 3.4: Key Indicators for Children (0-59 months) by Food-Expwnditure Crtftile FOCO-EXPENDITUtE SMATILE I-t flRST TILE SECOI tURTILE THIRD D IURTILE FOURTH I TI&LE 1. Nean fight (KO) 10.38 10.34 1O.73 10.6? 2. Nsmn height (CM) 81.3 80.9 82.4 82.7 3. X children above - 250 weight-for-age 52.07 54.85 58.87 59.97 4. X children above - 2SD height-for-age 37.86 42.17 47.03 50.72 5. S children above - 2SD ueight/height 91.05 90.57 90.32 91.88 6. Nea Calorie Intake 805 7m3 828 837 7. Ronm Food Expenses 13.29 56.76 75.39 93.33 8. Kean vatue of "asset index" 82.24 124.87 145.30 141.20 9. Nen age (Nonths) 31.65 30.76 31.20 31.66 Source: Nlb, General Nutrition Survey, 1991. - 45 - B. Nutritional InDuts Food Consumotion and Enerav Intakes 3.14 Diet Composition. The Vietnamese diet is dominated by rice, which supplies 80-85% of total intake of calories (Table 3.5). Carbohydrates contribute between 71-83% of total energy intakes of adults and between 62-83% of energy intaken of children under five in different ecoregions. Consumption of milk and milk products is minimal even among young children. Fish and aquatic foods substantially add to dietary quality, except in the north- mountain midlands and the central highland regions. Mean per capita consump- tion of meats in also high, though standard deviations around the mean are large, implying large differences in individual intakes. Fat consumption is low, averaging less than half of the 18% of total energy intake recommended by the National Institute of Nutrition (NIN). 3.15 The predominance of rice in the Vietnamese diet is of particular concern, especially in the case of young children and pregnant and breastfeeding women, since the caloric density of rice is very low and these individuals may thus be unable to consume enough calories to meet their special energy needs. Additionally, since rice alone is deficient in several important nutrients, such as iron, thiamine, Vitamin A and fats (lipids), women and young children are especially susceptible to diseases associated with thees deficiencies, such as anemia, beriberi, and xerophthalmia (which can ultimately lead to blindness). 3.16 Adult Enerav Intakes. Mean daily rice consumption per capita is 452 grams, with the mean value for each region being above 400 grams. Average daily energy intakes for adults are 1,928 calories. However, once again, these mean values mask substantial interhousehold as well as intrahousehold variations in food availability. According to one estimate, between 6-41% of households surveyed in different ecoregions had daily per capita energy intakes of less than 1,800 calories.32/ More recent data from the NIN General Survey Report indicate that in 1987-90, adults in nearly 9% of the sample households were consuming less than .1,500 calories, while adults in over 25% of households had energy intakes below 1,800 calories per day. Interregional comparisons suggest that the North Mountains and Midlands as well as the Central Highlands have the most equitable distribution of calorie. with about 12-15% of the adults consuming 1,800 Calories or less in these regions. On the other hand, the proportion of adults consuming fewer than 1,800 calories was about 24% in the Mekong River Delta and 35% in Hanoi and Ho Chi Minh cities. Figures for other regions ranged between 25 and 30 percent. MI1 Tu Giay, "General Survey on Food Consumption and Nutritional Status in Viet Nam,* National Institute of Nutrition, 1989 (mimeo). - 46 - 3.17 Recional Differences. There are substantial differences in mean adult energy intakes across ecoregions. While the Central Coast of Southlands has a mean adult energy intake of 1,861 calories per day, adults in the North- Mountain Midlands and the Central Highland regions consume more than 2,000 calories per day (Table 3.6). Interestingly, the mountain areas also show lower rates of child malnutrition than most other ecoregions in Viet Nam. The high average energy intakes are inconsistent with the generally low agricultural productivity and low incomes in the mountain areas. There are several possible explanations of this inconsistency in the data. As noted earlier, this discrepancy between production and consumption might arise because the NIN sample may be non-representative of the "true" mountain population, since the NIN sampled the more accessible and better-off house- holds. A second hypothesis is the existence of lower prices for food and other consumption goods in these areas than prevail in other rural areas, thereby enabling poor families to obtain more nutrition with their incomes. Alternatively, the inconsistency between food intake and food production may be explained by the existence of illicit trade on the mountain borders between Viet Nam and China and the additional unreported incomes from it. The latter explanation would imply that households in the mountain areas of the North produce more rice and earn higher incomes than generally believed. While this hypothesis does not seem plausible in the face of all the evidence that shows the mountainous regions to be poorer, conclusive evidence to resolve this issue can only be obtained through surveys covering the remote mountain areas.33/ 3.18 Child Energy Intakes. Energy intakes of young children under six (Table 3.3) are estimated at 827 calories per day (excluding breast-milk intakes). Figures for the North-Mountain Midlands are the lowest (727 calories), while those for the cities are the highest (944 calories).34/ These figures, if representative, suggest that energy intakes of young children do not parallel those of adults in the eight ecoregions. In fact, there seems to be an almost inverse relationship between adult and young child energy intakes. Awareness about the nutritional needs of young children may be one of the factors that may explain this differential. The inverse relationship may imply re-distribution in intra-family food distribution wherein children's food allocation is increased at the cost of adults. 33/ The National Institute of Nutrition is currently investigating this issue by undertaking more representative surveys in the mountainous regions. 34/ Figures for the Central highlands are not available. - 47 - Tbl, 3.S3 Avmag Deily Food CmAption per capita wd per chi td wdr 5 wrs, 1967w-9 DalLy per capita cansu ption Daily per capita consumption among children under 5 (grams) Food (gr) Nean Std. Dev. Mean Std. Dev. Ric* 451.6 4.0 188.3 34.94 Other coroels 6.21 10.79 4.25 10.70 Tuber 37.6 37.3 6.56 13.34 gems& pe 2.79 5.77 1.06 2.98 Tofu 6.80 7.44 2.45 6.27 Nutsesam 3.79 5.22 0.85 2.75 V". leaves 124.8 35.5 16.48 15.76 V". tubers 46.53 27.53 4.87 9.40 Fruit 4.09 7.45 9.49 13.32 Suger 0.76 1.82 3.09 6.22 sauces 24.73 10.73 4.66 6.09 Fats & oal 3.01 2.68 0.61 1.52 Nests 24.41 14.39 13.01 10.42 Egg & mi lk 2.93 3.65 13.30 13.96 Fish 42.1 16.6 35.32 22.33 Other efood 7.85 8.03 3.45 6.48 Surce: NIN, aeneral Nutrition Survey, 1991. 3.19 DVaaJn,jnts of Child Enerav Intakes. Analysis of data on child (0-5 years of age) energy intakes from the General Nutrition Survey indicate that the availability of running water in the child's house is the most important factor associated with energy intake (Annex B). As noted earlier, this variable-may be a proxy for income or for general environmental hygiene in the household. Interestingly, neither food expenditure nor affiliation to the Viet ethnic group is significantly associated with energy intake. - 48 - 3.20 Iicronutrient Intakee. Iron and Vitamin A intakes for adults and young children are listed in Table 3.7 by ecoregion. Intakes of iron, iodine and vitamin A are generally low throughout Viet Nam. The daily per capita intake of iron is only about 10 mg, much of which is in the unabsorbable form (non-homs iron). Thli is particularly problematic for pregnant women who need to absorb at least 6.3 mg/day in the last two trimeaters.35/ This equates to 120 mg/day for women eating foods containing only non-heme iron or 60 mg/day for women eating foods containing some hems iron.36/ In hospitals in Hanoi, 40% of children under the age of 15 years of age, and 64-70% of children 1-18 months of age, are observed to be anemic. Typically, children under 2 years of age are anemic from iron deficiency alone, while anemia in those 3-7 years of age may also be due to parasitic and other infec- tions .3/ 3.21 Iodine deficiency is a problem, especially at high elevations where iodine-depleted soils prevail. Up to 2.5% of the population in some mountainous areas have cretinism (the most severe form of deficiency) (Due, et al, 1989) with primarily young children being affected.j3/ Vitamin A deficiency is also a problem. According to the WHO, there is a public health problem when more than 0.5% of children exhibit clinical signs of vitamin A deficiency.j3/ All geographical areas in Viet Nam, with the exception of the Southeast and Hanoi City, report clinical signs above .5 percent. The North Mountain Midlands, Red River Delta, Mekong River Delta, and Ho Chi Minh City all report the highest rates of vitamin A deficiency. These high rates of vitamin A deficiency do not correspond with per capita retinol consumption for the general population. In the North Mountain Midlands, where vitamin A deficiency is highest, per capita consumption of retinol is also the highest. This probably means that children are not being fed foods containing vitamin A (or vitamin A activity). 35/ . M. DeMaeyer, "Preventing and Controlling Iron Deficiency Anaemia Through Primary Health Care," W.H.O., Geneva, 1989. 36/ H. Levin, B. Pollitt, R. Galloway, and J. McGuire, "Micronutrient Deficiency Disorders," in D. Jamison and H. Mosley, eds., Health Sector Priorities Review, The World Bank, 1991 (draft). 37/ Brun, op. cit., p. 11. 38/ D. T. Due, L. Mv, and J. Dricot, *Viet Nam," IDD Newsletter 5(2): 23-24. 39/ Levin, et al., op. cit. Table 3.6: Key Nutrition Indicators by Eco-Region, 1987-89 ECO-REGIONS' INDICATORS 1 2 3 4 5 6 7 1. Nean Adult Calorie 2107 1878 1880 1861 2059 1924 1891 1886 Intake/capita/day 2. X energy from 80.1 82.6 79.9 76.2 81.3 77.8 80.9 71.0 carbohydrates 3. X of fanilies with 11.8 27.2 30.4 30.1 15.0 27.1 25.3 35.1 adults consuhing < 1800 catories/day * Eco-Region Code: 1 * North Mountain & mid-lands p. 2 a Red river delta D 3 a Central coast of northlands 4 u Central coast of southlands 5 a Central Highlands 6 a NorthEast of southlands 7 a Mekong river delta 8 = Hanoi & Ho Chi Minh cities Source: WIN, General Nutrition Survey, 1987-89 Table 3.7: Nicro-Nutrfent Intakes by Eco-rgloan, 1967-69 l"QICATBtS 1 2 __: 4 5 3 6 7 a 1. Newt Iran Intakes 10.31 9.60 9.81 6.79 9.01 9.64 6.91 10.06 (adults) 2. Nu n Iron intakes 3.10 3.67 3.35 3.96 ... 4.10 4.17 4.76 (0-6 yar) 3. Keen Vit A Intakes 0.01 0.01 0.01 0.04 0.01 0.03 0.02 0.13 (amults) 4. NKen Carotene 2.78 3.32 2.0? 1.44 3.53 2.26 1.61 2.41 fntake (adults) 5. Wean Vit A Intakes 0.05 0.07 0.02 0.03 ... 0.04 0.02 0.20 (0-6 years) 6. Wean Carotene 0.25 0.61 0.17 0.16 ... 0.60 0.12 0.93 O Intak (0-6 years) * Eco-Region Code: 1 a North Nountain £ nid-lands 2 a Red river delta 3 u CentraL coast of northlands 4 a Central coast of sauthlnds 5 * Central Highlands 6 * NorthEast of southtands 7 - Nekons river delta 8 = Hanoi & Ho Chi Minh cities Source: MIH 6eneral Nutrition Survey, 1987-89 - 51 - BreastfeedinQ Patterns 3.22 Breast-feeding is almout universal in Viet Nam, with between 97-99% mothers breast-feeding. Mothers over thirty years of age, those living in rural areas, those working in the agriculture sector, and those from the northern regions are slightly more likely to breast-feed than others (Table 3.8). However, differences between these groups are small. The KIN reports a small (but steady) decline between 1987 and 1989 in the percentage of mothers breast-feeding at 9 and 12 months of age. The mean duration of breast-feeding is 14.5 months, with rural mothers and those working in the agricultural sector breast-feeding for a little over one month longer than urban mothers (Table 3.8). Basic literacy is associated with a longer period of breast- feeding, an advantage that seems to be lost with higher education (due to a concurrent shift in work patterns away from agriculture). The Vietnamese government allows maternity leave for 180 days, a factor which promotes breast-feeding among urban mothers. The most common reasons for discontinua- tion of breast-feeding among mothers (NIN data) are maternal perception of lack of milk and the mother's return to work. 3.23 A study on 2,579 infants aged 0-23 months from lowland and mountainous provinces of the North and the South reported 100% breast-feeding between 1984-1989.40/ Duration of breast-feeding ranged from 18-21 months. Major reasons for discontinuing breastfeeding in urban areas were maternal employment, and, in rural areas, maternal perception that the "child had grown up". Rooming-in after birth was more common in rural areas, but not in urban areas. Over 81% of urban-born infants were separated from their mothers at birth, and prelacteal feeds were common among 49-78% of newborns. 3.24 According to one NIN study, in the third month of lactation, only 21% mothers produce 600 ml or more of breast-milk, with 56% mothers producing between 240-500 ml. However, another NIN study on 35 mothers (origin not known) concludes that "the average milk secretion during the first three months varies at 500-600 ml per day ...".41/ There has been extensive re- search on the sufficiency of breastmilk with conclusions that some women have insufficient milk because of poor nutrition or because they work.42/ It is more likely that these women are not breastfeeding their children exclusively, so that milk production is low and intake inadequate. In addition, premature introduction of supplemental foods greatly increases the risk of infection in the small infant. 40/ Nhan Nguyen Thu, Hofvander, et al. "Breastfeeding and Weaning Practic- es," KIN, Hanoi, 1990, mimeo. ill Kim Nguyen Thi, "Nutritive Composition of Breastmilk of Vietnamese Mothers," NIN, undated (mimeo). 2j/ Brun, op. cit. - 52 - Tdble 3.6: Mam Duratlon of Breratftsdug by Selected backgrosmd Characterlsttea 190 Duration of Breastfb.dfng (n Background Characteristfc months) Weighted Nuiber of Births Under 30 years 14.1 672 30 years or older 14.8 916 Urban/Rural Residence Urban 13.4 252 Rural 14.8 1333 Realon North 14.8 824 South 14.2 767 Educatfon lltiterate 14.6 92 Can Read or Write 15.7 288 Primary 14.5 960 Secondary 13.4 248 Professi Agriculture 14.9 1138 Other productlon Nonproduction sector 13.7 276 TOTAL 14.5 1588 Source: Viet Nan Demographic and Health Survey (VNDHS). 1988 3.25 Since over 90% of births in Viet Nam take place at health facili- ties, hospital and institutional practices play an important role in the promotion of breast-feeding. According to a breastfooding and fertility survey in the district of Uong Bi in Quang Ninh Province, breast-feeding was initiated after 72 hours for 35% of the infants.43/ Among others, it was delayed for 1-3 days after birth. Almost 99% of the mothers waited until 36 hours after birth before they initiated breastfeeding, presumably because of an incorrect perception that colostrum is an inferior food. Direct information on exclusive breastfeeding was not given, but it was reported that 90% of infants were receiving supplementary food by 4 months of age. (Mean duration of breastfeeding was observed to 13.3 months.) Bottle-feeding was not generally practiced because formula was not available; instead mothers supplemented with solid foods. 43/ N. T. Nga and P. Weissner, "Breastfeeding and Young Child Nutrition in Uong bi, Quang Ninh Province, Viet Nam," Journal of Tronical Pediatrics 32 (3)s 137-139, 1986. - 53 - supolementarv Feedina of Infants 3.26 Information on duration of exclusive breast-feeding is limited. The VNDHS estimated the mean duration of breastfeeding nationally to be 14.5 months. According to the breastfeeding and fertility survey in Uong Bi, complementary feeding of infants is started at 3 months. Age at starting supplementary feeding in four provinces studied by the NIN is listed below: 3.27 Another study in the Central province shows Hanoi (North) 3.4 - 4.2 months that weaning foods are, Tien Giang (South) 4.9 - 5.4 months introduced early at 2.8 Ha Na Minh (North) 3.0 - 4.2 months months, while breast-feeding Lang San (Mountain) 0.7 - 2.3 months is prolonged until 16.3 months.4j/ (Sources NIN) 3.28 Traditional beliefs proscribe the feeding of vegetables and meats to infants before one to two years of age. Use of fats and oils is believed to lead to diarrhea. The most common supplementary food for infants is rice gruel (prepared with rice flour and salt, with small amounts of meat, fish, mung beans or meat added at times), commonly given after 4 months of age. Cow milk is usually not given to infants. Use of commercial infant formula is limited to urban areas where up to 48% of the infants of factory workers are bottle-fed. Viet Nam has yet to implement the International Code for Marketing of Infant Foods. One small NIN study found a significant difference in diarrhea incidence between supplement- ed and non-supplemented infants aged 0-3 months and 4-24 months. 3.29 A weaning food consisting of rice flour, germinated "mung bean" and salt/sugar is manufactured by a government-owned, UNICEF-assisted factory close to Hanoi. The low-cost weaning food with a shelf-life of four months is packaged and sold in 200 gm plastic bags. Plans are under way to substitute some of the World Food Programme rations (given to lactating mothers) with this weaning food. Feedina in Child-Care Facilities 3.30 Since a large proportion of urban and rural mothers work, child- care facilities are crucial in determining breast-feeding patterns. The creche system and the formal nursery schools were integrated in 1987 and are now administered by the Department for Children's Protection and Education. The system now covers all pre-school chidren. These creches are financed either by the state-run cooperatives, or by communes/parents themselves, with marginal support from government budgets. In 1988-89, 28,122 creches were functional, providing service to just about 15% of the total 0-3 age popula- tion, this figure being 5% less than that for 1987-88. A majority of these (80 percent) are located in urban areas and in the northern part of Viet Nam. Much of the need for child-care facilities is met by "unregistered" or Al/ Khan Nguyen Cong, Thin Duc Hoang, and An Quoc Tran, "Recent Observations on Breastfe.ding Practices at Three Communes of the Central Provinces," NIN, Hanoi, 1990 (mimeo). - 54 - informal family-run creches housed in cramped private homes. The exact numbers of such creches are not known, but they are fairly common, especially in urban areas. In rural areas, many of these creches are set up to meet seasonal demands such as during harvest time. 3.31 In one typical creche visited in Quat Dong commune, mothers leave their infants (0-36 months) at the creche early morning before they leave for work, return once during the day to breast-feed, and pick up their children at the end of the day when returning from agricultural work. Mothers leave food for the infants, and contribute towards the cost of the creche services. Infants are fed rice-soup during the day on demand. Opportunities for breast- feeding during the day are obviously severely curtailed. Interestingly, surveys indicate that creche facilities are availed as a second resort when family support is not available, and girls are more likely to be left in a creche than boys.4S/ 3.32 In conclusion, breast-feeding in Viet Nam is initiated late, supplementary foods are low in energy and inappropriate, and a majority of the mothers (especially in rural areas) return to work soon after child-birth, leaving infants in the care of older siblings, grand-mothers, or in some cases, creches. Opportunities for breast-feeding infants in creches are limited, so that many infants survive on rice-soup as a substitute for breast- milk through the day. The reduced opportunities for suckling suppresses breast-milk production, thereby promoting the cycle of inadequate infant- feeding. The Government is still considering implementation of the Interna- tional Code for Marketing of Breast-milk substitutes. C. Food Production and Security 3.33 Food Production and Exports: Per capita food production in Viet Nam has been increasing almost continuously since the mid-1970s (Figure 3.1). Since 1987, rice production -- indeed, total cereal production -- has been increasing even more rapidly (although there was a small dip in 1990), in part because of far-reaching policy reforms in the agricultural sector (Figure 3.2). The increase in rice production is attributable primarily to an increase in the total area under cultivation rather than to an increase in rice yields.46/ Rice exports have increased dramatically between 1985 and 1990 (though recent reports indicate a drop in exports for 1991 as a conse- quence of lower production). 3.34 Food balance sheets prepared by the FAO show that average daily availability of calories and protein per capita, which fell dramatically during the 1970s, have been increasing secularly during the 1980s. At pres- ent, energy availability is estimated to be more than 2,250 calories per day (Figure 3.3). This figure is higher than the recommended level of energy intake for a moderate activity adult in Viet Nam. Therefore, undernutrition 45/ Stewart H. Fraser, "Viet Nam: Maternal Child Health and Education -- Notes, Aspects and Prospects," Population Education Studies, La Trobe University, Victoria, 1990 (mimeo). 3j/ Tu Giay, "A decade of Food and Nutrition Development in Viet Nam: 1980- 89," NIN, Hanoi, 1991 (mimeo). - 55 - in Viet Nam is not a problem of inadequate food production, but instead one of distribution and demand. A poor procurement, storage and transportation system continues to hinder the efficient dLitribution of food grains from the food-surplus to the food-deficit areas withln the country. Indeed, government encouragement of regional self-suffLel-ncy in rlce in the past has prevented development of storage, transportation and dlitribution systms. In addition, in recent years, the removal of price controls and abolltion of food subsidies and rice allocation to public employees may have adversely affected the demand for food, and consequently nutritional outcomes, among the poor. 3.35 Without detailed information on issues such a woP'stI as regional rice production, ss5 food transport systems, the regions from where rice ex- I sos ports originate, and season- al variations in nutritional 9S - status, it is difficult to la -55 say whether the dramatic 85 - /\ la increase in rice exports in recent years -- a policy . A actively encouraged by the 2\ Government -- has had any | repercussions on the nutrl- tional status of the popula- tion. For example, if rice L d in95 ,9h .eb s.. *ub exports are being channelled vor from rice-surplus areas, as appears to be the came, the Figure 3.1 adverse effects on nutrition are likely to be minlmal, since a poor transportatLon system in any case did not allow for re-distribu- tion of this surplus rice to rice-deficlent areas (though it is not clear how rice stores are being mobilized for export despite inadequate transportation facilities). More information is required on where the exportod rice origi- nates, how it is transported, and changes in food availabLilty (both at macro and micro-levels) and consumption ln these areas between 1989 and 1990.51/ Reports for 1991 indicate that rice exports from Vlot Nam may be lower than those in 1990 as a conseuence of lower productlon in this year. It li hI/ A food and nutrition surveillance system (NFSS) assisted by UNICSF and managed by an inter-agency commlttee composed of the Stat- Plannlng Committee, General Statlitics Offlce and the National Institute of Nutrition has been set up by the Government. Tho system is intended to monitor the food and nutrition sltuatlon ln 15 food-deficit provinces. Reports on the food and nutritLon sltuation are done seml-annually and reported to FNSS commLttees at the natlonal and provLncial levels. In 1991, the report of the natLonal FNSS committee to the Council of Ministers called attention to the impendLng daterloration of nutrltlon indLcators ln the food-deficit provinces. Ac a result, a pollcy decLsion was taken to drastically reduce rice exports, Lncrease the volume of rice distributed from the southern provinces to the food-deficit provLaces, and impos price controls on selected food commodltLes. - 56 - unclear whether this is a reflection of governmental sensitivity to the food- security situation, or whether the decreased exports simply reflect the Government's inability to procure adequate quantities of rice for export. 3.36 Exports of frozen sea products and coffee have more than quadrupled since 1980, while others like eggs and tea have registered more modest improvements. Exports of fruit have declined dramatically, though it is not evident whether this is related to decreased productlon or to lncreased local consumption. 3.37 Data show that productlon and consumptLon of foods other than rice have remained almost static over the decade. Dlverslflcation ln food production and consumptlon may be one of the keys to enhanced food-securlty in Vlet Nam. 3.38 Reaional Dlfferences in Food Production and Potential. In the past, most reglons wlthln Vlet Nam, with the exceptlon of the Mekong Rlver Delta, ran large food deficits (Table 3.9). Because transportation systems wlthln the country are poor, the Government developed a policy of regional self-sufficlency in rice. This discouraged the transfer ofric mcapita cereal ad rice pro .'an, 1980-90 transfer of rlce fromX .' f surplus areas to deficit 1 30- areas. Since 1989, rice productLon is sufficlent to support the populatlon ln all regLons; but lt is not clear how these new surplus- 2 - /290 es have affected malnutrl- tlon. (The latest data on 2 - -27 nutrLtLon from the NIN cover 5 260- 20 the period 1987-89). 3.39 The North's Red -230 Rlver Delta, known as the 215 "heartland of Viet Nam," has 1 i 9 a large populatlon relative Yer to the amount of arable land available and the food pro- Figure 3.2 duced. Historically, there has always been an "open border" policy wlth the South to encourage redLstribution of the populatlon. The Red River Delta houses 26% of the total populatlon of the country, but has only 13% of the country's cultivated land and produces only 19.9% of lts food (in rice -equLvalents). Because of overpopulation, this area has suffered from soil eroslon and general soil deterioration which in turn exacerbates the low productive capacity of the land. The potential for yield improvements does exist, primarily through greater input (especially fertilizer) use and, in the longer term, through rehabilitation of existing irrigation facllities and new small-scale irrigatLon development in minor watershed areas. Nonetheless, because of the small average soze of farms (0.3 hectares on average) and projected growth in population, the Northern region wlll continue to be food deficit for the foreseeable future. - 57 - 3.40 The South's Mekong River Delta, where food surpluses abound, provides Viet Nam with 38% of its cultivated land and 42% of its food crops (in rica equivalents), but houses only 22% of its total population. Possibilities for increasing production in the South are good. Limiting factors are the acid sulphate soils, defoliation due to years of war causing drainage problems, and water shortages during certain times of the year.A4/ 3 * 41 The mountains are A'9g doily nutrient avalleflity, per capita. 1965-56, considered unproductive 2300- areas, and many parts have 2250- been heavily deforested. - 22W0-0 However, the sparsely popu- lated highlands and the more 2150- densely populated coastal 2100- areas both have high devel- J 2050- opment potential. Together, 2000- A-2 these two areas contain 55% /2 of the country's total popu- - 59 - -.0 lation, 51% of its cultivat- f ed land, but only provide -4B 38% of the country's food crops. Both these areas 5 1951t918911 9741971 198 193 suffer from lack of agricul- I Yer tural services, since the benefits of agricultural Figure 3.3 extension and research ac- crue disproportionately to the Red River and Mekong River Deltas. 3.42 The major potential for reclamation is in the Central Highlands (upland food and cash crops), where both food and labor shortages are real constraints to future development. This area has also not received the research and extension attention it requires to increase productivity. In addition, the country's past policies of emphasizing regional self-sufficiency in rice forced these areas to neglect alternative food crops that were easier to grow and could be important sources of food for the people there. 3.43 Another particular feature of several regions in Viet Nam, particularly in the North and the Center, that has a bearing on food security is their susceptibility to natural calamities, especially typhoons. The typhoons typically result in large crop losses and disruptions in transport and supplies. There is, therefore, a large variance in agricultural production in these regions from year to year, which, in the absence of smooth interregional flows of food grains, results in a large variance in nutrient intakes and incidence of malnutrition. 3.44 Aaricultural Policy. Agriculture is the most important sector in Viet Nam. It currently accounts for 38% of the country's total output value, 49% of national income, and employs about 62% of the national labor force. At AA/ M. Beresford, "Viet Nam: Socialist Agriculture in Transition," Journal of ContemDorarv Asia, 20 (4): 466-486. - 58 - 19 percent, agriculture (including irrigation) takes up a significant share of total government expenditures. 3.45 Although agricultural production has been increasing secularly during the 19800, it is not impressive when compared to other countries in the region. Growth rates in agricultural production are below those gained two decades earlier by most other South and Southeast Asian countries. The average paddy rice yield in Viet Nam is 2.8 tons per hectare, while the yields in the neighboring Chinese province of Guangxi are 3.9 tons per hectare.49/ 3.46 Agricultural production suffered from government policy immediately after the war. In an effort to gain more control over agriculture production and ensure food security, the Government created farming collectives. This extended a policy developed in the mid-1960s in North Viet Nam when, after bombing by the United States government, per capita food supplies fell to near subsistence levels requiring the Government to become dependent on food imports 50/ The Government forced procurement of rice at low prices to ensure food security for the urban industrial sector but this resulted in profound disincentives to farmers. One of the policy reforms Viet Nam has instituted in the last couple of years has been to reduce its control over agricultural production and input prices so that market pricing now predomi- nates. 3.47 After 1981, when it was clear collectivism was not working, the Government broadened its policy so that land could be distributed to individuals under a contract system. After 1989 this policy was further broadened to give 15 years of tenure security to farming families. Decen- tralization of production and marketing decisions to households is virtually complete (except for certain industrial crops). In addition, collectives, which were the primary distributors of inputs and services in the past, have been disbanded in the South and converted to cooperatives in the North. There is recognition within the Government that the farming family unit will be developed further in the future. The VAC51/ system, which integrates agriculture, livestock and fish production, is currently being promoted to further strengthen family-sized plots, particularly in the North. 3.48 The effect of these reforms has paid off. Agriculture yields and production increased slowly but steadily from 1981 to 1985 (approximately 5% per year). After 1987, there has been a dramatic increase. In 1989, Viet Nam became the world's third largest rice exporter. 3.49 Despite productivity level below potential under the collective system, increased production has resulted from land reclamation that increased W2/ World Bank, "Viet Nam: Stabilization and Structural Reforms: An Economic Report," Country Operations Division, Country Department II, 1990. 0/ A. Pforde, "Specific Aspects of Collectivisation of Wet Rice Cultiva- tion: Reflections of Vietnamese Experience," Discussion Paper No. 159, Department of Economics, Birkbeck College, University of London, 1984. l/ VAC stands for yaon (garden), Ao (pond), and Chan nuoi (animal husbandry). - 59 - the amount of cultivated land by 19% (to 5.6 million hectares). Most of this land was planted with industrial or tree crops. Future increases in production will probably come from further land reclamation and increased use of fertilizers and high-yielding varieties. 3.50 Until recently, the objective of regional self-sufficiency in rice took precedence over that of improvements in per capita food consumption or nutrition.ja/ The policy of promoting self-sufficiency in rice within each region was meant to mitigate the need for transporting rice between food surplus and deficit areas. Even then, the magnitude of government-managed rice flows was not inconsequential. Prior to 1989, when national rice production averaged about 16.0 million tons in paddy equLvalent (nearly 10 million tons of rice), the Government imported about 300,000 tons of rice and shipped about the same amount from the South to the North. During this period (1984-88), with an annual national foodgrain production of 18.3 million tons, the Government procured an average of 3.8 million tons and distributed 2.7 million tons of foodgrains per year. Detailed information on non-governmental flows is not available. In early 1991, amidst forecasts of a poor winter- spring, the Government tripled the volume of rice it shipped from the South to the North. 3.51 In late 1990, overcommitted rice exports accompanied by speculative rice stocking and criminal disappearance of government security stocks drove up prices to such an extent that prlces in Ho Chi Minh City were even higher than those in Hanoi. This unprecedented situation even saw some private- sector movement of rice from the North to the South for a very brief period. During this same period, "over-exportation" of maize at the expense of domestic consumption also took place, largely because of the extreme pressure to earn foreign exchange. 3.52 The FAO reports that the Government's 1995 projections for agricultural production would improve per capita food consumption by 20 percent.AJ/ In turn, energy intake would increase to a level of more than 2,400 kcal. per day -- a gain sufficient to reduce malnutrition in a signifi- cant proportion of the population (assuming a population growth rate of under 2% and attenuated post-harvest losses). The projections for agriculture production imply a 1985-95 trend growth rate of per capita farm incomes as high as 7.7% per annum.54/ China achieved similar growth during 1978-87. However, as personal incomes increase and the demand for higher quality food- stuffs increases concomitantly (as it usually does with increasing incomes), it is likely that the country's existing crop mix, which is dominated by rice, will not adequately serve increasing consumer preference for food variety. While the increasing demand for foods other than rice could be met through imports, it may also make sense for the rice-deficit regions in the country to exploit their comparative advantage in crops other than rice. la/ Regional food self-sufficiency is no longer pursued as a priority policy (and has not been vigorously enforced for some years) by the Government. iJU Food and Agriculture Organization, "Viet Nam: Agricultural and Food Production Sector Review," UNDP, FAO and The World Bank, 1989. 9A/ FAO, op. cit. - 60 - 3.53 Food Subsidies. An of 1989, food subsidies in urban areas have been discontinued to accommodate unified pricing. Increases in food prices have been compensated for by increasing the wages of most urban dwellers, although there is little information on the effect of the subsidy removal on the ultra poor who may not have benefitted from increased wages. Speculation ie that the urban unemployed and upland farmers have benefitted less than lowland farmers from the liberalization of prices, especially for rice. Information on the nutritional status of landless wage laborers and individu- als working in the informal sector is not available. D. Nutritional Policy and Strateay 3.54 Vietnam does not have a clearly-articulated national nutrition policy or strategy. Perhaps because regional food self-sufficiency was a paramount objective of government policy in the past, improvements in per capita food consumption or nutritional status were secondary goals. However, a significant effort to fill the strategy gap took place through a recent UNIC$F/NIN national seminar which drafted recommendations towards a plan of action to achieve improvements in nutritional standards by the year 2000. Participants in the seminar included representatives of the State Committees of Science and Planning, the Ministries of Health, Agriculture and Food, Industry, and Education; the General Statistical Office, and organizations representing farmers, VACVINA, women's groups and academic institutions. The ten objectives of the plan of action were: (a) Improvement of weight gain of mothers during pregnancy from 8 to 10 kilograms; (b) Reduction in the prevalence of low birth weights from existing 15- 20% to 10 percent; (c) Reduction in the incidence of malnutrition (weight-for-age below two standard deviations of the NCHS standard) from more than 50% to 25-30 percent; (d) Reduction in the incidence of nutritional anemia among women (15-49 yre) from 50-60% to 35-40 percent; (e) Eradication of Vitamin A deficiency and Xerophthalmia in areas where the program is implemented; (f) Promotion of breastfeeding and adequate child feeding practices; (g) Reduction in the prevalence of goiter in the mountain areas by 50 percent; (h) Promotion of growth monitoring; (i) Extension of the family VAC program to cover 50% of households; and (j) An increase in average daily energy intakes per capita to more than 2,100 Calories in all regions, with fewer than 70% of the calories derived from cereals. - 61 - 3.55 The seminar also proposed that the plan of action be developed through an intersectoral nutrition focal point to be set up in the Government. The plan of action would be included in the State plan for general economic and social development, and reflected in grass-roots sectoral plans for agriculture, health, child care, and family planning. To that end, a national "Programme of Action for the Survival, Protection and Development of Children" was adopted by the Government in late 1991. The Council of Ministers is responsible for general oversight of the program and will carry out formal progress reviews in 1993 and 1995. A national Committee for the Protection and Care of Children (CPCC), whose chairman iL the vice-chairman of the Council of Ministers, will monitor overall progress and promote implementation through a network of CPCC committees down to grassroots levels. The program document implies that financial rather than operational limitations pose the greatest constraint to successful execution. However, implementation capacity and coordination among concerned agencies clearly are key factors in determining the degree to which those ambitious nutrition goals will be met. Whether strategic issues of planning, design, coordination, program efficiency, management capacity and resource allocation for nutrition should be considered by the Council of Ministers, the CPCC or the State Planning Committee (SPC) remains to be determined. Nevertheless, it is clear that some national focal point for nutrition policy and strategy formulation and promotion remains to be established. E. Current Nutrition Interventions 3.56 Viet Nam's interventions generally have tended to center on three sets of activities: promoting the VAC ecosystem, correcting micronutrient deficiencies, and supplementary feeding. Some nutrition education efforts also have taken place through the Ministries of Education and Health. However, few of the interventions have been national in scope. 3.57 The VAC Ecosystem. The VAC ecosystem is probably the single most important nutritional intervention promoted by the Government. The concept of VAC, which was promoted by Ho Chi Minh and is rooted deeply in national traditions, is based on the strategy of renewing solar energy and reusing all kinds of waste. Under the program, rural households, communities and cooperatives are encouraged to set up an interlocking and interdependent system of gardening (vuon), aqua culture (ao), and animal husbandry (chan nuol). The end-products derived from the three activities, such as vegetables, pulses, tubers, fruit, poultry, eggs, pigs, fish, etc.) are used for human consumption, while the by-products are recirculated among the activities. For example, fodder from gardening is fed to household animals and fish. Animal waste is used as organic fertilizer for the garden and feed for the fish. The pond is used for watering plants, and so on. - 62 - Table 3.9: Regional Difference in Population, Cultivated Land, and Food Production, 1991 Noun- Central Central Central North- tan & Red River Coast Coast Hightand east Mekong Mid- Delta North- South- South- River lands tend land land Delta 1986 Population (000) 10,068 13,576 8,573 6,655 2,490 7,797 14,172 X)_ 15.9X 21.4X 13.5X 10.5X 3.9X 12.3X 22.4X Annual Paddy Production (000 tons) 1,969 3.743 1.563 1,606 396 836 8,883 Annual per capita Rlce 195 276 182 241 159 107 627 Production (kog.) K of Total Food Crop Production (rice aquivalent) 12.0X 19.9X 9.0X 8.8X 2.8X 4.92 41.92 K of total pigs production 24.5X 23.0X 16.5X 11.4K 4.7X 4.8X 15.0X Source: GSO (1991) 3.58 Thanks to the efforts of VACVINA, a semi-private organization of retired horticulturists, the VAC ecosystem is widespread in the country. In certain rural communities, it is estimated that as much au one-half to three- quarters of total household income is generated from VAC activities (Tu Giay and Duong Hong Dat 1988). One of the reasons for the VAC system's popularity is that it builds upon an age-old Vietnamese tradition. In addition to its income supplementation objective, the VAC system also seeks to increase the availability of food at the household level and enrich the diet of rural households. In particular, by encouraging the consumption of animal proteins and vegetables, it seeks to introduce greater variety in the otherwise monotonous and rice-dominated Vietnamese diet. Since 1988, UNICEF has provided support for the promotion of the VAC system among poor families, especially those with malnourished children and pregnant women. As of 1992, UNICEF has provided VAC-related assistance to about 500 communes in the country. Other NGOs are developing pilot projects in health and sanitation with the VACVINA as executing agency. - 63 - 3.59 Micronutrient Sugiplemontation. The Government also has had a number of micronutrient supplementation programs. Goiter surveys and planning for control programs began in 1971. In 1976 a salt iodination program was introduced in 11 provinces, covering 3 million people.-5/ The prevalence of goiter in Hasn Binh Province was reduced from 63% to 25% after iodized salt, which was continuously supplied, was introducedj56/ A similar reduction was experienced in Ha Tuyen Province where golter prevalence dropped from 80% to 40 percent. In 1986 a renewed strategy of IDD control was lnitiated by the Government with the help of CEMUBAC, a Belgium NGO, and UNICEF. The main objective is to prevent new cases of cretinism and to reduce golter prevalence to below 10% in 7-15 years old. Plans were to iodize 45,000 tons of iodized salt for 8 million people and give 1 million people injections of iodized oil. While some salt is iodized at the provincial lovel,57/ much of the salt is iodized at the village level using simple equipment. With the help of a parallel educational program, the campaign has been well accepted by the population and goiter prevalence has been reduced consider- ably. Some communes have been selected to monitor the progress of the program. In 1989 the State Committee for Science and Technology has investi- gated designing iodination equipment to fit local needs more closely. 3.60 Control of vitamin A deficiency has also been given attention, although programmatic activities have been sporadic. UNICEF has proposed a program to help the Government with prevalence surveys, seminars, training, nutrition education, and universal distribution of vitamin A capsules to children under 3 years of age and lactating women.L5/ Also planned is a household food production project to increase the production and use of foods high in vitamin A activity. FAO is considering a similar food production program to increase the availability of foods high in vitamin A. 3.61 Supolementarv Feeding. The World Food Programme (WFP) project 2651 was primarily a feeding project initiated in 1984 which has since been substituted by a follow-up project (WFP 3844). The original WFP project was largely a nutrition supplementation project, providing supplemental foods (viz., flour, sugar, dried skimmed milk, fish powder, and oil) to women in their last trimester of pregnancy, women in their first three months of lactation, children under 3 years of age at creches, and hospitalized children under 15 years of age who need rehabilitation. The project covered only 11 of the 44 provinces in the country, and only a limited number of communes in each province. As such, the WFP intervention, although useful, has been basically on a pilot basis. The new project (3844) attempts to link nutrition services to primary health care at the commune level by utilizing food commodities as jj/ T. Ma and T. Z. Lu, "Iodine Deficiency Disorders in the Western Pacific Region, in B. S. Hetzel, J. Dunn and J. Stanbury, eds., The Prevention and Control of IDD, Elsevier Press, North Holland, 1987. 25/ Due, et al., op. cit. 57/ UNICEF has assisted six provincial governments in setting up salt iodization plants. IfV/ In 1991, for instance, UNICEF assistance provided for the distribution of some 3.6 million capsules in 36 provinces. - 64 - Lncentives for pregnant women and familles with malnourlshed chlldren under throe to regularly vLsit commune health centers. 3.62 In addltion, the FAO ls supportlng a Nutrltion Improvement/Vitamln A control project ln selected areas. UNICEF is playlng an actlve role in health-oriented programs implemented through the exiutlng primary health care Lnfrastructure. other intervention efforts are limlted to small-scale experiments by non-governmental organizations, such as Save the Chlldren Fund, the Mennonltes (MCC), and CIDSE (Cooperation Internatlonale pour le Develop- pement et la 80idairit6). There is a need to revlew the collectlve experi- ence of all national, international and bilateral agencLes as well as the NGOs ln the field of nutrltion so as to draw lessons for future plans. Such a revliw may also facLILtate a convergence of development activities initiated by different agencies so as to promote a coordinated approach towards health and nutritlon impiovements in Vlet Nam. F. Institutional Arrangements 3.63 No strong focal point for nutrition advocacy or nutrition planning, strategy formulation or personnel development exlsts at present in Viet Nam. Programs of Ministries like Health, Education and Food touch, but do not focus, on nutrition. The national planning process has yet to recognize nutritional improvement as an explicit human resource development objective. The Natlonal Institute of Nutrition (NIN), which was set up in 1980 as a central specialty instltute under the Ministry of Health, has been the key institution for implementing nutrition-relevant actions in the country. The NIN played a key role in the implementation of the World Food Programme supplementary feedlng project. However, the NIN's role seems to be leos signlficant in the follow-up project. In addition, the NIN has been developlng educational materials to help mothers improve the nutritional status and health of thelr children. The NIN sees itself principally as a technical resource focusing on nutrition and food research, food quality and hygLene, and training cadres for community food and nutrition responsiblilties. Training programs in community nutrition are extremely limlted. Efforts are being made to include nutrition training in the medical currlculum by setting up a Department of Nutrition in the Hanoi Medical College. No trainlng or research programs in applied or community nutrition are yet in place in Vlet Nam. 3.64 The General Statistical Office (GSO) is another lnstitutional resource for nutrition-related activities. The GSO currently has primary responsibliLty for the large UNICEF-assisted nutrition-surveillance system get up in 1990. The surveillance system ls guided by the State Plannlng Committee and the MinLstry of Health. G. Issues and Recommendations 3.65 There has been an impressive increase in agricultural and food productlon ln Vlet Nam during the last decade. Consequently, Viet Nam is now self-sufficient in food; in other words, domestic food production is adequate to meot the energy needs of the population. However, food self-sufficlency does not mean imply an adequate level of food intake for everybody in the - 65 - country. Indeed, there is compelling evidence that a significant proportion of the adult population -- anywhere from 11 to 35%, depending upon the region -- may be consuming fewer than 1,800 calories per day. The magnitude of child and maternal malnutrition is also surprisingly high in comparison with the infant mortality rate. Most recent estimates put the proportion of children malnourished at 45% by weight-for-age and 56.5% by height-for-age. These high ratas are comparable to those observed in Bangladesh. Further, although the evidence is not conclusive, it appears that malnutrition, especially among infants under one year of age, may have increased in the late 1980s. 3.66 The high rates of malnutrition, especially among children, arise for a number of reasonss a poor food transportation, storage and distribution system that cannot redress the regional imbalances in food supply &nd demand; the susceptability of many regions in the country, especially in the Center, to natural calamities (viz., typhoons) that cause crop loss and food shortages; a worsening distribution of income that results in lack of food purchasing power amongst the poor; relatively high rates of fertility; boor maternal health and nutrition that result in low birth weight babies; suboptimal breastfeeding and weaning behavior; and a high incidence of diarrneal and other infections among children, caused in part by poor sanitation and lack of health care. 3.67 The fact that child malnutrition rates in Viet Nam are so high, especially in relation to the infant mortality rate, suggests that the country has concentrated more on qyantlty rather than the quality of life in its health achievements. The large numbers of children who survive infanthood face a bleak future of malnutrition and doprivation. It goes without saying that this imbalance between quantity and quality needs to be redressed urgently. 3.68 Agricultural Policies. The cause for high rates of malnutrition cannot be found in inadequate food production in the country. Rather, the problem lies in the poor food procurement, transportation, storage and distribution system in the country. Indeed, the Government's encouragement of regional food self-sufficiency in the past retarded development of the food distribution system. A poor distribution system combined with the existence of major agricultural productivity differences across ecoregions has resulted in the population in some parts of the country being prone to chronic undernutrition. In addition, the preoccupation with rice self-sufficieincy resulted in the neglect of other subsidiary food crops, which are important in increasing dietary balance and variety. As a result, micronutrient deficiencies are pervasive in the entire country.59/ 3.69 Major improvements in the food procurement, storage, transportation and distribution system will need to be undertaken, so that food can move from surplus to deficit areas efficiently and quickly. These improvements may 59/ Another important efficiency implication of the past emphasis on local self-sufficiency was the country's inability to exploit regional comparative advantage. Hence, although agricultural production has increased impressively, it remains well below the country's full potential. - 66 - require large initial investments in transportation and storage infrastruc- ture. After having made these infrastructural investments, however, it may be best to rely on private-sector foodgrain flows to reduce interregional variations in food availability. Of course, since private-sector flows typically respond to price incentives, the problem of food availability in the food-deficit areas will not disappear overnight, since consumers in these areas do not have the purchasing power to bid up the price paid for foodgrains from the surplus regions. In fact, at present, it is financially more rewarding to export rice outside Viet Nam than to transfer it to the deficit regions within the country. Indeed, as private-sector grain trade expands, the availability of food in the deficit regions may initially decline before it improves. 3.70 In the long run, however, the only sustainable means of improving food security in the deficit regions will be by letting them exploit their comparative advantage in crops other than rice. There is substantial scope for increasing agricultural production in the deficit areas. Greater geographical equity of supply of fertilizer and other inputs, combined with official financing of research and extension work targeted on the crops and farming systems of the poor areas (which currently cannot afford to "buy" the attention of the research system), will allow the food-deficit areas to increase production of crops in which they specialize and thereby exploit their comparative advantage. As personal incomes increase in Viet Nam, the demand for crops other than rice is likely to increase. The rice-deficit regions in the country would be well positioned to serve this demand. As noted earlier, diet diversification is likely to have important nutritional benefits as well. 3.71 Role of Economic Growth in Reducinc Malnutrition. Food insecurity is often more closely linked to personal incomes and consumer purchasing power than to food availability in a monetized economy. The 1987-89 General Survey of Nutrition shows positive associations between indicators of socioeconomic status (such as presence of a bathroom or availability of running water in a house) and child nutritional status. Energy intake also appears to be responsive to household socioeconomic status. If these results are accurate, the economic growth that Viet Nam has experienced since 1989 (and will most likely experience in the future) may improve average nutrient intakes. However, it is not clear this will necessarily translate into lower rates of malnutrition for three reasons. First, the evidence from other countries indicates that income growth alone can take an inordinately long time to eradicate undernutrition.A6/ Second, since food subsidies were also abolished after 1989, the purchasing power of households (especially, employee households that were the major beneficiaries of food subsidies) may not have improved -- and, in fact, may have worsened -- despite rising money incomes. Third, there is some indication that economic liberalization and the resulting emrgence of the private sector in Viet Nam have widened income disparities substantially. As a result, the growth in average per capita income that Viet Nam has been experiencing in the last few years may be masking a decline in absolute incomes and an increase in poverty in a section of the population. la/ See Jere R. Behrman, Anil B. Deolalikar, and Barbara L. Wolfe, 'Nutrients: Impacts and Determinants," The World Bank Economic Review 1(7), September 1988. - 67 - If this is indeed the case, nutrient intakes may be falling, and the incidence of malnutrition worsening, among this group of households. 3.72 Consequently, general economic growth alone cannot be relied on to alleviate the problem of malnutrition. Shorter-term measures are needed to avert or reverse growth faltering in children below 3 years of age through a combination of behavioral and health interventions, as well as targeted food supplementation designed to optimize the use of available food supplies. 3.73 Nutrition Education. Overall household food insecurity is a proximate cause of malnutrition, but its effects can be compounded by suboptimal breastfeeding and weaning behavior and the impact of infection, diarrheal and other diseases. The journey into child stunting can begin with low birth weight which in turn may originate partly from the mother's malnutrition as an adolescent. There are two cultural practices in Viet Nam that may exacerbate child malnutrition. First, breastfeeding is generally initiated 3-4 days after birth, largely because of an incorrect perception that cologtrum is an inferior food. Second, although the mean duration of breactfeeding is 12-14 months, exclusive breastfeeding is not common, and infanto are introduced to supplementary foods as early as 2-3 months of age. Since infant formula is not available widely, breastfeeding is not supplemented with bottle-feeding but instead with solid foods. The premature introduction of supplemental foods greatly increases the risk of infection in small infants. 3.74 Another social-behavioral problem is that the typical Vietnamese diet -- dominated by rice -- is inadequate for older children and pregnant and lactating women. The caloric density of rice is very low, with the result that young children and pregnant/lactating women are often unable to obtain enough calories from rice to meet their special energy needs. In addition, rice does not contain several important nutrients, such as iron, thiamine, Vitamin A and fats (lipids). As a result, women and young children on an exclusive rice diet are vulnerable to dioeases associated with these deficiencies, such as anemia, beriberi and xerophthalmia (which can lead to blindness). The combination of low overall calorie intake and little variety in the diet create a fragile nutritional balance for the most vulnerable groups. 3.75 There thus appears to be an important role for providing nutrition education, especially on infant nutrition, breastfeeding and dietary variety, to pregnant women and mothers of young infants. Whie the National Institute of Nutrition has been developing educational materials to help mothers improve the nutritional status and health of their chidren, there is no institutional mechanism or program for widely disseminating such materials. The Government urgently needs to designate an agency to take the national lead in improving family nutrition behavior and to promote development of a strategy and program to do so. 3.76 A National Nutrition Prooram Indeed, given the high rates of child malnutrition, a strong case might be made for a national child nutrition program that would include growth monitoring, selective short-term supplementation, and nutrition education. Such a program could be implemented by the primary health care system. To some extent, the World Food Programme Project 3844 attempts to do, although on a piot basis. At present, few of - 68 - the commune health centers in the country perform child growth monitoring on a regular basis. 3.77 In order to contain costs, the nutrition intervention program could be area-targeted (to regions having the highest child malnutrition rates), age-targeted (concentrating exclusively on children 6-36 months of age and pregnant and lactating women), and need-targeted. Targeting by need could be achieved by monitoring the weights of all children 6-36 months old in the project communes, and enrolling only those children whose weight gain over a certain period falls below standard. Those children would be singled out for special health monitoring, food supplementation and intensive nutrition education for their families. It would be important to enroll children in the nutrition supplementation program only for the duration of time that there is a lapse in their weight gain, so that long-term dependence of the beneficia- ries on nutrition supplementation is discouraged. The linking of health and nutrition services would be a critical element of the program. Children who do not respond to nutrition supplementation could be provided health services, including check-ups and referrals, treatment of diarrhea, deworming, and immunization. These services chould also be available to pregnant and lactating women. In addition, the program could include intensive counseling of mothers in nutrition and hygiene education. In designing such a program, Viet Nam might be able to derive lessons from the experience of other developing countries that have experimented with such integrated nutrition programs, often with the assistance of the World Bank.ilI 3.78 Institutional Arranaements. Adoption of the Programme of Action for the Survival, Protection and Development of Children" (see para 3.55) underscores the need to establish a focal point within the Government for formulation, coordination and oversight of a national nutrition strategy and program. Institutional arrangements for nutrition advocacy, to promote more efficient and effective program implementation and for training particularly in community nutrition also need to be strengthened from national to local levels. Careful consideration should be given to possibly incorporating these functions into the role of the State Planning Committee. ]Jj See Alan Berg, Malnutrition: What Can Be Done? Lessons from the World Bank Experience, Baltimore: Johns Hopkins, 1987. - 69 - IV. HEALTH A. Health Status: Achievements and Current Status Infant and Child Mortality 4.1 National Estimates. Although there is no unanimity on estimates of crude death rates, infant mortality, and average life expectancy in Viet Nam,62/ there is little doubt that Viet Nam has experienced considerable improvements in all of these indicators during the last three decades. Figure 4.1 shows that by 1989 the crude death rate had fallen to less than one-third of its level in 1950. The infant mortality rate was roughly halved from about 156 per 1,000 live births in 1960 to around 83 in 1979, and was then approximately Crk Rate, Vietnun, 1950-89 halved again during the next X ten years (to a level of 45 in 1989). The 1990 Census est mat- ed average life expectancy at birth to be 65 years -- 63 years for males and 67% i_ years for females. This was d up from a rate of about 34 in 1936 (for North Viet Nam), 50 in 1970 (for North Viet Nam), and 60 in X __ 1978 .fi/ 4.2 Little is known about the under-five mortality rate. The Year Institute for the Protection ____ of Child Health has Figure 4.1 estimated this rate to be Al/ The two competing estimates of infant mortality are from the 1989 Population Census and the Viet Nam Demographic Health Survey (VNDHS). The two estimates differ, since the former uses information on birth history of women and the Brass method to indirectly estimate infant mortality, while the latter utilizes direct survey data on mortality of children. The VNDHS comes up wlth an infant mortality rate of 33.5 for the period 1983-88, which is significantly lower than the estimate of 45 provided by the Census. Since the VNDHS sample size is very small in comparison with the *Lsi of the 1989 Census Sample, the sampling error of the VNDHS estimate is llkely to be greater than that of the Census estimate. Unless otherwise noted, the Population Census estimates of infant mortality will be used in the remainder of this report. fa/ Ministry of Health of the Socialist Republic of Viet Nam, "Health Service in the Socialist Republic of Viet Nam," Hanoit Ministry of Health, 1981, m$meo. - 70 - 108 in 1987 and 98 in 1988. An under-five mortality rate of around 100 is high in relation to an infant mortality of 45. If correct, it would corroborate the evidence on high levels of malnutrition among children aged 2- S discussed in Chapter III. 4.3 Causes of Infant PDeaths: The Institute for ouap in infant wwtaity Rate et~wbn ig7g an ige. the Protection of Child JPovine Health reports that, of all - the perinatal deaths during i the first month of life, 25% a - occur due to prematurity, 18% due to respiratory infections, 8t due to -35 tetanus, and 6% due to X 0 congenital anomalies . However, since these data -m are compiled on the basis of referral cases seen at the Institute (and which -ml typically are complicated 5n 7t d - Tl5 rnfwit HctaUty RateIn17 cases), they may not be indicative of the situation Figure 4.2 in the country. Nevertheless, these data suggest that premature births and neonatal tetanus, Infat Mortality and Per Capita GNP in 14 Asian Countries. 1989 both of which are generally : preventable by good antenatal care, are important, if not major, causes of infant deaths in Viet Nam. 4.4 Regional Differencess An average infant mortality rate of 45 v ; \ for Viet Nam is misleading in view of the wide mortality variations acrosstsss regions and between rural so _ o __bo 3000 500'0 and urban areas The Per Capita Go (USO disparity in infant mortality rates between Figure 4.3 rural and urban areas is of the order of 38%. The variation across regions is also considerable; for example, while the average infant mortality rate for the Central Highland region is 56.4, that for the Northeast Southland (which includes TP Ho Chi Minh) is merely 33.9 (Statisti- cal Annex Table 4.1). There are provinces in the Mountain and Midland Region of the North and in the Central Highland Region that have infant mortality rates exceeding 60, while tho provinces of Hai Phong, Tien Giang and TP Ho Chi Minh have infant mortality rates of 30 and below. - 71 - 4.5 Further, the data indicate that the infant mortality rate fell unevenly across regions and provinces during the 19805 (Statistical Annex Table 4.1). The relative decline in the North-East Southland Region was the greatest (53.5%), while that in the Mountain and Midland Region was smallest (27.5%). The province of Lai Chau in the Mountain and Midland Region, which had a high infant mortality rate of 72 in 1979, experienced a decline in infant mortality rate of merely 8.2% between 1979 and 1989. On the other hand, TP Ho Chi Minh, which had a relatively low infant mortality rate to begin with (viz., 60), experienced a steep decline of 54.4% over the same period. However, Figure 4.2 shows that provinces that had high infant mortality rate in 1979 did have larger absolute declines in infant mortality rate over the next ten years as compared with provinces having low infant mortality rates in 1979. These results imply a narrowing of interprovincial disparities in the infant mortality rate during the 1980s. 4.6 Comparison with Other Countries: Viet Nam's LU1f Expbctanc and Per Capita GNP in 14 Asian Countriaes lg infant mortality rate and life expectancy are impres- 75 sive even when compared with the health indicators for countries having considerably higher per 65- capita income. Figure 4.3, which plots the infant ii mortality rates of 14 Asian countries against their La respective per capita GNPs 2 in 1989, shows Viet Nam to be a significant (negative) outlier. Viet Nam has an 45 infant mortality rate that 4 7wdo sbo obo 5odo is one-half of the infant PMr Capita GNP USS mortality rate predicted for its income level, given the estimated relationship be- Figure 4.4 tween infant mortality rate and per capita income across Asian countries. The same finding is observed in Figure 4.4, which presents a scatter plot of average life expectancy against per capita income. Viet Nam is observed to have an average expectancy of life that is significantly greater than the average life expectancy predicted for its income level. Adult and Maternal Mortality 4.7 National Estimates. Unfortunately, there is no authoritative source on estimates of the maternal mortality rate. The official MOH statis- tics, given in Table 4.1 below, suggest a sharp increase in maternal mortality between 1975 and 1980 but a steady decline since 1980. The maternal mortality rate is estimated by the MOH to be 1.1 deaths per 1,000 deliveries. However, these estimates are contested by the MOH's own Institute for Protection of the - 72 - Mother and Newborn, which reported an average rate of 5.76 per 1,000 deliv- eries from a 1985 hospital survey of seven provinces.§4/ Table 4.1t Maternal Mortality, 1975-90 Maternal Mortality Rate Year (Per 1,000 deliveries) 1975 0.9 1980 2.0 1985 1.4 1989 1.2 1990 1.07 Source: Ministry of Health, 1990 Table 4.2: Maternal Mortality by Region, 1990 Maternal Mortality Rate Region (Per 1,000 deliveries) Mountain and Midland 1.5 Red River Delta 0.5 Central Coast Northland 1.2 Central Coast Southland 1.2 Central Highland 1.8 North East Southland 0.9 Mekong River Delta 1.0 Average for all of Viet 1.07 Nam Source:' nf.stry of Health, 1990 4.8 Causes of Maternal Deaths. The study undertaken by the Institute for Protection of the Mother and Newborn in seven provinces and Hanoi in 1985 cited earlier found that 32% ofithe mothers had died because of infection, 21% because of severe anemia, 11% owing to post-partum or post-abortion hemor- rhage, 7% on account of kidney failure as a result of toxemia, and the remaining 9% because of other complications. Many of these conditions are likely to be caused or exacerbated by inadequate or incompetent antenatal care M) UNICEF, "Situation Analysis, SocLalist Republic of Viet Nam," October 1987 (aniio). - 73 - and improper use of contraceptives. It is estimated by UNICEF that only 7% of pregnant women in 1988 had received complete anti-tetanus vaccinations. 4.9 Reaional Differences, The figures for the national average again mask substantial regional variation in maternal mortality. As in the case of infant mortality, official Ministry of Health (MOH) statistics show maternal mortality in 1990 to be highest in the Central Highland (1.8) and Mountain and Midland regions (1.5), and lowest in the North East Southland (0.9) and the Red River Delta regions (0.5) (Table 4.2). The hospital survey of seven provinces undertaken by the Institute for Protection of the Mother and Newborn in 1985 found maternal mortality rates as high as 9 deaths per 1,000 deliveries in some of the provinces.65/ Of course, these unusually high rates reflect the fact that a large number of women with high risk pregnancies are referred to hospitals for having their deliveries. Morbidity 4.10 Mortality, being an extreme manifestation of poor health, occurs relatively rarely in a population. General morbidity -- viz., illness episodes, infections and outbreaks of preventible diseases -- is a much more common consequence of poor health, and accounts for a great deal of absentee- ism and productivity loss in the labor market. Unfortunately, not even any rough estimates are available of how many annual illness episodes occur on average in individuals belonging to different age groups. 4.11 Patterns of Diseases. The data on leading causes of mortality and morbidity indicate that vector-borne diseases -- in particular, malaria, followed by diarrhea and respiratory diseases -- account for the majority of reported illnesses in 1988 (Table 4.3). A WHO report suggests an even steeper increase in the number of malaria deaths than is indicated in the above table -- from a rate of 1.67 per 100,000 persons in 1985 to 5.37 in 1989, an increase of over 200% in five years (Statistical Annex Table 4.2)66/ The same report estimates that the actual number of deaths due to malaria are 10 to 20 times greater than the number reported. Within malaria, the chloro- quine-resistant P. tslclparum strain has become widespread, especially in the South. The resurgence of malaria has occurred due to several reasons: resistance of the parasite to chloroquine, reduction in insecticide spraying programs, and the return of infected troops from Cambodia in recent years. The revival of malaria has coincided with an increase in dengue fever. Indeed, a serious outbreak of hemorrhagic dengue fever, which has a four-year seasonal recurrence, can be expected in 1992 if appropriate mosquito control measures and community mobilization are not promoted soon. 4.12 In a 1991 MON survey of public and private health providers in three provinces, acute respiratory infections and diarrhea were found to be the leading causes of patient visits. Almost 27% and 10.4% of the patients attended by the health providers during the two weeks preceding the survey had been treated for acute respiratory infections and diarrheal infections, respectively. Other important diseases for which patients had sought treat- ill UNICEF, 1987, ibid., p. 99. 5il Nolineaux, "Report of Duty Travel to Viet Nam," November 1990 (WHO report). - 74 - ment were skin infections (4%), accident-related injuries and burns (3.6%), malaria (2.6%), and tuberculosis (2.5%). Since the survey had been conducted during March and April -- months in which malaria is not a major problem -- the number of malaria cases are most likely understated. 4.13 Viet Nam has higher morbidity rates associated with vaccine preventable childhood diseases, such as measles, diphtheria, and pertusais, than other large countries in the region, including India, Thailand, Myanmar, Indonesia and Bangladesh. The morbidity rate associated with tetanus has actually been increased between 1984 and 1988.6Z/ Tablo 4.3: Causes of Mortality and Morbidity (per 100,000 population) Mortality Morbidity Ailment 1978 1988 1978 1988 Tuberculosis 5.0 3.7 138 106 Malaria 7.0 2.7 1,241 2,166 Watery diarrhea 1.9 835 1,132 Bloody diarrhea 410 337 Heart diseases 1.4 Dengue fever 1.0 1.2 218 196 Bloody dystentry 2.0 1.2 Tetanus 0.8 1.0 Pneumonia 0.9 77 88 Rabies 0.8 Nutritional disorders 0.8 44 Eye infections 140 Peptic ulcers 0.2 73 Respiratory infections 119 257 UNrCEF, Viet Nam: The Situation of Children and Women, UNICEF, Hanoi, 1990, p. 73. B. Government Health Policy 4.14 Because of its international isolation and severe resource limita- tions, Viet Nam adopted a health strategy that emphasized self-reliance and austerity from early on. The basic tenets of this strategy included preven- tive services, use of traditional drugs and medicine in conjunction with 67/ UNICEF, Situation of Children and Women, Hanoi, 1990, p. 74. - 75 - modern methods, community-based health delivery systems, and self-reliance in the production of pharmaceuticals. Recent Measures 4.15 In March and April 1989, the Council of Ministers promulgated a series of bold measures designed to liberalize the health sector and mobilize new resources for the sector. These new measures included the following: (a) Introduction of user fees for health care: Basic health care was formerly free for all, regardless of a patient's ability to pay some or all of the cost of care. on May 1, 1989, a fee system was introduced in the three higher (district, provincial and national) levels of the health-care delivery system,68/ requiring those patients who are able to do so to pay at least a minimal part of their health care. The handicapped, families of health personnel, and individuals able to produce certification of indigency from their neighborhood or village People's Committee are still treated free of charge. (b) Leaalization of Drivate Dractice: When health care was free for all, health providers were forbidden to practice privately, and everyone was expected to use the public health system. Since 1989, doctors are allowed to practice privately (i.e., charge fees) after office hours in the public health facilities. There are also a growing number of physicians and traditional healers who engage in full-time private practice. (c) Sale of druas and medicines on the open market: Drugs and medi- cines were formerly dispensed free, though the public health network, to all patients. In July 1989, the Government began to allow pharmaceutical factories to open retail pharmacies and sell drugs and medicines to individuals, hospitals and health centers directly. Patients able to produce a certificate of indigency can still receive medicines free. (d) Liberalization of the oharmaceutical industry: Until April 1989, pharmaceutical factories, like all other industrial units in Viet Nam, had little control over decisions involving production and financial matters; production, input procurement and pricing (including employee wage levels), sale and pricing of outputs were all determined by state planners. In practice, pharmaceutical factories sold drugs and medicines to public hospitals and health centers at subsidized prices fixed by the state. In April 1989, pharmaceutical factories (and factories producing or repairing medical equipment) were given the freedom to make their own decisions concerning: (a) use of inputs, including wage levels; (b) production levels; and (c) selling prices of drugs and medicines. Pharmaceutical factories were also informed that they could no longer rely on government subsidies. ill Primary health care at commune health centers continues to be free (again in theory), but fees are paid at intercommunal polyclinics. - 76 - 4.16 These decisiona place Viet Nam in the forefront of socialist econo- mies attempting to restructure their health care systems. Current Policy Priorities 4.17 Current government health policy, as outlined in the document, "Strategy for Socioeconomic Stabilization and Development up to the year 2000," identifies the following six prLorities for the health sectort - strengthening prevention and control of infectious diseases - improvement of the basic health services network at the grassroots level - reduction of the population growth rate to 1.8% by the year 2000 - development of self-sufficiency in essential drugs - promotion of traditional medicine within the framework of primary health care, and - increased access to safe water and environmental sanitation. 4.18 These policy statements establish a clear framework for goal setting and action programming. The Government has also identified six national health programs, described below, as having the highest priorities and has included them in the National Plan of the Council of Ministers.69/ Two of these health programs are directed at improving the health services network, and the remaining four are personal health interventions directed to target populations, namely, mothers and children and people living in malaria- affected areas. The six programs are:70/ - Consolidation of health services at the village and district levels to implement the ten elements of Primary Health Care.7/; - Provision of maternal and child health (MCH) care, includings i) family planning; ii) control of diarrheal diseases (CDD); iii) malnutrition control; iv) acute respiratory infections control (ARI); - Strengthening of basic health services, including: i) expansion of diagnostic and treatment services; ii) raising the quality of first aid; iii) combining modern with traditional medicine; and iv) promoting non-drug therapy; - Malaria control; j2/ Ministry of Health, "Program of Action of the Health Services for the 1989- 1990 period," February 1989. 2Q/ See Annex C for a complete description of these programs. 1X1 Viet Nam has added the following two elements to the eight ones promoted by the Alma Ata Declaration on Primary Health Care: i) integration of modern and traditional medicine; and ii) sustaining the efficiency of the public health network. The other eight elements are: 1) promotion of health education; 2) food supply and proper nutrition; 3) adequate water supply and sanitation; 4) maternal and child health (MCH) and family planning; 5) immunizations; 6) prevention and control of endemic diseasea; 7) appropriate care of common diseases and injuries; and 8) provisior. of essential drugs. - 77 - - The Expanded Program of Immunization (EPI)l and - Essential drugs and materials. 4.19 Although the need for improving human resources has not been separately listed in this policy statement, it has been considered in a separate proposed Health Manpower Development Plan for the period 1990-2005 (see Annex D). Similarly, while financing policies are not highlighted in the policy statement, the Government has included the following measures as part of health servies¢ strengthening: improved collection of reasonable medical and hospital user fees for establishing village and district health and welfare funds; the opening of private maternity homes; the establishment of private clinics; the opening of private drug stores under speclfic regula- tions; and the promotion of health and accident insurance schemes. Also, the MOH has requested tax exemptions on imported drugs, medical equipment and other medical supplies from the Government, and has promoted a cost-accounting system for the local production of drugs by autonomous government factories. C. Structure of the Health Care System 4.20 Health services in Viet Nam are organized along a four-tiered pyramid. At the top of the pyramid is the Ministry of Health, which is the main national authority in the health sector and, together with the Provincial and District Health Bureaus and the Commune People's Committees, formulates and executes the health policy and programs for the country. The Ministry manufactures and distributes pharmaceutical, trains doctors, coordinates medical research, and is ultimately responsible for the provision of all curative and preventive health services in the country. The Minister of Health, a member of the Council of Ministers, is assisted by two vice minis- ters, one located in Hanoi and the other in Ho Chi Minh City. 4.21 The Ministry is assisted in its activities by a number of central specialty institutes, which function as tertiary care referral centers and professional training and medical research hubs. Among these are the Insti- tute for Protection of Mothers and Newborn, the Institute for Hygiene and Epidemiology, the National Institute of Nutrition, and a number of other institutes responsible for research, training and patient care in the areas of cancer, pediatrics, tuberculosis, otorhinolaryngology, traditional medicine, pharmacy, surgery and ophthalmology. 4.22 The Ministry of Health operates at the provincial level (and in Hanoi, Haiphong and Hoc Chi Minh City) through a Provincial Health Services Department, funded by the central government, which coordinates treatment and prevention activities, trains assistant doctors, manufactures pharmaceutical, and provides referral facilities (including laboratories for hygiene and epidemiology) to the districts. In each province, there is also at least one general hospital with 500-1,000 beds that typically has all seven departments: internal medicine, obstetrics and gynecology, surgery, pediatrics, infectious diseases, traditional medicine, and an emergency ward. The provincial hospitals are intended to be referral centers only, but the fact is that the referral system does not work in practice. Few of the patients cared for at these hospitals are referred from outer communities; the vast majority reside in the general vicinity of the hospitals. In addition to the general hospi- tal, each province may also have one or more specialized centers or hospitals - 78 - (e.g., oncological hospitals, cardiology centers, psychiatric hospitals, or centers of traditional medicine). 4.23 In each district there is a district general hospital, including a laboratory and a post for hygiene, epidemiology and malariology. Typically, a unit for MCH care and family planning is attached to the district general hospital. District hospitals are supposed to serve as referral institutions for all intercommunal polyclinic. in the district. They also provide training facilities for health staff working in intercommunal polyclinics and commune health centers in the district. Each district also has (i) brigades of hygiene and epidemiology, commanded by the regional branches of the central (national) specialty institutes, which move around the district providing support to categorical health programs, and (li) two or more intercommunal polyclinics, which are commune health centers that have been upgraded with selected laboratory and surgical equipment and 4-5 specialist doctors (usually an ophthalmologist, otorhinolaryngologist, dentist, internist, and clinical laboratory specialist). Intercommunal polyclinics are located strategically in relation to the commune health centers they serve. The Government's goal is to supplement the network of commune health centers with one intercommunal polyclinic for every 4-5 commune health centers. To reach this goal, over 1,900 intercommunal polyclinics would be needed in addition to the 700 existing ones. Each intercommunal polyclinic currently serves an average of 92,000 people; at the desired level of coverage, each would serve about 34,000 people. 4.24 At the bottom of the pyramid are the commune health centers. Each of the approximately 9,800 commune health centers in Viet Nam is responsible for providing primary health care, including preventive, ambulatory and inpatient services, to between 2,000 and 10,000 persons, and for referring complicated cases to upper levels of care. The commune health center is supposed to be staffed by a team of one assistant doctor in charge of admlnistering the center and training the staff; another assistant doctor trained in pediatrics and obstetrics/gynecology; and one pharmacist responsi- ble for dispensing drugs. Sometimes this team is complemented with an assistant doctor in traditional medicine, a health worker responsible for immunizations and sanitation, and an auxiliary nurse. In exceptional cases, a full physician is also part of the health center staff. Commune health centers are expected to implement the national programs of MCH and family planning, ARI, EPI, and CDD.72/ They are assisted in this effort by a Hygiene and Epidemiology Brigade worker from the district health services who is supposed to visit the commune for several days each month. D. Health Care Inputs Public Facilities 4.25 Numbers of Facilities of Different Tvyes. Although there has been a rapid expansion of health facilities in Viet Nam since 1968, Table 4.4, which shows the number of health facilities in relation to the population, shows that the per capita availability of hospitals and health centers wj/ But the lack of logistical support and well-trained staff reduce all program efforts to perfunctory activities. - 79 - actually fell sharply after 1975 (the year of reunification), largely because of th. lower per capita supply of health facilities in the South. Yet the flgure of 166.7 (commune and urban) health centers per million population in 1989 for Viet Nam is considerably higher than the corresponding figures for Indonesia (32), China (63), and Thailand (141).73/ 4.26 Data on the number of beds available, indicating inpatient care capacity in the country, indicates one inpatient bed per 389 persons (representing an average of one hospital bed per 628 people and one confinement bed per 1,019 people). These hospital bed/population ratios are much better than those prevailing in other low-income Asian countries, such as China (1/465), Philippines (1/647), Thailand (1/665), India (1/1,489), and Indonesia (1/1,743).74/ In fact, Viet Nam has one of the most favorable hospital bed ratios in Asia and in the entire developing world. The trend in overall beds/population ratio has, however, remained flat over the last decade. 4.27 Thus, potential population coverage by basic health facilities (ambulatory and hospital) in Viet Nam is almost complete. The official figure of health services coverage is 100% in urban areas and 75% in rural areas. These figures do not indicate a need for additional expansion in the total number of health facilities in the country. Indeed, in view of the very low occupancy rates at district hospitals (see discussion below), there might even be scope for reducing the ratio of district hospital beds to population by as much as 50% without any deterioration in the population's access to inpatient care. 4.28 Reaional Disnarities. The national averages for Viet Nam do, however, conceal wide provincial differences in the average number and spatial diLtribution of health facilities. At first glance, the mountainous provinces in the North and the Central Highland provinces appear to be favored over provinces in the Red River and Mekong River Delta. For example, while approximately 48,000-55,000 persons share a hospital in the three Central Highland provinces, a hospital in most provinces in the Mekong River Delta serves well over 100,000 persons (Statistical Annex Table 4.3). However, tince population densities are so much lower in the highland regions than in the Red/Mekong River Delta, the average person lives at a greater distance from a health facility in the highlands despite the greater per capita provision of health facilities. 4.29 Statistical Annex Table 4.3, which also reports the average radius of the service area of hospitals and polyclinics, shows that access to health facilities is considerably worse in the central highland and northern moun- tainouu provincs (e. g., Son La and Gia Lai-Kon Tum) than in the Red River and 2W World Bank, "Indonesiat Poverty Assessment and Strategy Report, " Report No. 8034-IND, Country Department V, Asia Region, Washington, D.C., 26 December 1988; and World Bank, "Indonesia: Issues in Health Planning and Budgeting," Report of Country Department V, Asia Region, Washington, D.C. 12 June 1989. 24/ The statistics, which are for the mid-1980s, are taken from Charles C. ariffin, "Health Sector Financing in Asia," Internal Discussion Paper, Asia Regional Series, Report No. IDP 68, The World Bank, August 1990, p. 76. - 80 - Mekong River Delta. For instance, a person living in Lai Chau would have to travel an average of 25.6 kms to reach a hospital, while in Hai Phong (s)he would have to travel only 5.3 kms. Furthermore, the average diotance figures understate the problem of access to health facilities in the mountainous and highland regions, where the terrain is difficult and the means of transporta- tion limited. 4.30 Thus, with the exception of the remote mountainous and highland areas of the North and the Center, the overall availability of health facilities as well as their spatial distribution appear to be adequate in Viet Nam. In part, access to health centers is generally good because communities have a say on their location through the Commune Council. The remote areas are, however, poorly served, with average distances to health facilities being extremely large in some provinces. Table 4.4: Health Facilities in Viet Nua, 1945-1989 Year Total No. of hos- Total No. of No. of hos- no. of pitals per no. of health cen- pital beds hospi- million pop- health ters per per 10,000 tals ulation centers mill. pop. population 1945 47 5.1 -- -- 1 1955 57 4.2 200 14.6 11 1965 252 13.8 5,463 299.2 16 1975 437 20.1 5,786 266.8 28 1976 -- -- -- 20 1980 676 12.7 9,886 185.4 25 1983 689 -- -- -- 24 1985 729 12.2 9,890 165 24 1986 757 12.4 10,573 173 24 1987 765 12.3 10,732 171.2 24 1988 767 12.0 10,716 167.5 24 1989 762 11.9 10,683 166.7 24 Notes: Data for 1945-75 are for the Democratic Republic of Viet Nam (i.e., North Viet Nam). Source: Data for 1945-1975 and 1983 are taken from Ulrich Vogel, "The Whole of Viet Nam can be Considered as One Well- Designed Project: Some Reflections on Primary Health Care Experiences in Viet Nam," unpublished master's thesis at the University College of Swansea Center for Development Studies, University of Wales, 1987, p. 24. Data for 1976, 1980, and 1985-89 were obtained from the Ministry of Health. - 81 - 4.31 The 1991 MOH survey of health facility users provides an indication of the distance that health facility users had to travel in each of the three sampled provinces. This survey indicates that, on average, users had to travel just over 2 kms. to the nearest intercommunal polyclinic and under 2 kms. to the nearest commune health center. However, these figures are likely to significantly underestimate the true proximity of health facilities for the population, since they represent averages over the sample of individuals actually using health facilities. To the extent that individuals do not use health facilities that are located too far away, a sample including users and nonusers would most likely have indicated poorer proximity to health facilities. 4.32 The provincial distribution of beds in hospital and in basic health care centers, shown in Statistical Annex Table 4.4, also indicates a wide variance in the size of the population served by a hospital bed. In 1989 the hospital bed:population ratio varied from a low of 1:375 (in Hai Phong) to 1:1,100 (in Thua Thien Hue). 4.33 Ouality of Health Services. The data on the number of health facilities in Viet Nam are impressive but misleading. For instance, although there are supposed to be over 10,000 commune health centers -- one for each commune -- field visits reveal that some commune health centers are simply housed in the homes of health workers, in the quarters of the People's committee, or in very inadequate structures. Further, since allocation of funds is related to the number of beds an institution has, beds are kept on the books even when they are rarely, if ever, used. The problem of poor quality is worse in remote mountainous areas. In these areas, the health centers are often essentially nonfunctional. The problem is compounded by the fact that, since there is no comprehensive system of supervision, monitoring or evaluation of grass-roots medical facilities in the health sector, no one at the planning levels, including the MOH or the provincial health services, knows how many of the 10,000 commune health centers in the country are operative in the sense of being able to offer even the most basic of health care. 4.34 In the 1991 MOH survey of health providers in three provinces, the overwhelming majority (91.1%) of providers cited low salaries, inadequate equipment (87.6% of providers), and inadequate drugs and medical supplies (82.6%) as problems affecting the quality of the health services they dis- pensed.75/ In contrast, 33.7% of providers believed the staffing at the health facility to be deficient, while only 15.7% felt that their problem was they were seeing too many patients. These survey results are useful in directing attention to the most important causes of poor quality of health services in the country, viz., low salaries, poor or nonfunctioning equipment, and inadequate (and, in many cases, nonexistent) medical supplies and drugs. Medical Eguipment 4.35 Overall Viet Nam has a serious shortage and obsolescence of medical equipment. This problem is compounded by the lack of maintenance, which Ii, It should be emphasized that this survey is not necessarily representative of all health providers in the country. - 82 - causes premature wear-and-tear on equipment and renders much of it unusable. In the 1991 MOH survey of health providers, only 49.3 and 58.4%, respectively, of rural commune health centers reported having a usable sterilizer and a unable weighing scale for infants (Table 4.5).76/ The corresponding figures for intercommunal polyclinics and for commune health centers in urban areas were somewhat higher. With no proper equipment with which to sterilize needles, the risk of spreading hepatitis infection and AIDS via injectable vaccinations is considerable. Similarly, it is unclear what, if any, type of child growth monitoring can take place at commune health centers if nearly 42% of them in the rural areas do not even have a usable weighing scale for infants. 4.36 The same survey shows that health providers are deeply aware of this problem. Nearly 40% of the staff of rural commune health centers reported the lack of medical equipment as a very severe problem -- second only to the problem of low salaries -- in the delivery of quality health services. 4.37 The physical deterioration of equipment is in large measure a result of inadequate maintenance. There is only one institute ia the entire country that trains medical equipment maintenance technicians. The institute opened in 1973 and has graduated a total of only 1,000 trainees to date. Almost all of these graduates are employed by national and provincial hospitals, with none reserved for the district or lower levels of health services. Further, since the graduates of the maintenance training institute are not trained with sophisticated medical equipment, they are rarely capable of maintaining such equipment. The repair and upkeep of more complex equip- ment throughout the country is handled by only two repair workshops situated in Hanoi and Ho Chi Minh City. The workshop in HCH City has no vehicles, so the repair technicians have to use public transportation to reach the provinces and districts. This can take up to several days in many cases. Table 4.5: Proportion of Public Health Facilities having Usable Medical Equipment Commune Health Center Intercommunal Polyclinic Type of Medical Equipment Rural Urban Rural Urban Weighing scale for adult. 45.6 54.3 63.0 52.2 Weighing scale - infants 58.4 73.6 75.9 56.5 Blood pressure gauge 72.3 74.1 93.7 87.0 Sterilizer 49.3 60.9 56.3 71 0 Source: MOH Survey of Health Providers, 1991 21/ These numbers are not necessarily representative of the situation in the country. - 83 - 4.38 Maintenance expenditures are severely underfunded in the national and provincial health budgets. The problem is compounded by the fact that external donors often fund the purchase and import of new medical equipment but not its upkeep and maintenance. In many cases, the equipment does not even come with appropriate manuals. Health Manpower 4.39 Catecories of Health Workers. The lowest level of health manpower are the Red Cross volunteers. They have been trained by commune health station staff and provide, on a part-time basis, simple first aid treatment and family planning advice. The next level up is the elementary level. This group include nurses, midwives, pharmacists and certain types of technicians, and typically receives training for 3, 6, 9 months at the district level. Certain categories of elementary health workers, such as brigade nurses attached to cooperatives and other state work units, have virtually disappeared during the last 3-4 years, as the funds available to pay these workers have dried up. 5econdary level (technician) health workers include nurses, midwives, pharmacists, aseistant doctors and various types of technicians. They are typically trained for 2V*-3 years at provincial secondary medical schools, after which they are assigned to commune health stations or to diLtrict or provincial hospitals. Finally, professional health staff include physicians and pharmacists, who are trained at a University Faculty of Medicine or Faculty of Pharmacy. Their training typically lasts for six years. This group of health workers almost exclusively work in district or provincial hospitals. Postgraduate specialization is also available, generally requiring flve years of experience. 4.40 Suoplv and Mix of Health Manower. Table 4.6 shows that the population:physician ratio in Viet Nam has been declining secularly since 1945, with the exception of the period immediately after reunification (1975- 80) when the ratio increased slightly. However, the per capita availability of all other health personnel -- particularly, assistant doctors -- decreased since 1975, most probably owlng to the poor supply of health workers in the South. In 1989 each physiclan on average served 2,694 persons, while each nurse served 760 persons. 4.41 The trend that is of concern in Table 4.6 is the falling ratio of all paramedical personnel to doctors over time. Thus, while there were 5.9 nurses to each doctor in 1980, the nurse:doctor ratlo was merely 3.5 in 1990. The ratio of midwives to doctors fell from 1.08 to .59. The increasing intensity of physicians in the health personnel mlx is not only expensive (since the cost of trainlng physicLans is much larger than the cost of training nurses or mLdwives) but also inapproprlate, since the dlsease pattern in Viet Nam does not call for a heavily intense use of phyelcian services. Indeed, for the type of infectious and parasitic diseases that are endemic to (and leadlng causes of mortality and morbidlty in) Vlet Nam, a relatlvely more lntenslve use of paramedlcal personnel inputs would be more relevant. - 84 - Table 4.6t Availability of health personnel, 1945-90 Population per Year Ass't. Doctor Doctor Nurse Midwife 1945 180,000 60,395 7,482 42,698 1955 126,685 24,302 1,334 6,737 1965 11,973 2,270 469 1,227 1975 3,816 907 512 2,547 1980 4,122 1,835 698 3,834 1983 3,572 1,711 713 4,079 1985 3,137 1,509 721 4,053 1990 2,694 1,381 760 4,568 Notes: Data for 1945-1975 are for the Democratic Republic of Viet Nam (i.e., North Viet Nam). Source: Data for 1945-1975 and 1983 are taken from Ulrich Vogel, "The Whole of Viet Nam can be Considered as One Well-Designed Project: Some Reflections on Primary Health Care Experiences in Viet Nam," unpublished master's thesis at the University College of Swansea Center for Development Studies, University of Wales, 1987, p. 34. Data for 1980, 1985 and 1990 were obtained from the Ministry of Health. 4.42 Comparison with Other Countries. At 2,694 persons to a doctor, the populationtphysician ratio is lower in Viet Nam than in most Asian countries (with the exception of Korea, Malaysia, China and India.) With a ratio of one nurse for 760 persons, Viet Nam surpasses all Asian countries, with the exception of Thailand and Korea, in the per capita supply of nurses (Table 4.7). - 85 - Tablo 4.7: Population per physician and nurse, Asia, 1984 Population Population Nurses to Country per physician per nurse physicians Bangladesh 6,730 8,980 0.7 China 1,000 1,700 0.6 India 2,520 1,700 1.5 Indonesia 9,460 1,260 5.4 Korea 1,170 590 2.0 Malaysia 1,930 1,010 2.0 Myanmar 3,740 900 4.2 Nepal 32,710 4,680 7.0 Papua New Guinea 6,160 890 7.0 Philippines 6,700 2,740 2.4 Sri Lanka 5,520 1,290 4.3 Thailand 6,290 710 8.9 Viet Nam 2,694* 760* 3.5* Notess *Data aro for 1990 Source: Charles C. Griffin, "Health Sector Financing in Asia," Internal Discussion Paper, Asia Regional Series, Report No-. IDP 68, The World Bank, August 1990, p. 33. 4.43 Table 4.7 shows that Viet Nam also has a better mix of nurses and physicians relativo to most Asian countries. However, at 3.5 nurses to each physician, Viet Nam's nurse:physician ratio is comparable to ratios observed in developed countries, where the heavily curative-based systems of health car require relatively intense use of physician services. Since the leading cause- of morbidity and mortality in Viet Nam are preventible diseases (primarily, infectious and parasitic), which typically do not need physician- intensity, there may be scope for substantial cost saving by relying more on a personnel mix with heavier emphasis in community health workers, nurses, midwives and assistant doctors. 4.44 Provincial Distribution of Hoalth Personnel: As in the case of health facilities, there is substantial disparity in the allocation of health personnel anross provinces. The population:physician and population:nurse ratio tends to be much smaller in the North, particularly in the Red River Delta provinces, and significantly higher in the South (particularly, the Mekong Delta provinces) (Statistical Annex Table 4.5). - 86 - 4.45 The relatively low supply of health personnel in the South iu not a post-reunification phenomenon, although Lt is likely that large-scale out- migration of health professionals LimmdLately after reunification in 1975 exacerbated the shortage of health care personnel. As early as 19G4, an American study enumerated 6,000 physicians practilcing in the North and only 750 physicians in the South, with most of the latter practicing in Saigon.f7/ About the same time, another report noted that South Viet Nam had "one of the most severe doctor shortages in Southeast Asia. Of approximately 800 practicing physicians, some 500 serve in the army, and another 150 are in private practice in Saigon. Thus about 150 doctors, or 1 for about every 100,000 persons are available for the rest of the country. "78/ 4.46 TraLnLna There are eight medical schools in the country: four in the north, two in the center, and two in the South. About two-thirds of all Vietnamese doctors have been trained at the Hanoi Medical School, the oldest and most prestigious in the country. In 1991, only 200 of the 4,250 applL- cants who applied to the School were admitted to the medical training program. 4.47 The *ix-year undergraduate curriculum has four modules: basic sciences (10% of-total curriculum hours), medical sciences (30%), clinical medicine (45%), and social medicine (15%). The first two years are devoted to basic and medical sciences (e.g., biochemistry, statistics, physics, anatomy, microbiology, and pathology)s the remaining four years are devoted to clinical medicine and social medicine. The clinical medicine curriculum includes semiology and clinical practice, whereas the social medicine curriculum includes public health and health sector organization. Postgraduate studies consist of master, specialty, and doctorate programs. Typically E-10% of those graduating from medical school pursue a 3-year master's degroe program; another 15% pursue specialty studies (firet-degree specialty is a fGur-year program in any one of the four basic specialties; the second-degree specialty is an additional two-year program to obtain a sub-specialty). About 10 to 20 students per year enter the four-year doctorate program of the Hanoi Medical School. Public health is a specialty of four years, but very few students apply due to its low status. 4.48 The Hanoi Medical School has a staff of 960 persons, of whom only one-half are full-time. There are 60 full professors in the School. The faculty/student ratio of one to two is remarkably favorable, but the total annual budget per pupil of the School was merely US$80 in 1991 -- 30% less than in 1990. Thie budget includes a US$36 subsidy (scholarship) given to 80% of the students. The other 20% of the students pay a US$24 annual tuition. The basic annual salary for a full-time faculty member in the School is 2/ Walter Reed Army Institute of Research, "Democratic Republic of Viet Nam, North Viet Nam," Walter Reed Army Medical Center, Washington, D.C.: Health Data Publicat$ons, October 1966; and Walter Reed Army Institute of Research, "Democratic Republic of Viet Nam, South Viet Nam," Walter Reed Army Medical Center, Washington, D.C.s Health Data Publications, January 1966. 2l/ H. H. Smlth, et &I., Area Handbook for South Vlet Ham, Washington, D.C.% U.S. Government Printlng Offlce, 1967, p. 129. - 87 - US$156; in addition, faculty members receive a bonus of US$48 and a housing subsidy. The food subsidy of 15 kgs. of rice and one-half kg. of meat per month was suspended in 1989. Full-time faculty members are not allowed to have private practice. The MOH sets the salary scale and pension require- ments. 4.49 There are two schools of traditional medicine, one in Hanoi and one in Ho Chi Minh City, both of which train assistant doctors. The faculty of the Ho Chi Minh school constitutes 5 full professors, 10 teachers with post- graduate training; 10 university graduates, and 10 assistant doctors; in 1991 there were 300 applications, 100 admittances and a total student population of 1,200 students. In 1990, it graduated 50 doctors, 100 assistant doctors, and 200 traditional practitioners. The training periods are 6 years for traditional medicine doctors and 3 years for assistant doctors and traditional practitioners. After 5 years of practice, an assistant doctor can apply for three additional years of training to become a traditional medicine doctor. Provinces send their candidates and pay the student tuition (US$20 annually) and board (US$10 per year). A full-time professor has an annual base salary of US$120 plus a bonus of US$0.40 per hour of teaching. An instructor earns US$60 annually and a US$0.10 bonus per hour of teaching. Faculty members are allowed to have private practice. 4.50 Until recently, there were no schools of public health in Viet Nam, reflecting the low priority given to training in public health and community medicine in the country. Now, there are two schools of public health, one in Hanoi and the other one in Ho Chi Minh City. There are English training centers incorporated to these schools that are supported by the World Health Organization (WHO). Both schools are underfunded, and require substantial support to increase the number of post-graduate trainees. 4.51 A generalized problem in the training of health auxiliaries, technicians, and professionals is the serious weakness in health management, community mobilization, and screening for health. Epidemiological approaches to confront health problems in a community are strongly needed in all health training programs. 4.52 Remuneration and Productivity of Health Workers. Health workers are employed by the state immediately after graduation, and are placed into a 3-5 year mandatory practice at a site selected by the MOH in consultation with state health facilities, agencies, bureaus, and factories. Only those completing the mandatory practice can take a residency entrance examination, the successful completion of which is required before entering a residency program. Only 10-12% of graduating students are selected into residency programs. Typically, about 10% of students completing post-graduate training are selected for teaching positions. 4.53 Most health workers typically stay in the same job for their entire careers because of difficulties in obtaining transfers, glut of physicians in the public sector, and an incipient private health sector. Emerging economic changes and greater autonomy in setting wage policies may lead to a greater liberalization of the job market and less retention on the same job. 4.54 Basic salary scales for different health workers are set by the NOH. A new directive (numbered 923) established that all staff working at - 88 - grassroots level will be paid by the district, which has the choice of increasing the salary level. Despite this directive, salaries of health workers ar- abysmally low, even by Vietnamese standards. In the 1991 MOH survey of health providers, physicians at commune health centers reported earning an average monthly salary of 57,500 dong (about US$7.50 in April 1991) (Table 4.8). Assistant physicians reported an average monthly salary of 38,000 dong (US$5.00). Given the average daily hours of work reported (and assuming a six-day work week and a 25-day work month), the hourly wage rate of doctors and assistant doctors works out to be 263 dong (US$0.04) and 154 dong (US$0.02), respectively. Staff at commune health centers who reported having a private medical practice (approximately one-third of the physicians and one- fourth of the assistant doctors) reported earning an hourly wage rate in this practice that was 2% - 4 times greater than their public-sector wage. Physicians and assistant physicians in private health clinics also reported earning monthly incomes that are 3-4 times the public-sector salaries, viz., an average monthly income of 271,240 dong (US$36.20) for physicians and 103,850 dong (US$13.80) for assistant physicians. 4.55 The statistics on numbers of health personnel in Viet Nam, like those on the number of health facilities, can be misleading since little information is available on how much time health workers actually put in delivering health services. Since salaries for health staff have been falling during the last 3-4 years, many health workers have to hold second and third jobs in order to make a living. The problem is most acute in rural areas, where a large number of health workers also engage in farming to augment their earnings. 4.56 In the 1991 MOH survey, all categories of health workers at public health facilities reported long hours of daily work. On average, doctors at commune health centers reported working 8.4 hours per day, while midwives reported working over 10% hours per day. At the same time, about one-third of the surveyed physicians and one-fourth of the assistant physicians at commune health centers reported spending more than two hours each day in supplementary private medical practice (Table 4.8). Since most health workers also hold other (in many cases, full-time) jobs outside the health sector, the data on average daily hours of work at the public health facilities are likely to be grossly overstated. The MOH survey of health facility users showed each commune health center with a staff of about 5 health workers seeing an average of only 6 patients a day 79/ (Table 4.9). Such a low daily case load is at variance with the 8-9 hours of work reported by the providers. 21/ In contrast, private physicians and traditional healers, operating essentially by themselves, reported attending to more than 6 patients each day. - 89 - Tablea 4.8: Averag working hours, n rur of patients examined onthly selary nd time In private practice, by type of heatth provider ;aj typ, of clInIc 1991 CoISSNE HEALTH CENTERS Ass't Traditional Variable Doctors Rctors NurTse fdiwivls HetLers Average daily working hours 8.40 9.50 9.17 10.50 7.18 Nuer of patients seen in uweek 55.32 44.68 35.60 27.99 33.22 Avg. no. of patients sean per hour of work 1.10 0.78 0.65 0.44 0.77 Monthly salary ('000 dong) 57.52 38.13 30.38 32.81 39.25 Avg. hourly wage rate (dong) 263.44 154.40 127.42 120.24 210.34 X having private practice 33.3K 24.5X 17.4X 20.5X 57.1X PrIOvte DraStIce: No. of pati nts In a week 15.10 10.55 8.37 7.37 12.56 Avg. daily working hours 2.00 2.42 1.94 2.30 2.75 Avg. no. of patients seen per hour of work 1.26 o.73 72 0.53 0.76 Monthly lncome from practice ('000 dbng) 32.22 40.38 19.67 27.12 43.67 Avg. rate charged per patient (dong) 492.41 883.43 542.54 849.57 802.20 Avg. hourly wage rate (dong) 619.62 640.49 390.75 452.96 610.77 S ple size 31 449 187 157 32 INTERCO)MUNAL POLYCLINICS Asset Traditional Variable Doctors Doctors Nurses Midwives Healers Average daily working hours 9.06 9.25 9.10 10.56 9.08 Number of patients seen In a week 45.54 38.90 39.38 55.20 42.46 Avg. no. of patients seen par hour of work 0.84 0.70 0.72 0.87 0.78 Monthly slaery ('000 dong) 51.03 42.97 37.09 38.16 39.00 Avg. hourly wage rate (dang) 216.68 178.72 156.78 138.99 165.25 K having private practice 41.7K 23.1K 12.6K 10.4K 36.4X PrIvate practice: No. of patients in a week 12.97 8.49 8.21 7.20 16.00 Avg. daily working hours 2.30 2.01 1.42 1.75 4.50 Avg. no. of patients seen per hour of work 0.94 0.70 0.97 0.69 0.59 Monthly income from practice ('000 dong) 35.73 21.31 23.50 28.00 60.00 Avg. rate charged per patient (dong) 635.66 579.17 660.20 897.44 865.39 Avg. hourly wage rate (dong) 597.49 407.95 638.01 615.39 512.82 saple size 87 263 124 50 13 PRIVATE HEALTH PROVIDERS Asslt Traditional VarIeble Doctors Doctors Healers Averaee daily working hours 7.18 8.08 9.45 Niber of patIents seen In a week 49.93 33.73 55.89 Avg. no. of patients seen per hour of work 1.16 0.70 0.99 Monthly Income from practice ('000 dong) 271.24 103.85 119.17 Avg. rate charged par patient (dong) 1,253.52 710.44 492.08 Avg. hourly wage rate (dong) 1,452.67 494.21 484.98 Splt size 46 15 62 Notes: A six-day work week nd a 25-day work month have been assumed in calculating productivTty and hourly wage variables. Source: MON Survey of Health Users nd Providers, 1991 4.57 Frequent absenteeism from jobs, low morals and low productivity are some of the consequences of low wages in the public sector. Health workers in public facilitles openly report low salaries as a serious problem in the - 90 - delivery of health services. In the 1991 MOH survey of providers, for instance, 54.5% of the health workers in commune health centers and 65.8% of the workers in polyclinics reported inadequate salaries as a "very severe" problem in the delivery of health services. No other problem, including that of inadequate drugs and supplies and of nonfunctional equipment -- received such a high rating. However, in spite of the low salaries, Viet Nam has not yet experienced the serious attrition in health staff at primary health centers that China experienced during its reform of the health sector. For example, owing largely to the withdrawal of local funding for their positions, particularly in the poorer regions of the country, the number of barefoot doctors in China decreased from 1.8 to 1.2 million between 1978 and 1982 -- the first years of economic reform in China.8Q/ 4.58 Private Sector Health Providers. By all indications, private health services, although still incipient, have been multiplying rapidly in recent years. The 1991 MOH survey of health providers found a relatively large proportion of public-sector physiciar. (one-third of those in commune health centers and nearly 42% of those in pelyclinics) engaged in private practice for an average of 2-2% hours per d&y. In addition, the survey discovered a large number of full-time private physicians and traditional healers operating in both rural and urban areas. These practitioners were doing brisk business, as evidenced by the number of patients they examined during the three days they were observed by the MOH survey team (Table 4.9). Since the staffing at private health facilities was meager (with only slightly more than one staff member per facility), the average daily number of patients examined by a private health worker was over five times as large in the public facilities. 4.59 The OH survey of health facility users can be used to examine the socioeconomic and demographic profile of patients using private health providers. The latter tend to be older and have more schooling and higher incomes than users of public health services. For example, patients visiting private physicians had incomes that were more than 25% higher than those of patients using commune health centers (Table 4.10). About 4.5% of the patients using private physician services had college education; in contrast, the corresponding figure for commune health centers was merely 1.3%. The public health facilities attended to more pregnant women than the private facilities, indicating that people tend to obtain preventive care (in particu- lar, prenatal checkups) at public health facilities, and use private health services generally for curative care. In terms of occupational backgrounds, users of public health facilities typically were farmers (both cooperative and private), while private farmers, professionals, factory workers, housewives, and students figured prominently among users of private facilities. Finally, the fact that patients traveled two times as far and paid over two times as much for a private health care visit as for a visit to a public health facility implies that they must perceive the quality of private health providers to be significantly better than that of public health providers. 80/ Joan Kaufman and Gita Sen, "Population, Health, and Gender in Viet Nams Social Policies under the Economic Reforms," Center for Population Studies, Harvard University, 8 September 1991, p. 25. - 91 - Table 4.9: Average daily came load at health facilities, 1991 Average number of pati- Average number of Type of facility ents examined per day health workers Commune health center 5.8 4.6 Intercommunal polyclinic 7.8 7.2 Private physician 6.3 1.1 Traditional healer 6.3 1.2 Source: MOH survey of health facility users and providers, 1991 Traditional Medicine 4.60 The Vietnamese government has placed strong emphasis on the use of traditional medicine. In part, this emphasis has arisen out of necessity. The international isolation and severe resource crunch that Viet Nam experi- enced forced it to accord an important status to traditional medicine, which makes good use of the abundant supplies of local herbs and flora. Unlike the situation in other countries, where traditional medicine is merely part of the folklore, traditional medicine is fully integrated within the health system in Viet Nam and is placed on an equal footing with Western medicine. For example, all medical and health personnel receive training in traditional medicine; in addition, all practicing assistant doctors at commune health centers receive continuing education in traditional medicine; traditional doctors or healers are included on the staff in national, provincial and district hospitals; and the Institute for Traditional Medicine is one of the central specialty institutes financed by the MOH. As a result, traditional medicine is actively practiced in the country. Indeed, integrated treatment using both Vietnamese and Western medicine is the rule, rather than the exception, in most health facilities. Safe Water and Sanitation 4.61 Health Consy"q2£g. The mortality and morbidity profile in Viet Nam is characterized ': .vily by diseases that are linked to water supply and sanitation. For example, malaria and dengue fever are among the important causes of mortality, while malaria is a leading cause of morbidity as well. Both of these diseases are spread by mosquitoes that breed in stagnant water ponds and inadequately covered water tanks. Excreta-related and water-borne diseases, such as gastroenteritis, dysentery, typhoid, cholera, and viral hepatitis, are also important sources of morbidity, especially among children. Almost 60% of the total reported diarrheal cases in 1986 in Viet Nam occurred in chLidren under 5 years of age.81/ An examination of clinic records at commune health centers in Yen Phung district revealed that 41% of all treat- ments offered were for diarrhea. Thus, any attempts to improve health jJ/ World Health Organisation, "Programme for Control of Diarrhoeal Diseases, Sixth Programme Report, 1986-87," Geneva, 1988, p. 5. - 92 - conditions in Viet Nam, especially those of children, should necessarily include foreoful interventions in the supply of safe water and sanitation. Table 4.10: Characteristics of Patients Using Public and Private Health Services, Viet NM, 1991 Conne- ELtire health Intercommunal Private Traditional Varidble SaLt centers .YjAtlnjcs Physicians bgtolrs_ Nteber of observatfons 7,294 3,393 1,941 1,165 795 Household size 5.127 5.145 4.878 5.379 5.285 Total household income (dong) 213,434 204,301 198,467 257,316 225,646 Age of patient (years) 29.759 29.048 27.529 30.154 37.644 Prooortion of fndividuals who have comoleted: primary school 0.494 0.495 0.538 0.453 0.436 secondry school 0.203 0.176 0.178 0.310 0.219 college 0.020 0.013 0.016 0.045 0.024 Promortiqn of individuals who are: cooperative former 0.222 0.273 0.246 0.103 0.116 private terer 0.297 0.301 0.359 0.211 0.255 factory worker 0.062 0.050 0.057 0.090 0.083 teacher 0.034 0.029 0.043 0.042 0.021 professional 0.024 0.018 0.024 0.034 0.040 artisan 0.029 0.030 0.018 0.040 0.035 construction worker 0.010 0.009 0.007 0.009 0.020 hous"wife 0.043 0.041 0.018 0.067 0.081 student 0.061 0.054 0.060 0.102 0.035 driver 0.009 0.008 0.004 0.015 0.021 uneeployed 0.041 0.043 0.020 0.064 0.055 Procortlon of individuals who ore: _el 0.410 0.401 0.417 0.439 0.393 pregnant women 0.082 0.095 0.112 0.037 0.023 nursfng mothers 0.042 0.046 0.040 0.045 0.025 Distance traveled (kms.) 3.085 1.452 2.382 5.914 7.603 Payent for visit (VND) 4.830 3.403 4.279 7.897 7.772 Source: NON Survey of Users, 1991. 4.62 Ponulatilon CoveraQQ. Although the Government has waged mass campaigns to encourage construction and utilization of wells, rainwater tanks, latrines and bathrooms over the last 35 years, a significant majority of the Vietnamese population remains without access to safe water and sanitation. It is estimated that. only about 54% of the urban population has access to safe drinking water.82/ The remainder obtain drinking water from shallow wells that are typically unprotected and highly contaminated, rainwater catchment tanks, rivers and ponds. Even in the towns having access to piped water, the facilities are poorly maintained and in a state of disrepair. In Hanoi and Ho Chi Minh City, water treatment facilities are adequate, but water is sometimes distributed untreated. Lack of chlorination, excessive leakage (approaching nearly 50), and problems with sewage disposal, among other things, make the dellvery of water in urban areas unsafe. IV United Nations Development Programme, "Report on the Economy of Viet Nam," December 1990, p. 188. - 93 - 4.63 According to the MOH, 33% of rural households have a well or rainwater collection tank.83/ Only 40% of these facilities, however, meet the MOH's quality standards. Furthermore, it is not clear whether the wells can be conoidered as safe supplies of water. Typically, they are shallow and easily prone to contamination. The rainwater collection tanks do supply clean water, but only in such small quantities that it can be used for drinking and cooking alone. Water for personal use and washing is generally obtained from contaminated sources. 4.64 The official statistics on water and sanitation show virtually no increase in the proportion of housing units with safe water and sanitation during the last 15 years. The percentage of housing units without safe water supply has hovered around 60% in urban areas and 30% in rural areas since 1976. In urban areas the proportion of housing units without sanitation facilities is small (around 2%), but in rural areas this figure has remained around 40% since 197644]J 4.65 Administration and Organization. One of the problems with the supply of drinking water and sanitation in Viet Nam is that no single agency has authority over this subject. Among the four ministries that have jurisdiction over water and sanitation are (i) the Ministry of Construction, which is responsible for designing and implementing urban water supply and sanitation systems, (Li) the Ministry of Health, which is responsible for implementation of rural sanitation programs and quality control of drinking water in urban and rural areas, (iii) Ministry of Water Resources, which looks after the development and management of surface water resources (typically flood control and irrigation), and (iv) the Ministry of Labor, Social Welfare and Invalids, which coordinates implementation of rural water supply projects (financed by UNICEF). The dispersion of responsibility, combined with the lack of coordination, among the various agencies makes it difficult to develop a coherent plan for improving water quality and sanitation. Pharmaceutical and Other Medical SupDlies 4.66 Viet Nam has produced some drugs and medicines since colonial times; however, the domestic pharmaceutical sector in Viet Nam is unable to meet domestic demand. Most essential drugs. (e.g., antibiotics and drugs against tuberculosis, infectious diseases, malaria, and parasites) are imported. one of the main objectives of the Government's Development Plan for 1990-2000 is to expand the domestic production of drugs and pharmaceutical and has established a goal of eventual self-sufficiency, at least in essential drugs. Viet Nam produces vaccines such as BCG and DPT; nevertheless, it still relies on UNICEF for a large part of its supply of these vaccines. Positive factors condueive to an expansion of domestic production of pharmaceutical a/ Of course, this is a big improvement over the situation in 1955, when fewer than one percent of rural households in the country had a well or rainwater storage facilities. See UNICEF, Viet Nam: The Situation of Children and Women, UNICEF, Hanoi, 1990. 94/ National Centre for Social Sciences of Viet Nam and Centre for Women's studies, SelScted Indicators on Womer Status in Viet Nam 1975-1989, Statistical Publishing House, Hanoi, pp. 109-111. - 94 - include (a) the country's wealth of vegetation which can be used in the production of traditional drugs and medicines, (b) a cheap labor supply, and (c) in the late 1990s, when several large hydroelectric projects will be completed, an abundant source of relatively cheap energy. 4.67 Growth and Production. Viet Namas pharmaceutical industry consists of 12 manufacturing units producing traditional Vietnamese medicines and a total of 54 manufacturing units producing western drugs. Of these, 46 are provincial and 20 are national units. 4.68 The total output of western medicine in Viet Nam in 1989 amounted to 3,107 million dong (3.3 million dong at constant 1982 price.) (Statistical Annex Table 4.6). This was a mere 1.5% of the corresponding output in 1980. There was a sharp decline in the output of pharmaceutical products during the 1980s, primarily due to a general lack of financial resources and the consequent inability to import raw materials and spare parts for equipment. It is estimated that, at the present time, domestic pharmaceutical factories are producing at only about 40% of capacity. 4.69 Four important characteristics have shaped the growth of pharmaceutical in the past and will continue to shape it in the near future. First, the availability of root plants and extremely cheap labor; second, acquisition of spare parts and equipment; third, the pricing and promotional policies for setting up local manufacturing facilities; and, fourth, government policy toward joint ventures. In the past, the establishment of packaging and dosage formulation facilities in the North (Hanoi), Central (Quang Nam Danang) and South (Ho Chi Minh City) has been carried out with varying degrees of success. But, owing to heavy reliance on foreign technology and imports of some raw materials, the savings in cost and foreign exchange are lower than might be expected. Most essential drugs, e.g. antibiotics and drugs against tuberculosis, infectious diseases, malaria and parasites, are currently imported. 4.70 Institutional Framework. Viet Nam has a relatively few nuAber of agencies involved in the pharmaceutical subsector. Responsibility for overseeing the pharmaceutical industry is shared by the MOH and Viet Nam Pharmaceutical United Enterprise (VINAPHA). The MOH has jurisdiction over formulation of policies regulating drug use and administration, quality control, supervision, license and registration. The VINAPHA, which was created in 1982 by the MOH, is directly responsible for all aspects of production, supply, exports and imports and R&D activities. With its head office in Hanoi and branch office in Ho Chi Minh City, VINAPHA oversees all national and provincial pharmaceutical enterprises. 4.71 Of the 54 pharmaceutical units in the country, 46 are provincial. Thus each province controls and manages at least one pharmaceutical factory. In addition to producing drugs and medicines themselves, the provincial companies supply inputs to the national enterprises and distribute drugs produced by the national enterprises in their respective provinces. The 20 national enterprisoo include 10 pharmaceutical factories controlled directly by VINAPHA, three trading companies specializing in medical supplies, two pharmaceutical wholesale companies, one transportation and distribution company for imported drugs, medicines, and medical supplies, one - 95 - pharmaceutical import-export company (VIMEDIMEX), and three plantations producing medicinal plants and herbs. 4U 72 Consumotion and Distribution. No detailed statistics on consumption of pharmaceutical by main therapeutic groups are available. At present, the per capita consumption of pharmaceutical products amounts to roughly US$1.00. These rates are too low, and reflect the widespread unavailab$lity of pharmaceutlcal products in the country. In addition to inadequate levels of production and imports, the dLstribution system for drugs ic also weak and is hampered by the lack of proper transportation between the central storage and district hospitals, poor refrigeration facilities, improper warehousing, and inadequate management of drug inventories. The situation is worse in the remote mountainous regions, where the difficult t-rrrain and isolation hinder distribution of drugs. Realizing this, the Government has initiated the Program of Price Support on Essential Drugs for Mountainous Villages, covering some 2 million persons, in 1992. Under this program, the Government pays the transportation costs for drugs, so that drugs can be available to consumers in these regions at the same prices as in the plains. 4.73 Sales. Imports and Exoorts of Druos. Since July 1989, the Government began to allow pharmaceutical factories to open retail pharmacies and sell drugs and medicines to individuals, hospitals and commune health centers directly. Commercial sales have increased in 1990, with individuals who can afford to buying drugs at market prices and those unable to do so receiving them free through the public sector. Also, patients who can provide a certlficate from their commune People's Committee that they are too poor to pay for drug. and medicines still receive them free. The introduction of this system has enabled the pharmaceutical factories to recover some portions of costs of production and has reduced the drain on the Government subsidy to the factories. Imports of medicine and drugs amounted to US$44 million in 1989 and fell to US$41 million in 1990 mainly due to the shortages of foreign exchange and drop in supplies from non-convertible countries at concessional prices. Viet Nam exports some pharmaceutical products to China and East European countries. But exports fell steeply from US$10 million in 1989 to US$6 mLilLon in 1990 due to the recent economic and political changes in the Bast European countries. Enterprises which finance their exports themselves are allowed to retain the foreign exchange earned from exports. Foreign exchange is retained by the Government if it finances the exports and pays an equivalent amount to the enterprise in local currency. 4.74 Recent Policy Reforms. The Government recently introduced two reforms having a major impact on the pharmaceutical industry. The first reform, introduced in April 1989, gave pharmaceutical factories the freedom to decide on the levels of input use, output, and sale prices of drugs and medi- cines. Pharmaceutical factories were also informed that they would have to cover costs or else close down (i.e., they could no longer rely on government subsidies). Initially, most firms had a difficult time. In an ideological context in which for years workers had been guaranteed employment, it was vlrtually impossible for enterprises to discharge the large numbers of redundant workers. Yet, in the same month as the reform was lntroduced, some 400 workers in the pharmaceutical industry were induced to take early retire- ment. All received a lump sum of 1-1.5 million dong (then US$250-400) from VINAPHA and a guaranteed monthly pensLon from the Ministry of Labor averaging - 96 - about 70% of their monthly salary. During the period Lmmediately after the reform, most of the pharmaceutical enterprises also raised prices for their products. The prices initially set by the enterprises, which simply reflected the high production costs, were too high, and it was difficult for the enter- prises to find buyers for their products. 4.75 The situation has improved since July 1989, when the Government introduced the second reform having a major impact on the pharmaceutical industry, viz., pharmaceutical factories were allowed to open retail pharmacies and sell drugs and medicines directly to hospitals, health centers, and private individuals, rather then through government-dictated contracts. The introduction of this measure has allowed pharmaceutical factories to recover their costs of production; firms have become more efficient, costs and prices have declined, and sales have increased. Faced with competition, both from other firms and from the imported drugs now flowing into the country, domestic firms have also improved the quality of their products. Although redundant labor is still a problem, most firms appear to be adapting well to the new economic context. 4.76 Medical Eauioment. Suoplies. and Instruments. Medical equipment and supplies are produced in government enterprises grouped together in an agency similar to VINAPHA -- The General Company for Medical Equipment and Sanitary Construction -- which is also a subsidiary of the Ministry of Health. This agency consists of seven national factories, including a factory producing metal and steel medical instruments, located near Hanoi; a repair factory for medical equipment, also located in Hanoi; a condom factory, located in Ho Chi Minh City; a Hanoi-based Center for Research in Medical Instruments (also used for technician training); two companies, located in Hanoi and Ho Chi Minh City, producing medical equipment and laboratory chemicals; and a company for sanitary construction (i.e., construction of hospitals and other facLilties belonging to the central government network). 4.77 The medical instruments factory was built with assistance from China and began production in 1975. Most of the equipment in use in the factory was imported from China in the early 1970s; all of the equipment is very old and needs to be replaced. The factory employs about 1,000 staff in ten separate workshops. It is the only factory in Viet Nam to produce medical equipment from steel and other metals. The product line includes basic instruments used for primary health care as well as more sophisticated instruments used for specialized health care, e.g., kits for obstetrical, gynecological, and otorhinolaryngological examinations and treatment. Instru- ments are produced not from stainless steel but instead from normal steel plated with nickel or chrome. As a result, the instruments are hard to sterilize (especially in a tropical climate), rust easily, and are uneconomi- cal in the long run. The factory has begun to produce, on a pilot basis, a very small output of instruments made with stainless steel. 4.78 The experlences of the medical equipment factorles with respect to the policy reforms, particularly involving redundant workers, have been simLlar to those of the pharmaceutical factoriess the services of some 300 redundant workers were temporarily terminated immediately after April 1989, but 120 of these workers have slnce been rehired. The situation of the factories Li still dLfficult but steadily improving. Market competition, particularly from imports, has forced them to lmprove the quality of their - 97 - products, such that they now are meeting international standards, and they hope soon to begin exporting simple medical instruments. Z. Utilization of Health Service. 4.79 While the provision of primary health services may well be a necessary condition for improved health status of a population, it is not sufficient. For the health care system to have an impact on health, individuals and households need to utilize health services effectively. A rough measure of overall utilization that is often used is the annual number of per capita contacts with the health services. 4.80 An adequate level of annual health contacts per capita depends on the age and sex distribution of the population as well as on incidence of morbidity and access to the health system. However, it is thought that an average of three to four annual contacts with the health services are adequate in achieving basic preventive health care goals. For instance, this level of contact with mothers and children would assure a high level of immunization of the child population and proper monitoring of pregnancies and deliveries. A few developing countries, such as China, Tanzania and Sri Lanka, average 4-D contacts per year.85/ Annual Per Capita Contact Rates 4.81 The calculation of per-capita contact rates requires information not only on the number of individuals visiting, say, health centers and private physicians, but also on the total number of all eligible users, viz., the entire population of the catchment area served by the health providers. Unfortunately, few countries, including Viet Nam, collect statistics on overall health services utilization that relate to an area-wide population base. Using the data on the total number of medical consultations that the MOH reports, a rate of about 1.22 consultations per person per annum is obtained for 1989 (Statistical Annex Table 4.7). This contact rate is observed to be down sharply from rates exceeding 2 during the period 1980-87. However, it is not at all clear how reliable the national data on number of consultations are in view of the fact that Viet Nam has no organized system of monitoring and evaluating the activities of commune health centers. 4.82 The survey of health facility users and providers undertaken by the NOH in three provinces in 1991 suggest outpatient contact rates with the primary health facilities to be very low -- in the range of 0.3 - 0.5 annual contacts per capita. Another survey of 3,502 women in seven provinces under- taken by the Center for Human Resources (MOH) in 1990 found 50.8% of the women reporting no contacts with the health services for any of their family members. In the mountainous northern province of Lang Son, 74.5% of the mothers reported no medical contacts. These rates of contact are Ml/ Oscar Gish, 1989, "Some Links Between Successful Implementation of Primary Health Care Interventions and the Overall Utilization of Health Services," Social Scence and Nedlcine 30 (4): 401-405. - 98 - significantly lower than those observed for some other developing countries,86/ and lower than the 3-4 annual contacts with the health services that are thought to be necessary to achieve basic preventive health care goals. Indeed, given the unequal distribution of service contacts in a developing country, an average annual contact rate of 0.3 - 0.5 implies that a significant majority of the Vietnamese population is effectively outside the formal health care system. Bed Occupancy Rates 4.83 The utilization of hospital health services in Viet Nam also appears to be low. Average occupancy rates for hospitals and primary health care centers (viz., commune and urban health stations) for 1979 and 1989 reported by the MOH are shown in Statistical Annex Table 4.8. The bed occupancy rate for hospitals ranges between 40 and 60% in most provinces. The rates for commune health centers and polyclinics are even lower. Further, while there was a modest increase in the occupancy rate for hospital beds between 1979 and 1989, there was a decrease in the rate for basic health centers. The provincial differences are strikings while TP Ho Chi Minh had an average occupancy rate for hospital beds of 79.9% in 1989, the Central Coast Northland province of Thanh Hoa had an occupancy rate of only 39.2%. The bed occupancy rate for basic health centers ranges from 4.4% to 57.5% among the provinces. 4.84 Interestingly, the data in Statistical Annex Table 4.8 suggests an inverse relationship between the occupancy rate in hospitals and that in basic health centers. For example, TP Ho Chi Minh, which has one of the highest hospital bed occupancy rates in the country, has the lowest bed occupancy rate for its basic health centers (only 4.4% in 1989). These statistics reflect the poor performance of the referral system (as well as of primary health cen- ters), since they show that individuals residing in provinces having high quality hospitals typically bypass the primary health centers and admit themselves directly into hospitals. Another reason for the very low bed occupancy rates of basic health centers across the board has to do with logistical support and the way funds are allocated to health facilities. Since the allocation of funds is related to the number of beds an institution has, beds are sometimes kept on the books even -vhen they are not effectively available for use. 4.85 In view of the facts that bed occupany rates at inpatient facilities are very low and that Viet Nam already has one of the highest ratios of hospital beds to population, there might be scope for a sharp reduction in the number of hospital beds, especially those in a serious state of disrepair and rarely used. 1_/ For example, surveys in Sri Lanka and in the Sichuan province of China indicate average outpatient contact rates per capita per year of 2.1 and 13.4 visits, respectively. Tanzanians average about four to five contacts with the health services per year. See World Bank, "Indonesia: Poverty Assessment and Strategy Report," Report No. 8034-IND, Country Department V, Asia Region, Washington, D.C., 26 December 1988; and World Bank, "Indonesiat Issues in Health Planning and Budgeting," Report of Country Department V, Asia Region, Washington, D.C. 12 June 1989. - 99 - Utilization by Women 4.86 A women's health survey conducted by the Swedish International Development Agency (SIDA) in 1989 in Vinh Phu and Ha Tuyen provinces (in the mountainous reglon of the North) found that 70% of the farming women and 40% of women working in forestry had not sought a treatment the last time they were ill. More than two-thirds of the surveyed women indicated they had to travel more than one hour for ante-natal care. Although the NOH reports 80- 90% of the deliveries ln Viet Nam taking place in health facilities under the supervision of a trained health professional, the SIDA survey found that one- half of the dellveries during the past five years had taken place at home. In only 6C% of all births was there a nurse or assistant doctor present. The remaining births were attended either by midwives, traditional birth attendants or relatives of the woman. 4.87 Another survey of 3,502 women in seven provinces conducted by the Center for Human Resources (NOH) in 1990 came up with several startling findings. Among these were: - About 56% of the surveyed women reported that their families had never been visited by a commune health worker. Another 14% reported being visited only once during the six months preceding the survey. - Only 2% of the women said that the commune health center would be their provider of choice in the event of an illness. Another 28% said they would go to the intercommunal polyclinic, and fewer than 1% expressed the district hospital as the provider of choice. - Nearly 30% of the women said they had obtained no antenatal care, and 65% said they had not been immunized against tetanus before their delivery. In the province of Lang Son in the mountainous region of the North, the percentage of women not seeking antenatal care was as high as 71% and the percentage not receiving a tetanus immunization was 94.1%. - Nearly 82% of the surveyed mothers reported that their children had not been weighed at the commune health center. An identical number reported not having any growth charts for their children. - Only 58% of the mothers were aware of the diseases that can be prevented by immunization. About 55% of the women were not aware of oral rehydration therapy, and 44% could not identify the symptoms of acute respIratory infection. The generally low utilization of antenatal health services, including anti-tetanus immunization, is at variance with the low officially- reported maternal mortality rates. - 100 - Utilization by Children 4.88 The official data from MOH, reported below in Table 4.11, indicate over 80% of all children under one year being fully immunized since 1989. Such high rates of immunization coverage are generally incongruous with the high (relative to other countries) morbidity rates associated with immunizable diseases in Viet Nam. The incongruity may be the result of ineffective biological quality of vaccines used (due to problems of production and cold chain), an increased death rate among children above the age of one year due to inadequate health care, and/or overestimated immunization coverage rates. There is strong support for the last possibility. In a survey conducted by the MOH in eight communes of Yendung District in Habac Province in 1991, only 25% of the 1,782 surveyed infants under one year of age had been fully vaccinated. Another 18.8% had been partially vaccinated, while the majority (56.2%) had not been vaccinated at all. In another MOH survey of 3,502 mothers conducted in seven provinces in 1990, while 72.8% of all surveyed mothers said their children had received adequate immunization, 35.6% could not tell what an adequate level of immunization was. The survey also indicat- ed wide provincial differences in the mother-reported rate of immunization coverage, Only 47.4 and 57.1% of the mothers reported adequate immunization of their children in the mountainous Northern province of Lang Son and in the Central Highland province of Lam Dong, respectively. These results cast serious doubt on the officially-reported figure of immunization coverage, and are consistent with the relatively high rates of morbidity associated with vaccine preventable diseases in Viet Nam relative to other Asian countries. Table 4.11: Percentage of Infants Reported to be Fully Imaunimed, 1985 and 1989 Coverage (%) accine Protection against: 1985 1989 Diphtheria/whooping cough/tetanus 42.1 88.7 Polio 62.1 87.3 Measles 19.2 88.7 Tuberculosis 50.0 93.6 Source: MOH, 1988, 1990. Determinants of Utilization and Provider Choice 4.89 If utilization rates of primary health services are to be improved, it is important to know the factors associated with low utilization. Analysis of interprovincial data on bed occupancy rates at basic health centers shows occupancy rates being inversely associated with average distance to health facilities (Annex E). Thus, provinces having more health facilities per capita and per unit of area (and thus having smaller average distances to health facilities) appear to have significantly higher occupancy rates than provinces having fewer health facilities. Bed occupancy rates also appear to - 101 - be higher in provinces that have well-staffed facilities. However, when the average number of health workers per facility is separated into two components -- the number of physicians and the number of other health personnel -- the results indicate that it is the staffing of facilities by nonphysician personnel -- not physicians -- that is associated with higher occupancy rates. The implication of this finding is that the mix of health personnel needs to be shlfted to a higher proportion of non-phyaicians to improve utilization performance. Thls ln turn implies that programs for training community health workers need to be expanded. 4.90 The Viet Nam Demographlc and Health Survey (VNDHS) shows a very clear relationship between the educational level of women and their propensity to seek adequate prenatal care (Table 4.12). In 1987-88, for example, nearly two-thirds of the women with secondary schooling -- as opposed to only 21% of illiterate women -- had sought adequate prenatal care during their pregnan- cles. Not surpriLsngly, infant and under-5 mortality rates for illiterate women aro significantly higher than those for women with at least secondary schoollng. Table 4.12 : Mortality and Prenatal Care during 1978-88, by Mother's EducatLon, VLet Kan.......... Mortality Rate (per 1,000 live births) percent of Mother's births with Educational Level Infant Child Under-5 adequate pre- natal care Illiterate 59.6 12.8 71.7 21.0 Read/write 41.9 11.4 52.8 40.0 Primary 32.9 12.5 44.9 54.0 Secondary and higher 25.3 6.3 31.5 61.0 Source: Vlet Nam Demographic and Health Survey, 1988 4.91 Analysis of data from the 1991 MOH survey of health facility users of public and private health services in three provinces, which can be used to analyze the cholce of health providers by Lndividuals,87/ shows high-income flu However, since the survey sampled only individuals utilizing health services (and excluded nonusers), it cannot be used to analyze the factors determinlng utilizatLon. In addition, since the sample was not a random, representatLve sample of tho entire population of the three provinces, the *xerclse should be treated as illustrative. - 102 - and better-schooled individuals relying much more on private providers for their heaith care needs than individuals with low income and low schooling (Annex F). These results imply that improvements and investments in commune health centers and intercommunal polyclinics will be automatically targeted to low-income, poorly-schooled individuals. The same analysis indicates that the demand for health care from commune health centers and polyclinics is not very responsive to the price that is charged at these facilities. Indeed, the results suggest that the shift in demand from public to private health services associated with an increase in user fees at public health facilities will be small enough,88/ so that total revenues at public health facilities will increase. Thus, there may be substantial scope for increasing user fees at health centers and polyclinics associated with improvements in quality. 89/ F. Health Inputs and Health Outcomes 4.92 So far the discussion has focused almost exclusively on the factors determining the utilization of health services. But insofar as the ultimate goal of health policy is to improve health status and reduce mortality -- and not merely to increase the utilization of health services per se -- it is more instructive to analyze the determinants of health outcomes. The link between health status and the utilization of health services is likely to be complex, since the latter is only one of several (possibly substitutable) inputs in the determination and maintenance of health status. For example, adequate provision of safe drinking water and sanitation can substitute, to some extent, for a poor health care system in maintaining the health status of a population. 4.93 Analysis of interprovincial data on infant mortality rates for 1979 and 1989 reveals the importance of access to health facilities and average health expenditures per facility in lowering infant mortality (Annex G). Access to health facilities, as measured by the average distance to a public health facility, is observed to significantly reduce the infant mortality rate in a province. In addition, average government health expenditure per facility is inversely associated with the infant mortality rate in a province. Since higher expenditure per facility is generally synonymous with a better supply of consumables, these results reinforce the importance of drugs, medicines and useable equipment in improving health outcomes. Finally, female literacy has a strong ameliorating effect on infant mortality. 4.94 The strong association of literacy with low infant and child mortality rates is also observed in the Viet Nam DHS data shown in Table 4.12 above. Numerous studies for Africa, Asia and Latin America have documented the immensely beneficial effects of maternal literacy on child morbidity and fl/ The price elasticity for demand is estimated to be -0.23. f8/ This point is discussed in more detail in the section on cost recovery in Chapter V. - 103 - survival outcomes.90/ Viet Nam is fortunate in having high levels of female literacy relative to other countries at its income level. The 1989 Census enumerated 81.8% of the women over the age of 10 as literate. These high rates of literacy are the direct result of the Government's strong commitment to women's education and literacy. Universal primary education was established as early as 1958 in North Viet Nam (and in 1978 in the South). Furthermore, unlike many other countries in Asia, there is no gender discrimi- nation in access to schooling, education or the labor force. Women partici- pate actively in the work force, and there are no formal or informal barriers to the participation of women in the political and administrative structures. G. Health ExDenditure 4.95 In Viet Nam, the public health budget contains four fiscal transactions by levels of government: the central and the three levels of local governments: provincial, district and commune. Since communes do not report their budgets to the Ministry of Health, a comprehensive database which aggregates the four level transactions into a consolidated account of sectoral expenditure is lacking.2l/ However, central, provincial and district governments account for about 90% of the national budget. 4.96 The provinces normally do not receive budgetary support from the central government except under special cases when some provinces need central financing for high-priority programs like malaria control, expanded program of immunization, or literacy campaigns. Districts receive most of their support from the provinces, although they have their own revenue sources. Since the communes are expected to be basically self-sufficient, they are excluded from the integrated national budget. But the neediest communes may receive some contributions from district governments. 4.97 Each year the provinces submit their budgets for approval to the Council of Ministers. The Ministry of Finance negotiates with the provincial governments on the proportion of revenues to be retained by the provinces.92/ After approval from the Council of Ministers, each province is provided with targets for total local revenues (excluding surpluses from previous year), total local expenditures (investment and recurrent), the 29/ S. H. Cochrane, D.J. O'Hara and J. Leslie, "The Effect of Education on Health," World Bank Staff Working Paper Number 405, Washington D.C., 1980; and J. C. Caldwell, "Education as a Factor in Mortality Decline: An Examination of Nigerian Data," Population Studies 33 (3), 1979, pp. 395- 413. 21/ The health services of the commune typically are paid for by the commune itself. Commune health center staff are trained at government expense in secondary medical schools, but the health posts are built by the commune and the small staff salaries are supplemented with food rations from communal production. 21/ Virtually all revenue is collected by provinces on behalf of the central government. - 104 - proportion of central revenue to be retained, and any grants to be received from the central government. The allocation of revenues to districts and communes is largely under the jurisdiction of the provincial governments. 4.98 Cross-Country Comoarisons. In 1990, per capita government expenditure on health war 5,664 dong (US$0.83), constituting about 4.4% of total government expenditure and 0.8% of GNP. While Viet Nam's per capita health expenditure of US$0.83 is considerably lower than that of any other country in Asia, including Nepal and Bangladesh (Table 4.13), its health spending relative to its per capita GNP and total government spending is not unusually low; for instance, Viet Nam spends roughly the same proportions of its per capita GNP and total government expenditure on health as Indonesia and Thailand. These facts indicate that Vietnam is devoting a fair share of national resources to health care, but that its revenue generation performance is generally weak. Therefore, additional health expenditures will have to either be financed through higher general revenues, user fees, or foreign assistance. Tabte 4.13: Government Neatth Expendftures in Asia. 1989 Goversment Health Expenditures in 1989 Country Us# as X of total as X of GNP (per capita) govt. exp. tangladesh 1.26 4.5 0.7 Shutan 2.23 2.6 1.2 China 2.88 4.2 0.8 India 5.04 6.7 1.6 Indonesia 4.51 3.8 0.9 Korea 29.74 2.2 0.6 Lao POR 3.03 4.9 2.0 Malaysia 55.41 6.8 2.7 yarmar 4.40 6.8 1.1 lepal 1.33 4.3 0.8 PhItippfnes 4.42 3.3 0.6 Papua New Guinea 30.14 10.0 3.4 Sri Larka 5.37 4.5 1.3 Thailand 13.64 6.1 1.1 Viet aM 0.83 3.3 0.7 Source: IWortd Bank data files - 105 _ 4.99 Tme Trends. Table 4.14 shows the trends in real government health expenditure between 1984 and 1990 (with Tables 4.9 to 4,11 in the Statistical Annex containing the details). Before analyzing these data, it is important to note again (see para. 2.55) that budgetary data are perhaps the least reliable of the statistics collected and reported by the Government. In 1985, for example, Viet Nam was experiencing hyperinflation, and the Government attempted a disastrous ten-for-one currency swap, introducing the "new dong." Reported government expenditure in all categories is much higher in 1985 than in neighboring years. Hence, the sharp decline in real health expenditure from 1985 to 1987 is probably a statistical artifact. Indeed, what is surprising is the sharp increase (of over 125%) in real health expenditure between 1987 and 1990 (Figure 4.5). Again, these _ numbers have to be GV alith U Failly PIvg Ex 1984-90 interpreted with caution, 2s0 since the method of paying all government workers 240 - changed in 1989, with monetary remuneration i - increasing and payments in at kind (mostly rice) S 160 - decreasing. Since pre-1989 X government expenditure data 120 did not include the implicit 22 value of in-kind payments to 17 government employees, it is likely that post-1989 7o government expenditures are 5 - ~ ~ ~ ~ ~ ~ t~~~96 14s 1986 16i7 19b8 Nsbg I9 artificially inflated. vim 4.100 The most Figure 4.5 likely scenario is that real health expenditure (adjusted for the change in method of compensatlng public employees) remalned falrly stable between 1985 and 1990. But this stability Ln health expenditure is impressive, considering that Viet Nam was experiencing hyperinflation, acute macroeconomic instability, and a sharp drop in total external aid during this period. of course, most of the external assistance that was cut sharply was from the Eastern bloc and lncluded little for the health sector. The modest resumption of Western aid Ln recent years has included a large component for the social sectors and for health in particular. Therefore, it is possible that the Government of Viet Nam has been able to maintain health expenditure largely because of external asistance from Western donors and the internatLonal agencies. 4.101 Reaional Disparities. Data are available for 1979 and 1989 on provincial government health expendltures. These indicate enormous disparity in the dlitributlon of provincial public health expenditures. TP Ho Chi Mlnh City had a per-caplta provinclal government health expendlture of 15,505 dong, while Gla Lai-Kon Tum had a per-capita health expenditure of merely 71 dong (Table 4.12 in Statistical Annex). Indeed, the interprovincial variance in government health expenditure is much greater than the variance in total - 106 - government expenditure and in GDP. 9/ Further, the interprovincial variance in government health expenditure war virtually unchanged between 1979 and 1989. 4.102 CpmDosition of Expenditure. Table 4.11 in the Statistical Annex reports the compositioa of government health expenditure from 1984 to 1990. Three revealing observations can be made from these data. First, recurrent expenditure accounts for a large share of total health expenditure, with investment expenditure having a meager share. For instance, with the excep- tion of 1985, investment accounted for merely 5-7% of total public health expenditure. Within investment expenditure, most of the spending was on equipment and civil works, with hardly anything being spent on rmanpower training. While the share of investment in total health expenditure has begun to improve since 1988, largely because of increased bilateral external assistance in the form of medical equipment, training continues to be underfunded. 4.103 Second, the share of salaries in total recurrent health expenditure is very low in Viet Nam relative to low-income Asian standards. For example, in 1988, salaries constituted only 15.7% of total recurrent health expenditure in Viet Nam.2A/ The corresponding figures for other countries in Asia were: 24.3% for China, 37.5% for Lao PDR, 44.6% for Myanmar, 52% for Indonesia, and 69% for Bangladesh. The low share of salaries in total recurrent health expenditure combined with the high intensity of physicians simply reflects the extremely low salaries of health workers in Viet Nam. 4.104 Third, the ratio of expenditure on curative to that on preventive health services is very high. Until 1988, this ratio ranged between 4 and 5. The expenditure data thus confirm the strongly curative bias of the health sector observed in the number of patients examined at primary health facili- ties for preventive versus curative care. Again, the data in Statistical Annex Table 4.11 suggest that the ratio of curative to preventive health expenditure has decreased since 1989, although the decrease is most likely the result of an increase in external assistance tied to preventive health interventions. 22/ The coefficients of variation for provincial health expenditure, total expenditure, and gross domestic product for 1989 are 233, 81 and 163, respectively. 24/ Of course, the ratio of salaries to total recurrent costs was exceptionally low in 1988. But between 1984 and 1990, it did not average more than 19-20 percent. - 107 - Tabl- 4.14: Composition and Structure of Health Expenditure, 1984-90 Variable 1984 1985 1986 1987 1988 1989 1990 As % of GDP 0.82 0.24 0.46 0.44 0.38 0.69 0.79 As % of Total Govt. Expenditure 3.2 4.0 3.5 3.2 2.0 3.3 4.4 Real Per Capita *xp. (1982 dong) 2.0 3.8 2.0 2.0 1.7 3.7 4.4 Source: Statistical Annex Tables 4.9, 4.10, and 4.11 4.105 Private Exoenditure on Health. All of the discussion so far has referred to government health expenditure. There are very few estimates of private (household) expenditure on health. Since it was only after 1989 that user fees in public health facilities were introduced, private medical practice was legalized, and medicines and drugs could be sold by the private sector, it is useful to examine private health expenditure only after 1989. There are two sources that provide some information on private health spend- ing. A survey of 3,502 mothers conducted by the MOH in seven provinces in 1990 obtained data on household expenditures on medical care and on medicines during the six months preceding the survey. Average household expenditure on medical care was reported to be 880 dong (or 1,760 dong annually), while expenditure on drugs and medicines was much higher, viz., 67,780 dong.95/ If one extrapolates these sample averages to the national population, one obtains an estimate of total private health (including health care and drugs) expenditure of 828.19 billion dong for 1990. Table 4.9 in the Statistical Annex shows that total government health expenditure was 364.80 billion dong in 1990. Thus, assuming that the sample of the MOH maternal eurvey is representative of the entire population, private health expenditure in Viet Nam may be over two times as large as government health expenditure (thus constituting over two-thirds of total health expenditure). 4.106 The other source is a household expenditure survey conducted by the General Statistical Office (OSO) in Hanoi City and in the provinces of Haiphong, Tiengiang, and Nghe Tinh. For 1989, this survey indicates that private household expenditure on health averaged slightly under 2% of total household expenditure. This budget share of private health expenditure is roughly similar to that observed in a number of low- and middle-income Asian countries.j1/ Extrapolating the per-capita health expenditure observed in this sample to the national population, a figure of 325.10 billion dong is obtained for private health expenditure in 1989. Since government health 95/ Average household size in the MOH sample was 5.34 persons. 21/ For example, the budget shares of private health expenditure were reported to be 2.5% in Bangladesh (IBA survey 1988), 1.7 percent in the Philippines (NSO survey 1988), 3.1 percent in Sri Lanka (Central Bank survey 1988), and 2.2 percent in Peru (2.2%, ECIEL survey 1985). - 108 - expenditure was 226.56 billion dong during that year, private health expendi- ture is estimated to account for 59% of total health expenditure. 4.107 As Table 4.15 below shows, while Viet Nam's ratio of private to total health expenditure is not unusually high in comparison with the share of private health expenditure in the other low- and middle-income countries of Asia, Viet Nam appears to be more liKe the Philippines, Thailand or Indonesia than China or Sri Lanka in this respect. 4.108 Another important observation is that, according to the MOH maternal survey, 97.5% of household health expenditure is on drugs and medicines, reflecting the paucity -- in many cases, the non-availability -- of medicines in the primary health facilities. Table 4.15: Private as a Percentage of Total Health Uxpenditure, Selected Asian Countries Country Year Percentage * Bangladesh 1987 60 China 1987 32 India 1987 59 Indonesia 1986 60 S. Korea 1987 66 Philippines 1987 69 Sri Lanka 1986 41 Thailand 1987 69 Viet Mam 1989/90 59-69 Source: Charles Griffin, "Health Sector Financing in Asia" Asia Regional Series 1990, P.168. The World Bank * Excludes insurance H. Health Financina 4.109 Sources of Health Financing. Table 4.16 shows the sources of financing in health and family planning since 1981. The domestic sources of financing include government budget (central, provincial and district) and user charges since 1989. The external sources include bilateral and multilateral grants. The Government mainly finances health expenditure through non-tax revenues, such as its operating surplus and transfers from state enterprises, although a small proportion of tax revenues, mainly from agricultural and municipality taxes, are also used to finance health activities. - 109 - 4.110 There are three trends that are apparent from Table 4.16. First, external grants and assistance have financed about 13% of total government health expenditure over the 1981-90 decade. However, this share hau been highly variable, ranging from a low of 5.5% to a high of 17 percent. In fact, the share of external grants in financing health expenditure have been increasing sharply since 1987, when external sources financed only 5.5% of total expenditure. In contrast, the share of external sources was as high as 17% and 16.2% in 1989 and 1990, respectively. The increase in external resources to the health sector is attributable to the re-entry of a number of bilateral donors into the country as a result of an improved economic and political climate. Second, after 1985, the burden of financing health services has increasingly fallen on provincial and district governments, thanks to the policy of fiscal decentralization and the deterioration in contral government revenues induced by a sharp drop in external grant assistance. As external donors have increased their contribution since 1989, the fiscal burden on provinces and districts has come down from about 75% of the total to around 66%. TAW 4.1 0: DWOlmboe Of Sowmi Of Kntmdng H.dth kpweitm.. 19S140 .191 1293 19 284 1985 1986 1987 9 1989 1990 bomsticbources MA 92. 28.39 83.2 76.5 92.0 94.5 87.8 86.6 89.0 CentrltGovt 19.0 21.8 26.0 22.5 24.2 15.7 16.3 13.6 19.8 18.0 Praw:rEcs "66.0 71.0 66.9 60.7 52.3 76.3 78.2 74.2 63.2 65.8 U"r Fe 3.6 5.2 Extemeltrants 15.0 7.2 7.1 16.8 23.5 8.0 5.5 12.2 17.0 16.2 DOmtfcSources 94.3 93.5 29.0 89.0 83L. Z94 CentralGovt 40.4 42.3 30.3 27.8 28.2 20.4 Prow es 53.9 51.2 66.7 61.2 55.2 59.0 amhatricts ExttmatGrants 5.7 6.5 3.0 11.0 16.6 20.6 I Sources:N0H and INCPFP Third, user fees, introduced in the health sector only since 1989, account for a small share of total health expenditure. Between 1989 and 1990, the share of user fees in total expenditure increased sharply (from 3.6 to 5.2%), but it li too early to know whether this reflects a long-term trend. At present, revenues from user charges are said to be primarily used for maintenance of equipment and medlcal supplies, although glven the fungibility of resources lt is not clear how this fact can be established. 4.111 Current Cost Recovery Ratio. We have already noted that the provision of adequate health servlces has been compromieed by the severe scarclty of fLscal resources ln Vlet Nam. Greater cost recovery through user fees may be an optlon to offset the decline in budgetary revenues. At pr esnt, only about 5% of total health expenditure (equivalent to about 6% of recurrent health expenditure) ls financed from user fees. This ratlo li not exceptlonally low ln comparLson with the cost recovery ratios ln other low- and mlddle-lncome Aslan countrLes. For example, the share of total health expenditure flnanced out of user fees was 0.6% ln Srl Lanka, 15.% ln PakLetan, - 110 - and 6.4% in the Philippines in the early 1980s (de Ferranti 1985). However, it was much higher in Indonesia (12.9%) -- a country that has committed itself to improving cost recovery in health and other social programs. 4.112 The cost recovery ratio of 5% in Viet Nam is somewhat misleading in view of the fact that since the Government health sector faces a severe shortage of medicines, drugs and medical supplies, individuals often have to obtain prescription drugs and medical supplies (such as bandages and injectables) on their own in the open market. These purchases are quite substantial in magnitude. Indeed, as was noted earlier, there is evidence to suggest that private health expenditure in Viet Nam (mainly comprising expenditure on medicines, drugs and medical supplies) far exceeds government health expenditure. If the Government health sector were dispensing such medical supplies and charging market prices for them, the cost recovery ratio would be significantly higher than the current estimate of 5%. 4.113 User Fee Structure. A system of fees for health care was introduced on May 1, 1989. The basic fee per consultation is 300 dong at district hospi- tals, 500 dong at provincial hospitals, and 1,000 dong at central hospitals. Primary health care at commune health centers continues to be free, but fees are paid at intercommunal polyclinics. This basic structure, established by the MOH, is apparently applied differently in different parts of the country. Local health workers have the discretion to waive fees for people able to produce certification from their neighborhood or village People's Committee that they are too poor to pay. The handicapped and medical personnel them- selves are also exempted from the user fees. Conversely, hospitals can (and most do) charge supplementary fees depending on the services provided and the drugs or other supplies consumed. A maternity, for example, costs 5,000 dong (the basic fee of 3,000 dong plus a supplementary fee of 2,000 dong) for the first and the second child; each additional child costs 10,000 dong (deliber- ately higher as a disincentive to having more than two children). An appen- dectomy costs 13,000 dong, which includes a 3,000 dong basic fee plus a supplement of 10,000 dong. However, these charges still represent heavy government subsidies to hospitals. For example, patients at a government hospital are typically charged 50,000 dong for a gastrectomy, which in fact costs 1.5 million dong. Thus, the extent of the subsidy for this operative procedure is about 97%. 4.114 Even commune health centers, which are supposed to offer their services free of charge, are not really free. In the 1991 MOH survey of health facility users, on average, 81.1% of all patients reported having to pay for treatment (presumably drugs and supplies) at commune health centers; they paid an average of 4,194 dong for each visit (Table 4.17). The average payment for a visit to an intercommunal polyclinic was 5,078 dong. (Of course, these fees were considerably lower than the 9,090 dong and 8,056 dong that patients reported paying for each visit to a traditional healer and a private physician, respectively.) - 111 - Table 4.17: Averg prce paid per visit to hetth proiders. 1991 ALL Rural Urban Variable Areas Areas Areas Average househotd size 5.1 5.0 5.3 Average monthly household income 213,434 199,377 240,040 Average monthly house-hold income 43,491 41,459 47,331 per capita Averace Drice Dmid per visit to:' Comunue health canter 4,194 3.966 5,054 Intercommunal polyclinic 5.078 4,784 6,292 Private doctor 8.056 4,857 9.447 Traditional healer 9.090 4,184 11,727 X of households obtafnina wafver from user fee at: Commune health center 18.9X 14.0X 33.1X Intercommunal polyclinic 15.7X 15.71 15.71 Notes: Averages co2puted onlt over individuals makfng a payment. Source: NOR Survey of Users, 1991. 4.115 ScoDe for Imrrovina Cost Recovery. The scope for increasing cost recovery in the health sector depends on three important factors. First, it depends on how much of the burden of total health expenditure is already being borne by individuals and households. In countries where the overwhelming share of health spending in shouldered by the Government and very little by households, there may be substantial scope for transferring some share of the responsibility for health expenses to the private sector. However, the available evidence suggests that in Viet Nam, unlike in other socialist economies, households are already taking on responsibility for a significant share of total health expenditure. Indeed, total private health expenditure in Viet Nam most likely exceeds government health expenditure by a significant margin. Much of the private health expenditure is on medicines and drugs, medical supplies (such as injectable., bandages, etc.), and contraceptives, although some of it L on health care itself. For instance, the 1991 MOH survey of health facility users indLcates that individuals are paying amounts that are not inconsequential (approximately, 3,403 dong per vlist) for health care from commune health centers, which are supposed to offer their services free of charge. Given the already significant extent of cost sharing by the public, it is not clear what scope there is for recovering a much larger share of health costs in the future from user fees. 4.116 Second, the financlal ability of individuals to pay higher prices for good care also determines the scope for improving cost recovery in the Government health sector. Financial ability can be inferred from how much individuals ace paying currently for prlvate health care. Here, the evidence is positive; data from the 1991 MO4 survey of health facility users shows individuals paylng nearly two timen as much for private health care as for health care from commune health centers and lntercommunal polyclinics. - 112 - However, it is likely that government health facilities will have to slgnificantly improve the quality of services they dispense before raiuing their fees to private-sector levels. 4.117 Third, the scope for improvlng cost recovery depends on the price elasticity of demand for health services. Here, the evidence is mixed. Many local health officials, hospital administrators, and health care workers in Viet Nam contend -- without providing much concrete evidence -- that the user fees that were introduced in 1989 have significantly reduced occupancy rates in district and provincial hospitals. However, the empirical analysis undertaken wlth the 1991 KOH survey data on health facility users suggests that the choice of government health facilities over private health care is not very responsive to price (with a price elasticity of -0.23) (See Annex F). Thus, an Lncrease in user fees at these facilities will result in an increase in total revenues.2.l 4.118 Protection of the Poor from User Fee Increases. Thus, there would appear to be some scope for increasing revenues from higher user fees. If user fees are increased, it is important to have a mechanism for protecting the poor from these increases. In Viet Nam, there is already such a mechanism in place; in principle, user fees are waived for people able to produce a certification of indigency from their neighborhood or village People's Commit- t-e. However, the MOH survey of users shows that there is little relationship between an individuals income and the average price (s)he had to pay for a visit to a commune health center or polyclinic (Table 4.18). Indeed, somewhat surprisingly, the proportion of patients for whom a user fee was waived is also not strongly related to patient income. For instance, while 19% of the poorest 25% of individuals (on the basis of monthly household income per capita) were waived from user fees at commune health centers, as many as 22.1% of individuals in the third income quartile were waived from user fees. of course, a smaller proportion of patients in the top income quartile (as compared with the bottom quartile) obtained waivers, but Table 4.18 shows evidence of rampant leakage of benefits to those not needing them. The leakage necessarily hurts the poor, who have to pay high prices. For example, the data in Table 4.18 suggest that, while an individual in the bottom income quartile would have to spend 4.6% of his annual income on two annual visits to a commune health center, an individual in the top income quartile would upend merely 0.9% of his income for the same level of care. An increase in user fees will thus be highly regressive unless better mechanisms to cushion the poor from these increases are put in place. The survey data clearly show that the system of exempting the indigent from user fees is not working in practice. 97/ Since the MOH data are facility-based and are not representative of the entire population of health facility users and nonusers, they relate only to the shift between public and private sources of health care; no inference can be drawn about how many people will stop seeking medical care altogether as a result of increased user fees. This information is critical to determine whether user fee increases will have an adverse impact on health outcomes. - 113 - Table 4.18: Priem paId for health care. by income quertile,' 1991 Bottom Qua- Seesd Quar- Third Quar- Top Quar- Variable rtilt tile tilte tile Average household size 5.5 5.1 5.0 4.8 Average monthly household income 86.853 139,381 204,954 443,820 Average monthly household fncome per capita 14,870 27,222 41,460 94,790 Averane pavment Per visit to:" Comiune health center 4,130 3,685 4,290 4,657 Intercomunal polyclinic 4,831 6,296 4,369 4,924 Private doctor 7,168 7,383 7,024 10,622 Traditional healer 4.601 8,963 9.129 13,206 X of individual obtainina waiver from user fee: Commune health center 19.0X 19.7X 22.1X 13.3X Intercommunal polyclinic 16.8X 17.5X 157X 11.7X Notes: * Quartiles calculated on the basis of monthly household income per capita. Thus, the bottom quartfle Includes the poorest 25K of individuals in the sample. * Average calculated only over Individuals making a payment. 1. External Assistance and Non-aovernmental Oraanizations Levels and Sources of External Support 4.119 The State Committee for Economic Relations with Foreign Countries (SCERFC) is responsible for the coordination of external assistance to Viet Nam. Specific specialized donor agencies are assigned to ministries or special committees which work closely with them in carrying out their mission. For example, UNICEF's health activities are conducted in conjunction with the Ministry of Health. 4.120 In the past, a significant share of the bilateral assistance to Viet Nam came from the Soviet Union, which was the single largest donor. Other important donors were Sweden (contributing around US$55 million annually), the German Democratic Republic (US$23 million annually), and Finland. During 1987 total development assistance (bilateral and multilateral) from noncommunist countries amounted to US$148.3. The greater part of this amount, US$60.7 mil- lion, was directed towards capital assistance, followed by technical and program assistance (US$55.5 million) and humanitarian assistance (US$32.2 mil- lion). The USSR provided a very large amount of foreign aid but no official data are available on this flow. About 63% of the total known assistance came from bilateral donors, 31% from the United Nations agencies, and 6% from non- governmental organizations (NGOs). Industry generally accounts for the single largest share of foreign assistance (26%), followed by natural resources (16%), humanitarian aid (14%), health (13%), agriculture (12%), welfare and - 114 - social programs (6%), and transport, population and science and technology (receiving 2% each). 4.121 In 1988 total external assistance to the health sector was estimated at approximately US$25 million, of which US$8 million was from UNICEF, US$5 million from UNFPA, and US$10 million from the Swedish International Develop- ment Authority (SIDA). Much of the external assistance is channelled through the Central government but some gets to the provincial level. At US$25 million, external assistance constitutes between 15 - 17.5% of the total health budget. 4.122 One source of external assistance that is generally overlooked is overseas Vietnamese, who send remittances amounting to approximately US$100 million annually. This is a substantial sum, amounting to almost two-thirds of all the other aid coming from noncommunist countries. A large part of these remittances are sent in the form of medications. Maior Donor Proiects 4.123 Most of the donor activity in the health sector was initiated in 1982, when WHO and UNICEF launched the Expanded Program of Immunization (EPI) and the program for the Control of Diarrheal Diseases (CDD). Both of these programs have received the full support and promotion of the Government, and have trickled done to the lowest levels of the health services in the country. The National Institute of Hygiene and Epidemiology (NIHE) is the branch of the MOH that is ultimately responsible for EPI. There is a strict vertical hierarchy in the implementation of the program: the NIHE in Hanoi, the Regional Institutes of Hygiene and Epidemiology in the provinces, the Hygiene and Epidemiology Brigade at the district level, and the Hygiene and Epidemiol- ogy extension worker in the communes. 4.124 The EPI has included social mobilization, including organized mass propaganda campaigns to educate the public about immunization; refresher training of health workers at all levels; and major investments (largely made by UNICEF) in the "cold chain" (viz., refrigerators, freezers, vehicles, sterilization instruments, and injectables). The results of these campaigns have been impressive; the percentage of children under one immunized with measles, DPT, and OPV vaccines increased from around 4-5% in 1984 to 40-50% in 1987 (Pond and Johnson: 82). However, the geographical coverage of EPI is limited, especially in the highland provinces, where the costs of coverage are high owing to low population densities and difficult terrain. Although on average 40-60% of the communes in the country are covered by EPI, the coverage in the highland provinces is only about 20-30%. 4.125 Another major program supported by external assistance is the Control for Diarrheal Disease (CDD), which is also a categorical program directed by the NIHE in Hanoi and funded largely by UNICEF and WHO. Like the EPI, the CDD program has been promoted vigorously with training courses for health workers and widespread informational campaigns for the public. But the bulk of the resources in the program are spent on distributing ORS packets. The CDD program, which was initiated in four provinces in 1982, now covers 80% of all children under five nationwide. However, these statistica are somewhat mLileading, since the entire population in a commune is assumed to have access to ORS once ORS packets are distributed to the commune health center. - 115 - 4.126 The national ARI program was initiated in 1983, again with the support of UNICEF and WHO. The program is directed by the National Institute of Tuberculosis and Respiratory Disease and the Communicable Disease Control Committee. The program includes the training of health workers, the supply of appropriate drugs, the standardization of case management in the lower-level health facilities, and, to a small extent, community health education. Unlike the EPI and CDD, progress in the ARI program was initially slow. In 1988 it covered only one percent of children under five years of age. Total coverage was 5t in 1989 and 10% in 1990. However, since then, coverage in the program has increased significantly. As of December 1991, the program covers 40 provinces, 250 districts, 3,395 communes and a total of about 3.3 million children (implying about 30% coverage). One reason why the ARI program has not expanded as rapidly as the other programs is that it involves imported relatively expensive drugs (viz., benzathin, procain penicillin, cotrimexazole, chloramphenicol, ampicillin and amoxycillin). It is estimated that a nationwide childhood ARI program would cost upwards of US$5.5 million annually in drugs alone. 9/ 4.127 UNICEF has been the major external donor in the water and sanitation area. The UNICEF-supported rural water supply project, initiated in 1981 in three southern provinces, was gradually extended to a total of 27 provinces by 1990. It involves construction of wells using simple low-cost Bangladesh design hand pumps that are locally made in cast iron. Over 40,000 wells are now in operation.29/ UNICEF has estimated that a total of 120,000 wells will be needed by the year 2000 to assure a supply of one well for 350-500 individuals (or 50-100 households) within a maximum distance of 500 meters. Some 400,000 wells will be required to assure better access to drinking water, viz., a well for 100-150 individuals within a distance of 250 meters. Nongovernmental Oroanizations 4.128 There are a number of NGOs, almost all foreign, operating in the health sector in Viet Nam. Among the major ones are CIDSE (International Cooperation for Development and Solidarity), the Mennonite Central Committee (MCC), Oxfam/UK, Save the Children Foundation, World Vision, and Medicine Sans Fronti6re. The MOH estimates that the total financial assistance to the health sector from external NGOs amounted to US$7.9 million, with NGOs from the United States, which has no bilateral aid program to Viet Nam, contrib- uting US$5.2 million. The situation is changing rapidly, however, as increas- ing numbers of NGOs get involved. For example, World Vision's budget for 1991 was over US$1.1 million, up from less than US$600,000 in 1990. 4.129 The case of CIDSE's involvement in the health sector deserves special attention, since, in addition to being the first NGO with a resident representative in Hanoi, CIDSE has given high priority to health issues. Based on an internal assessment of the health sector in Viet Nam, CIDSE worked with the People's Committee and health authorities in Bac Thai province to 2/ Robert S. Pond and Bekki Johnson, "Health in Viet Nam: A Report to CIDSE," 15 November 1988, (mimeo) p. 97. 22/ UNICEF, Viet Nam: The Situation of Children and Women, UNICEF, Hanoi, 1990. - 116 - prepare a diutrict level primary health care project, which was later (Novem- ber 1990) funded. Activitie, to date have lncluded qualitative research to determine community health needs and problem. of the health care dellvery system, training of health workers in primary health care (PHC) concepts and interventions, plans for upgrading selected commune and district health centers, and a baseline survey to establish key health, demographic and development indicators. Particular attention during the first year has focused on improving immunization coverage, especially in three ioolated communes that had no immunization coverage up until September 1990. 4.130 Save the Children Fund (SCF/USA) has been working in four of the poorest communes in Than Hoa Province, with a total population of about 20,000. The focus of the project has been on developing an integrated PHC and community-based, low-cost income-generating model which can be later extended to other communes. Besides having a strong element of community involvement, the project has concentrated on developing a management information system that can be used to evaluate the impact of activities. Aid Commitments for the Future 4.131 Commitments to the Ministry of Health already made by external donors for the next four years suggest that external flows will be nearly 50% higher than the $20 million that have been flowing to the health sector annually during the last 3-4 years. However, all indications are that even this level of support will be inadequate to undertake the major investments that are needed to restructure the health sector. The Government will most likely incur large financial gaps, unless its fiscal situation improves and external donors come up with substantial additional commitments. Never- theless, it is important to note that fiscal problems could be ameliorated to a significant extent by improved domestic resource mobilization, better aid- coordination, and better targeting of health expenditure to poorer provinces. - 117 - V. MAJOR ISSUES AND STRATEGIC DIRECTIONS FOR THE HEALTH SECTOR A. Vietnam's Health Sector in Historical Persoective 5.1 There is a mystery in the details on the health sector in Viet Mam presented in this review. On the one hand, all the evidence points to a health sector that is languishing. Health facilities are suffering from a severe shortage of medical supplies and equipment; the morale of health workers is down; and the utilization of primary health services is extremely low. On the other hand, Viet Nam has made impressive gains in reducing infant mortality and fertility during the last 2-3 decades. Even countries that have much higher levels of per capita income do not have infant mortality rates as low as Viet Nam's. How could a country with such poorly equipped and under- utilized primary health facilities improve health outcomes so dramatically? 5.2 One can only conjecture on what happened to the health sector in the past, since there is very little historical information on the functioning of the health system. Most likely, a number of factors jointly contributed to the dramatic health gains achieved by Viet Nam. The most important factor was, of course, political resolve and imperative. From the very beginning, the country's leaders had decided that the development of social sectors -- in particular, literacy, education, poverty, health and female emancipation -- would have the highest priority in development. In this sense, Viet Nam was not very different from other centrally-planned, socialist economies, most of which perceived health as a basic human right and as a political imperative. This resulted in the allocation of considerable commitment and resources to the health sector. 5.3 The generous allocation of resources was apparent in the establish- ment of a vast network of primary health facilities throughout the country after 1954, but particularly after 1968. The expansion greatly increased people's access to primary health facilities, except in a few provinces (mostly in the remote mountainous regions of the country). Within provinces, health centers were generally well-located, since people had a say in the location of health centers via their Commune Councils. Unfortunately, since there are no empirical surveys or studies on the functioning of typical commune health centers under the old DRV (Democratic Republic of Viet Nam) system, we do not know how well the health centers were staffed, what type of useable equipment they had, whether they had adequate supplies of medicines and drugs, and how well they were utilized by the community. However, casual empiricism and anecdotal evidence suggests that even though many of the problems afflicting the health sector today may have been present since early times, they were relatively minor during the 1960s and early 1970s, and became much more acute only during the 1980s. 5.4 In addition to the vast network of health facilities, the country invested considerable resources in developing a number of very effective categorical health programs to deal with priority health problems, such as malaria, diarrheal diseases, and immunizable diseases. Despite the fact that officially primary health care was the dominant slogan, the categorical health programs, many of which were vertically organized and not integrated into a primary health system, were central to the country's health strategy. Again, - 118 - Viet Nam was not unusual in following this strategy. other centrally-planned, socialist countries, including China, have emphasized categorical interven- tions for priority health problems despite the rhetoric on primary health care. 5.5 The network of primary health facilities and the categorical health programs cannot by themselves account for all of Viet Nam's success. In addition to these factors, Viet Nam has the advantage of a highly literate population. Of course, near universal literacy and promotion of education, especially among women, are themselves an achievement of the DRV system. Because of the high literacy rates, the receptivity of the population to health messages and health campaigns is much greater than in other low-income countries. 5.6 Finally, the importance of contextual factors cannot be discounted. During the period when the health sector achieved its most impressive gains, Viet Nam was continuously at war. It was easy to mobilize the health cadres, and indeed the entire population, in that political milieu. 5.7 By the mid to late 1970s, Viet Nam had already realized most of the achievements in health that it was to accomplish. After that time, a number of events occurred that adversely affected the health sector. The first of these was reunification. After reunification, the Government of the Socialist Republic of Viet Nam (SRV) attempted to bring the level of health services coverage that existed in the North to the South. Since the South had a significantly poor network of primary health facilities, the extension of health facilities to the South was an enormous task that placed a big burden on the health budget. Health resources had to be spread more thinly, and it is likely that nonsalary, recurrent items in the health budget, such as supplies and maintenance, were cut back during this time. The expenditure cutback may have set in motion the process of deteriorating quality of health services and decay of physical infrastructure. 5.8 Reunification also brought about with it an enormous loss of skilled health manpower in the South. Because the SRV disallowed private practice, a large number of physicians and highly trained health workers emigrated out of the country in 1975. The combination of poor health infrastructure and greatly reduced numbers of health workers worsened health conditions in the South immediately after 1975. 5.9 At the same time, the flow of external resources into Viet Nam, especially to the health sector, began dwindling. After 1979, much of the multilateral aid that Viet Nam received was cut off. While the external assistance from the Eastern bloc continued, much of it was directed to sectors other than health. The decline of Western aid and Viet Nam's political isolation coincided with a period of hyperinflation and acute macroeconomic instability in Viet Nam during the 1980s. The combination of sharply reduced multilateral assistance to the health sector and a deepening macroeconomic crisis must have adversely affected the flow of resources to the health sector. 5.10 The 1980s also marked the end of external hostilities after nearly three decades. In the absence of war, it was difficult for the Government to engender the same degree of enthusiasm and morale among health workers. - 119 - Indeed, it could be argued that the political imperative traditionally accorded the health sector began diminishing at the same time as the external hostilities were ending, the macroeconomic environment was worsening, and external assistance to the health sector was falling. With the political imperative lessened, inputs to the health sector -- drugs, equipment, medical supplies, maintenance -- declined, thereby adversely affecting the quality of care offered by the primary health facilities. This in turn changed the perception of the population about the usefulness of the health services, and lowered utilization rates. Thus, a once well-functioning health service system has been breaking down as a result of a number of factors, including the difficult transition from a centrally-planned economy to a market economy. 5.11 The experience of the Vietnamese health sector is not unique. The same story has been unfolding in many other centrally-planned, socialist economies that are in the process of a transition to market economies. Large health structures and other priority social programs that were built up with considerable resources from the State have been left languishing as a result of fiscal crises and changes in political values. B. Issues and Strategic Directions 5.12 The important issue in Viet Nam, as in other countries in the process of transition, is not what happened to the health sector but what to do in the future. Clearly, rebuilding the entire public health sector as the single provider of health care is out of the question. The Government has neither the resources nor t..v political imperative to restore the public health sector to its past eminence. Further, it is not at all clear that this would represent an efficient use of scarce resources. But, at the same time, it is important to arrest the erosion of past achievements in health that appears to have already begun. Maintaining the health indicators at their generally good levels is much less difficult than restoring them from signifi- cantly deteriorated levels. The strategy for the Government in the future should be to focus on a few key interventions that it can do well, and permit the emerging private sector to increasingly shoulder a larger share of the responsibility for curative health care. Risk Reduction at the Community Level 5.13 Water and Sanitation. The highest priority should be interventions at the level of the community that significantly reduce the risk of infection and other preventable problems among individuals. These interventions, which have significant externalities associated with them, are ideal candidates for being publicly-provided goods.100/ In this regard, the supply of safe water and sanitation is one of the most cost-effective community-based health interventions. The mortality and morbidity profile in Viet Nam is character- 1002/ This does not necessarily mean that there is a low private willingness to pay for these goods. For instance, in many developing countries, people have shown a strong willingness to pay for clean drinking water, although drinking water supply is clearly a public good. - 120 - ized heavily by diseases that are linked to water supply and sanitation, such an gastroenteritis, dysentery, typhoid, cholera, viral hepatitis, malaria, and dengue fever. Yet, only about one-half of the urban population and one-third of the rural population in Viet Nam have access to drinking water, despite mass campaigns to encourage construction and utilization of wells, rainwater tanks, latrines and bathrooms over the last 35 years. Much of the water that is available is not safe to drink. Further, the official statistics on water and sanitation show virtually no change in the proportion of housing units with safe water and sanitation during the last 15 years. 5.14 The focus should be on assisting communities with implementing simple water and sanitation technologies. The UNICEF Rural Water Supply Project and the UNICEF Sanitation Project are useful examples of low-cost water and sanitation projects that rely on substantial community involvement and maintenance. While the goal of providing each rural family with a drinking water well is unrealistic, it is important in the Vietnamese context to provide wells within a one-mile radius of most households. Since unsafe surface water is available readily everywhere, people have a strong tendency to use this contaminated water if the access to safe water is not adequate. UNICEF has estimated that 120,000 wells or taps would be needed by the year 2000 to assure a minimum of one well or tap for 350-500 people within a maximum distance of 500 metres. This goal could be achieved by 1996 if current implementation rates are maintained. However, a more appropriate level of coverage would be one well or tap for 100-150 people within a maximum distance of 250 metres. This level of coverage would require nearly 400,000 additional wells or taps to be constructed by the year 2000. External resources would be needed to meet this target, in spite of substantial community involvement in the construction of the wells. Likewise, external donors could support pilot programs that seek to provide incentives and subsidies to families to invest in appropriate sanitation technologies, such as the sulabh latrine. 5.15 Intearated-Child Nutrition. Another pressing problem is that of moderate and severe malnutrition, especially among infants, children and pregnant and lactating women. In addition its long-term effects on cognitive development, moderate malnutrition increases the risk of infection and disease. Severe malnutrition can result in death. Therefore, efforts to reduce the extent of malnutrition are essentially preventive interventions. 5.16 Several cultural and social factors are responsible for the rela- tively high levels of child malnutrition in Viet Nam. The lack of variety in the diet has several adverse effects on nutrition. At one level, the predomi- nance of rice in the diet means that children and pregnant/lactating women are unable to consume enough calories to meet their special energy needs, since the caloric density of rice is very low. At another level, since rice alone is deficient in several important nutrients, such as iron, thiamine, Vitamin A and fats (lipids), women and young children are especially susceptible to diseases associated with these deficiencies, such as anemia, beriberi, and xerophthalmia (which can ultimately lead to blindness). Infant feeding practices in Viet Nam also exacerbate malnutrition and infection among very young children. Since breastfeeding is generally initiated 3-4 days after birth -- because of an incorrect perception that coloetrum is an inferior food -- the infant begins his journey into malnutrition at birth. Viet Nam's early weaning culture results in infants being introduced to supplementary -- - 121 - usually solid -- foods as early as 2-3 months of age. The premature introduc- tion of supplemental foods greatly increases the risk of infection in small infants. 5.17 A strong case could be made for a national child nutrition program that would include growth monitoring, selective short-term supplementation, and nutrition education. To contain costs, the program would have to be tightly targeted to children (say, 6-36 months old) whose weight gain over a certain period falls below standard. These children would be singled out for special health monitoring, food supplementation, and intensive nutrition education for their families. The program could be implemented through the commune health centers In designing such a program, Viet Nam might be able to learn much from the experience of other developing countries that have experimented with such integrated nutrition programs. Cateaorical Health Proarame in Selected Reaions 5.18 At the same time, the Government should selectively rehabilitate some of the existing categorical programs built around specific and well- defined interventions.I0/ Diarrhea, malaria, and acute respiratory in- fections (ARI) are leading causes of morbidity and mortality in the country, and categorical programs addressing these diseases have been among the most successful of health interventions in Viet Nam. These programs should be continued and even upgraded. However, their management and organization need to be changed. Currently, many of the categorical programs operate indepen- dently of each other and independently of the primary health care system. Some of them rely too much on a hierarchical (vertical) top-down management system. For long-run sustaLnability, greater community support and partici- pation will be essential in all such programs. In addition, substantial economies of scale could be reaped by integrating these programs with each other and with the primary health system. The cost-effectiveness of the categorical programs could be further enhanced by targeting them to those regions having generally poor health indicators and high prevalence of specific diseases, such as malaria and acute respiratory infections. For effective targeting, however, a comprehensive geographical mapping of the country's disease profile -- something that does not yet exist -- is needed. Privatization of Health Services 5.19 By all indLcations, private health services, have grown significant- ly as a result of recent reforms, so that private expenditure for health now accounts for about two-thlrds of total health expenditure. This pattern is broadly consistent with other low income Asian countrLes. The public sector stlll possesses assets -- mainly manpower and facilities -- that were appro- priate at an earlier perlod of the country's development when the public system accounted for a much larger share of health consumption. In this rapidly evolving system, it is inevitable that the public sector health service structure will need to adapt to a more limited role and restructure n192 The nutrition program dLicussed in the previous aection can also be viewed as a categorical Lntervention. - 122 - its inputs and programs accordingly. Some shift of resources out of the public sector will be appropriate in recognition of the larger private role in curative and clinical services. This would be desirable for two reasons: first, it would significantly reduce the financial burden on public resources while freeing up the health budget to address salary shortfalls, maintenance needs, and other recurrent operational improvements; second, it would allow a more manageable public service to improve quality and compete with the growing private sector, which by all evidence is operating at a higher level of productivity and quality. 5.20 The growth of the private sector in Viet Nam can play a beneficial role in the sector through improvements in service quality, increased competi- tive forces in the sector, and meeting the demand for specialized services the public sector is unequipped to provide. There are also risks involved in this shift to private provision in such areas as cost escalation, inequitable access to health services, and the deemphasis of critical preventive public health programs. Thus, a balance must be struck between the internal effi- ciency gains that accompany private sector growth, and the equity and health outcome benefits of targeted public health services. Mixed health systems everywhere face this difficult policy challenge. 5.21 Since 1989, the Government has permitted private practice in medicine, but additional reforms are needed. More active public support and regulation of the private sector must accompany the liberalizing measures taken to date. This could include contractual arrangements with the private sector to provide essential drugs, rental of government health facilities and equipment of private providers, and the introduction of private services in existing government hospitals. The public sector role in this changing system would also involve a regulatory function, which would need to be developed rapidly to assure minimum standards in private and public sector health services. The public sector would also be a source for trained health manpower. In addition, as discussed earlier, the Government would still be involved in the provision of basic health services, targetted to women and children, vulnerable and underserved groups, the poor and directed for the most part to control of communicable diseases. This entails a strong service delivery function for the public sector with some inevitable overlaps with the private sector. Reformino the Primarv Health System 5.22 Much of the evidence presented in this chapter suggests very low levels of utilization of primary health services. Annual contact rates with the health services average between 0.3 and 0.5 per capita for the overall population, with wide variations across provinces and regions. Bed occupancy rates for hospitals also appear to be less than 50%. These low levels of utilization, esp cially in rural areas, indicate a general dissatisfaction with commune health centers among a large segment of the population. 5.23 One does not have to look too far to locate the problems of the primary health sector in Viet Nam. The 1991 MOH survey of health providers invited 1,525 health workers in 379 public and private health facilities to list their priorities for future investment. Their responses, shown below in Table 5.1, are instructive in identifying the major problems of the sector. The highest priority for the future, as reported by health workers at commune - 123 - health centers and polyclinics, was increasing wages and salaries of health personnel, followed by expanding the supply of medical drugs/suppliee and purchasing medical equipment (in that order). Expansion/renovation of the health facilities received a fairly low score, and only about 2% of the health workers at public health facilities said they would like to spend additional resources in hiring more health workers. Table 5.1: Investment priorities of workers in health facilities Percentage of health workers declaring the item of expenditure as having the highest priority Additional resourc- es to be spent on: Commune Intercomm- health cen- unal poly- Private Traditional ters clinics physicians healers Hiring more health workers 2.3 2.1 6.8 2.5 Paying higher sala- ries to workers 36.4 46.2 25.6 17.1 Buying more drugs & medical supplies 33.3 34.5 50.0 58.5 Purchasing more medical equipment 24.0 29.0 37.5 8.5 Kxpanding/renovat- ing health facillty 19.5 7.4 14.0 34.7 Number of health workers surveyed 860 542 61 62 Number of health facilities surveyed 191 77 57 54 Notest The hypothetieal question that was asked of health workers was, "If your facility were to obtain additional resources, where would you llke to spend these resources (in order of prio- rity)? " Source: MOH Survey of Health Providers, 1991. 5.24 All of these priorities have signlficant budgetary implications. For example, a pay increase for all government health workers will result in an enormous Lncrease ln government health expenditure, as will a program of rehabliltating physLcal plant and equlpment ln commune health centers. Unless other cost-offsettLng changes are lntroduced simultaneously, it is not clear that the above "wLsh-lLst' can be fulfiled. 5.25 What Ls needed to revltalize the primary health sector is a package of options, some of which would Lncrease costs and others that would offset these cost Lncreases. It li unllkely that such a package would be expendi- - 124 - ture-neutral, but the budgetary implications of this package would be rela- tively modest. Listed below are some elements of the package. 5.26 Healt. Worker Wages. There is no question that the wages of health workers in the public health facilities are abysmally low. Further, it appears that many health workers are paid their monthly salaries several months behind schedule. Since public health workers can often double or treble thoir wages in private practice, the low public-sector salaries create severe incentive problems, and engender absenteeism, low productivity, and low morale among public health workers. The problem is most acute in rural areas, where a large number of health workers also engage in farming to augment their earnings. it is clear that effective implementation of categorical health interventions and a revitalization of the primary health-care sector will require the active cooperation and support of grass-roots health workers. Wage increases and improved fringe benefits, including health insurance, for these workers should be considered a necessary -- although not sufficient -- condition of improving the quality of health services. There are reports that the Government will announce a major reform in health worker salaries in 1992- 93, under which the wages of health workers will be brought in line with salaries in other sectors. If true, this would be a major step in raising the motivation of health workers and improving efficiency in the sector. 5.27 Retrenchment of Government Primary Health Workers. An increase in the wages of government health workers would almost necessarily have to be offset by a major reduction in the number of government health workers. The important question is: will this retrenchment have adverse health impacts? For a number of reasons, the answer is no. First, Viet Nam already has one of the lowest population:physician ratios in the developing world -- almost one- third of the rate in Indonesia and one-half of the rate in Sri Lanka, Thailand and the Philippines. Second, the evidence presented in this report suggests that significant improvements in the utilization of health services and in health outcomes are unlikely to be achieved by an increase in the number of clinical physicians. Low health worker wages, lack of equipment and supplies, and the generally poor quality of health services -- not health manpower -- are the real constraints to further reductions in the infant mortality rate. Third, retrenchment of government health workers need not imply a decrease in the total number of available health workers. Indeed, as pointed out earlier, a relatively painless way in which retrenchment could be undertaken would be by allowing government health workers to move to the private sector. In many canse, the workers could continue to practice in the same commune health centers in which they currently servel the only difference would be that they would pay a rental fee for using government facilities and equipment. 528 Refurbishing and Re-eauiDping Health Facilities. It is clear that, although Viet Nam has a vast network of health centers, hospitals, and other health facilities, the physical quality of these facilities and the extent to which they are equipped is extremely poor. For instance, many distrLct hospitals have very inadequate or no kitchens, laundry units, boilers, linen, or furnlture. Medical and surgical equipment, such as weighing scales, sterilization equipment, and refrigerators, are sparse, antiquated, and barely functional in many cases. A significant proportion of commune health centers are housed in buildings that are in a state of serious disrepair, and have virtually no drugs or injections to dispense. While a major rehabilitation of the prlmary health facilities appears to be essential, it is not clear how it - 125 - can be financed in a resource-constrained environment. Again, selective privatization of government health services might be a sound alternative. In areas where the private sector can step in without much difficulty, the Government could pass along the burden of rehabilitating health facilities to the private sector. In areas where the potential for a vibrant private sector in health care is limited, the Government could undertake selective refurbish- ing of health facilities, financed in part by increased user fees. 5.29 Cost Recovery in Health Services. Even if the Government downsizes the public health sector, it is going to have to raise user fees in order to finance salary increases for government health workers and refurbishing of government health facilities. There are a number of reasons why raising user fees at health facilities is a prudent option. First, survey data show that individuals are already paying considerable amounts for treatment at commune health centers, whlch are supposed to offer their services free of charge. Patients using intercommunal polyclinics are also paying significantly more than officially-established user fees. It is not clear where these additional collections are going. Most likely, the "informal" user fees are not flowing into the Government health system. By increasing official user fees, the Government health sector could tap into this existing revenue stream. Second, the same survey data show individuals paying nearly two times as much for private health care as for health care from commune health centers and intercommunal polyclinics. Hence, at least the better-off individuals in the country have the capacity to pay significantly higher fees for public health care. Of course, it is unlikely that government health facilities could raise user fees to private-sector levels without significantly improving their quality of care. Third and finally, the limited empirical analysis undertaken here with facility-based survey data suggests that the demand for government health facilities is not responsive to price (with a price elasticity of -0.23). The fact that the estimated price elasticity is less than unity suggests that an increase in user fees will raise total revenues. 5.30 However, there are two qualifications to the proposal to increase user fees. First, user fees should be increased only for curative services for which there is typically private willingness to pay. Preventive services, typically provided through the categorical programs discussed earlier, have a strong public-goods character, and should continue to be provided free of charge by the Government. Second, there should be a mechanism for protecting the poor from user fee increases. Although there is already such a mechanism in place in Viet Nam, the system of exempting the indigent from user fees is not working in practice. Survey data show little relationship between a patient's income and the average price paid for treatment at a commune health center or polyclinic. In fact, at the present "informal" level of user fees, an individual in the bottom income quartile would have to spend an unaccept- ably high share (nearly 5%) of his annual income on merely two annual visits to a commune health center. 5.31 Risk Coveraae and Insurance Schemes. Full-cost user fees cannot be introduced for high-cost items, such as hospital inpatient care, until a large segment of the population participates in risk-sharing insurance schemes. Currently, such schemes are non-existent in Viet Nam. However, a number of developing countries have successfully experimented with health insurance. In particular, China has been successful in covering the majority of its urban population, and a non-negligible proportion of its rural population, with its - 126 - risk-coverage programs. Since 1983 Thailand hau introduced an innovative "health card" system in its rural areas, under which households can purchase an annual health card, typically priced at le than the average annual household health expenditure, that entitles the bearer to a fixed number of treatments and an unlimited number of visits for preventive services, such as maternal and child health care and immunizations. Evaluation surveys indicate that the health card system is very popular among rural residents. Many other developing countries have village-level drug revolving funds that are one form of local health insurance. 5.32 A survey of 3,502 mothers in 7 provinces of Viet Nam found consider- able receptivity to health insurance and cooperative funding of village chemists. Over 93% of the surveyed mothers indicated that they would accept health insurance coverage, on average, survey respondents were willing to spend roughly one-third of their actual monthly expenditure on drugs and health care on health insurance premiums, with very wide variations across provinces (Table 5.2). In addition, the respondents were willing to make one- time contributions of between 2,730 and 7,450 dong for setting up a chemist in their neighborhoods. Although this sample was not nationally representative, the results are indicative of substantial interest among rural residents in participating in risk-coverage schemes. Already, with the assistance of some external NGOs, some provinces have been experimenting with village-level drug- revolving funds on a pilot basis. Some government officials also assert that a national health insurance system is being discussed at the highest policy levels in the country. It would be very important for the Ministry of Health to contribute to this discussion by launching pilot schemes nationwide to explore the viability of community risk-sharing arrangements, such as health insurance and drug-revolving funds. Tabte 5.2: Wlliingnes to Pay for Heltth Insurance, Viet NM, 1990 | Lang Haf Thue Lam Khanh Long Song Total Son Hung Thien Dong Hoa An Be Household expenditure on 110 30 640 2,390 1,540 900 70 880 health care during the last 6 months (dong) 9iv,i;choLd expenditure on 10,850 12,220 32,730 71,300 49,070 24,120 20,890 33,890 medicines during the last 6 months (dong) Willingness to pay for 3,780 2,730 7,450 4,070 4,260 3,260 3,500 3,910 opening of a chemist's store in neighborhood (done) Willingness to contribute 1,120 580 710 6,350 1,450 920 4,300 2,210 monthly of a chemist's store in neighborhood (dong) Source: Ministry of Health, Center for Humn Resources, Survey of Health Proarams in 7 Provinces, 1990. 5.33 Cost Recovery in Medical Trainino. Greater reliance on the private sector for curative health care needs to be accompanied by a fundamental change in the way in which physicians obtain their medical education and - 127 - training in Viet Nam. Although physicians are expensive to train, medical training is virtually free. This policy may have been practical in a govern- ment-run health care system, but it would make little sense in a heavily- privatized health sector. While it may still be efficient for the Government to supply trained manpower to the private sector, it would be imprudent for it to subsidize the medical education and training of private-sector health workers. Tuition fees would need to be set and collected, so that public subsidies for medical training could be significantly reduced or eliminated. 5.34 Manpower Retrainina. In addition to the introduction of tuition fees in medical schools, there is urgent necessity for changing the medical curriculum. Despite the Government's emphasis on primary health care and preventive medicine, there is a strong curative bias in the medical curricu- lum. The basic approaches to modern public health management are not being taught to health personnel, with the result that the concept of primary health care is still not widespread in the medical community. The bias in the medical curriculum is reflected in the strongly curative approach to medical treatment that is adopted by most health workers at all levels. To some extent, the Manpower Development Plan of 1990-2005 does seek to address this problem by proposing to retrain and upgrade a large number of general assis- tant physicians in commune health centers to become specialists in public health. Also, some NGOs have supported training of public health workers in nearby countries, particularly Thailand and the Philippines, but the numbers are still far few to have a major impact. 5.35 Institutional Canacitv Building in the Health Sector. The PHN sector lacks an institutional capacity to monitor, manage and evaluate its programs. Indeed, a part of the difficulty facing the MOH in assessing the performance of the health sector and in determining an appropriate strategy for the improvement of health and family planning services stems from the lack of an organized management information system encompassing health-sector activities. There is no comprehensive system of supervision, monitoring or evaluation of grass-roots medical facilities in the health sector. 5.36 This lacuna is most visible at the level of the district. One could argue that it is only through major improvements in district-level management of health services that primary health care can be upgraded. The district has to have the managerial, administrative and technical capabilities to make medium-term health plans and efficiently allocate resources, including staff, equipment, and drug supplies, for their implementation; to identify projects, evaluate their feasibility, and execute them; and to monitor and evaluate national health programs operating in the district. Planning and budgeting for efficient resource use in the sector, especially at the lower administra- tive levels, should be strengthened. At present, most district-level leaders lack these capabilities. 5.37 Limited information that would be useful for monitoring, evaluation and management purposes does exist, but it is poorly organized and difficult to retrieve. For instance, each of the categorical health programs, such as the Expanded Program of Immunization, Acute Respiratory Infections Program, and the Control for Diarrheal Diseases Program, requires the commune health center to maintain a separate register of participants. As a result, a single child's name is often entered in multiple registers. Not only is this method of recording data too onerous on the health workers, it makes subsequent data - 128 - retrieval for monitoring and evaluation purposes very difficult. It would make senus to move to a simplified and integrated system of record-keeping that is individual or family based. 5.38 There is also no system of regularly collecting information on the time spent by health workers in various activities and on worker performance and productivity. Such information is essential not only for better overall management but also in formulating wide-ranging policies on attrition in and redeployment of the country's health manpower. 5.39 Further, to make the record-keeping at comune health centers useful, there should be a mechanism for the data to be transmitted to the higher levels of decision-making for evaluating the impact of alternative health interventions. Currently, no mechanism of ongoing data transmittal (other than statistics on family planning) exists, with the result that the data, even when they exist at commune health centers, cannot be used for planning purposes. In addition, the district and provincial health services themselves lack the capabilities of meaningfully analyzing field data on programs for the purposes of supervising, monitoring, and evaluating program interventions. 5.40 Finally, an important limitation of existing data on the health sector is that they are based entirely on information obtained from the public sector. Admittedly, the public sector is the dominant sector in health delivery. However, since the legalization of private practice in 1989, the number of individuals seeking treatment ln the private health sector is expanding rapidly. As this trend continues, an information system based only on the public sector will provide data that are increasingly inaccurate. Internal Efficiencv of Government Health Exoenditure 5.41 Even with external assistance and availability of additional domestic resources, simultaneous efforts for improvement in internal efficien- cy through compositional shifts in recurrent budget should be continued. For example, a disproportionately large share of government health expenditure is spent on curative relative to preventive functions. since preventive services are public goods, it would be more efficient for the Government to subsidize preventive services and either leave the provision of some curative services to the private sector or sharply increase user fees for these services. Another area in which the internal efficiency of government expenditure could be improved is in the manpower mix. Althrugh Viet Nam has a more balanced mix of nurses and physicians relative to most Asian countries, lts ratio of 3.5 nurses to each physician is comparable to ratLos observed in developed countries, where the heavlly curative-based systems of health care require relatively intense use of physician services. Since the leading causes of morbidity and mortallty in Viet Nam are praventible diseases (primarily, infectious and parasitic), which typically do not need physician-intensity, there may be scope for substantial cost saving by changing the personnel mix with more emphasis on community health workers, nurses, midwives and assistant doctors. Additlonal examples of inefficient composition of expenditure are the disproportionately small share of contrae-ptive procurement in the family planning budget and the low ratio of family plannlng administrator to service delivery personnel salaries. Both of these trends are allocatively ineffi- cient. A final example of internal inefficlency is the wide disparLty in - 129 - provincial government health expenditures. Since the provinces that spend the least on health servlce. are typically those that have highest levels of infant mortality, a strategy of redistributing government health expenditure from richer to poorer provinces will not only promote equity goals but will also bring about a larger aggregate decline in the infant mortality rate. External Resources for the Health Sector 5.42 Without enhanced foreign aid flows, the Government's intended health sector development goals probably will not be feasible. Increased domestic resource mobilization may be impaired by inadequate private savings. More generous aid is therefore needed. However, equally important is the effective utilization of external aid flows in the long run. Although considerable improvements in foreign aid utilization can be achieved through institutional strengthening, improvements in procedural aspects, including choice of core investments to be financed within capital budget, procurement, the use of technical assistance, and staffing of project entities need immediate atten- tion. Critical problems like salary, equipment, medical and contraceptive supplies, manpower training, information/education/communication (IEC) should be tackled first, both through program and project assistance by bilateral and international agencies. For instance, although the salary issue cannot be resolved except at the macro level, the provision of better equipment, improved supplies, and manpower retraining within an individual project will have a positive impact on raising the morale and productivity of health and family planning workers. Economic Growth. Income Dietribution and Poverty: Implications for PHN 5.43 An important issue is the effect of economic growth and liberaliza- tion on health and nutritional status in Viet Nam. There are several indica- tions that economic liberalization and the consequent emergence and growth of the private sector have widened income disparities in the country. At the same time as entrepreneurs and individuals working in the incipient private sector have experienced rapidly growing incomes, public-sector employees and other salaried persons have found their real purchasing power eroded, as their salaries have failed to keep up with inflation and the food subsidies that benefited them have ended. The widening disparity in income is a relatively new phenomenon for Viet Nam (particularly, the North), and has important implications for the population, health and nutrition of the poor. In other countries the combination of worsening income distribution and price decontrol (resulting in higher prices for food, health care and contraceptives) has reduced the food consumption of the poor and their utilization of health and family planning services. 5.44 Unfortunately, little information is available on the growth of income disparities, poverty, and consumption in Viet Nam. This gap will be filled to a great extent by the Living Standards Measurement Survey being undertaken jointly by the World Bank and the General Statistical Office. With the availability of these data, much more research on the interactions between income, fertility, nutrition and health will need to be undertaken. This research would be useful to the Government in finding ways of cushioning the poor and other disadvantaged groups from the sharp price increases associated with increased privatization and greater cost recovery in the government health sector. - 130 - Emeraina Environmental Problems 5.45 Another issue that wlll become increasingly Lmportant in the future is the lmpact of worsening environmental and ecological conditions on health. Crowding, pollution, stress and occupational hazards are already beginning to adversely affect the environment and the quality of life in Viet Nam. Hazardous solid wastes from industry and agriculture are usually collected jointly with other common wastes, and the most common methods for waste dieposal are open dumping an open burning, both of which produce health hazards, air pollution, and sanitary discomfort. Pesticides and other chemical insecticides are becoming increasingly common in agriculture, resulting in contamination of rain, surface and underground water and contami- nation of the food chain. In addition, increaeLng population pressure, combined wLth a housing stock that is expanding very slowly, is resulting in overcrowding and unsanitary living conditions. With an average of 3.1 persons for every room, Ho Chi Minh City already ranks as a city with one of the scantiest living spaces per capita in the world. The addition of another 30 million or so people to the Vietnamese population over the next 2-3 decades may put a severe strain on environmental health. Development of the Pharmaceutical Industry 5.46 The Government has listed the development of the domestic pharmaceu- tical industry as a high priority. According to the UNIDO classification, Viet Nam falls in the group of countries which have began to repack formulated drugs and process bulk drugs into dosage forms. A detailed examination of whether Viet Nam has a comparative advantage Ln the production of pharmaceuti- cal is beyond the scope of this study. However, Lt is clear that, for a sustaLned growth Ln domestic pharmaceutical output, the country will require a large market, a threshold level of logistics, well-qualified scientists and skilled personnel, access to cost-effective technology, existence of a chemical industry, and generous funds for R&D activities. Although labor and other material costs (e.g., energy) are extremely low, the country is severely deprived of financial resources for Lnvestment in new plant and equipment, licensing of internationally-available technologies, R&D activities, and imports of raw materials and spare parts. Indeed, owing to these problems, local factories are producing at only 40% of capacity. Unless the factors affecting capacity underutilization are resolved, it may be premature for the Government to make major investments in new pharmaceutical enterprises. 5.47 While the Government should continue to seek technical and financLal support as well as joint ventures for upgrading existing factories to facili- tate increased production, there is a need for an in-depth evaluation of the pharmaceutical sector that would address such issues as future trends in the demand for drugs induced by the changing age structure of the population, the pattern of disease vector and morbidity rates, alternative procurement of drugs and supplies, and the costs, institutional and manpower requirements of domestic pharmaceutical production. Such feasibility studies might form the bases for possible donor involvement in assisting selected pharmaceutical enterprises in expansion and upgradation. - 131 - ANNEX ANNEX A: DETERMINANTS OF INTERi^OVIZCIAL FERTILITY DIFFERENTIALS 1. In order to determine the effectiveness of policy interventions in influencing fertility outcomes, an attempt was made to study the association between provincial fertility rates (available from the 1979 and 1989 Censuses) and provincial-level health infrastructure, manpower, and public expenditures. The explanatory variables in the model relate to the broader health sector rather than to the family planning sector for three reasons: (i) the provision of family planning information and services is integrated with the provision of basic health care services in Viet Nam, (ii) the 1988 VNDHS indicated that 82% of the family planning users in the country obtained their contraceptive supplies and services from commune health centers and intercommunal polyclin- ics, and (iii) famiiy planning expenditures are included in the budget of the Ministry of Health. 2. Nearly 84% of the interprovincial and interternporal variation in fertility rates can be explained by the model (Table A.1). The empirical results indicate the importance of five variables in influencing fertility. First, access to health facilities, as measured by the average distance to a public health facility, has a significant inverse association with the total fertility rate (TFR). The lower the density of health facilities per unit of area in a province (thereby implying a farther distance on average to each facility), the higher is the TFR in that province. Further, the effect of distance to health facilities on fertility outcomes appears to have increased substantially between 1979 and 1989. While the elasticity of fertility with respect to distance was 0.09 in 1979, it increased to .26 by 1989. The average size of population served by a health facility is also positively associated with fertility, but this relationship is not statistically signifi- cant. Second, average health expenditures of the provincial government per health facility have a significant negative effect on the TFR. Since higher expenditures per facility are generally synonymous with a better supply of consumables, including contraceptives, these results stress the importance of contraceptive supplies in lowering fertility. Third, higher levels of female literacy in a province are associated with low levels of fertility. Fourth, there is a strong positive association between the TFR and the level of infant mortality. In other words, even after controlling for all the other vari- ables, provinces that have lower infant mortality also tend to have lower fertility. Since the estimated elasticity of fertility with respect to the IMR is 0.176, the fall in the IMR of about 46% between 1979 and 1989 produced an 8% ( - .176 x 46) decline in fertility. Since the fertility rate actually fell by about 25% between 1979 and 1989, this implies that the fall in infant mortality was responsible for roughly one-third of the decline in fertility. Finally, the empirical results indicate a strong income effect on fertility, such that a 10% increase in provincial per-capita income reduces fertility by about 1%. - 132 - ANNEX A Table A.ls DETERNINANTS OF TOTAL FERTILITY RATE, 44 PROVINCES OF VIET RAM, 1979 and 1989 Elasticity Parameter T- in Indeoendent Variable Estimate Ratio 1989 Intercept 1.069 1.90 Avg distance to health facility 0.087 1.98* Population per health facility 0.032 1.02 No. of health personnel per health facility 0.013 1.03 Provincial government health expenditure per facility (xlOOO) -0.032 -2.16* Female literacy (%) -0.187 -1.67 Infant mortality rate 0.176 2.16* Dummy variable for 1989 2.599 2.13* Dummv for 1989 interacted with: Avg distance to health facility 0.260 3.90* 0.347 Population per health facility 0.105 1.42 0.137 No. of health personnel per health facility -0.051 -0.75 -0.037 Provincial government health expenditure per facility (xlOOO) 0.010 0.44 -0.022 Female literacy (%) -0.294 -1.56 -0.481 Infant mortality rate -0.148 -1.10 0.029 Per-capita provincial domestic product (GDP) (xlOOO) -0.104 -2.86* -0.104 F Ratio 32.070 R Squared 0.839 Number of observations 87.000 Notes: Regression is estimated by OLS. Since all variables are in natural logs, the coefficients may be inter- preted as elasticities. GDP data are available only for 1989. * T Statistic significant at 5% level - 133 - ANNEX A 3. Most of the empirical results discussed above are intuitive and have been widely documented for other countries and historically over time. For instance, one of the most ubiquitous empirical regularities in the literature on fertility is the inverse association between fertility and indicators of women's status (such as their literacy).j/ The positive association between fertility and infant mortality has also been documented widely. Under a high mortality regime, couples attempt to bear more than the desired number of children, even if none dies, as a form of insurance against subsequent deaths. They practice family planning only when they are reasonably sure about improved survivorship ("the hoarding strategy"). High infant mortality may also breed higher fertility for biological reasons. For instance, breastfeeding delays the return of regular ovulation, so the interval between a birth and the next conception becomes shorter when a baby dies.2/ The negative effect of income on fertility is also widely observed and may reflect the desire of parents to substitute child quality (e.g., schooling, health, nutrition) for child quantity as their incomes increase. Finally, in a situation where family planning services are provided free, the distance to a family planning clinic acts as the effective price. Hence, as distance increases, the demand for family planning services is likely to decline and fertility is likely to increase. 1/ See Nancy Birdsall, "Economic Approaches to Population Growth," in Hollis Chenery and T. N. Srinivasan, eds., Handbook of Development Economics, Volume 1, North Holland: Elsevier Science Publishers, 1988. Also, see T. Paul Schultz, Economics of PoDulation, Reading, MA.: Addison-Wesley, 1981. 3] For further cross-country evidence, see Samuel Preston, "Causes and Consequences of Mortality Decline in Less Developing Countries" in Richard Easterlin, ed, Population and Economic Chanae In Develocino Countries, National Bureau of Economic Research, 1990. - 134 - ANNEX8 ANNEX 8: DETERMINANTS OF CHILD NUTRITIONAL STATUS AND NUTRIENT INTAKES 1. Data from the General Nutrition Survey have been re-analyzed to identify factors associated with nutritional status. Tables B.1 and 8.2 present results from analyses of variance, using weight and height as depen- dent variables. In each case, the dependent variable is first regressed on the more direct independent variables (i.e., variables that are likely to have a more direct effect on the dependent variable). Indirect variables are then added to this model. Changes in the coefficients of the direct variables indicate the possible ways in which the indirect variables affect the depen- dent variable. 2. Tables B.1 shows that the age and sex of a child and the existence of a bathroom in the house (a possible proxy for the level of sanitation) are positively associated with weight, i.e., as age increases by one month, the weight of a child is expected to increase by 141 gms; male children are 190 gms heavier than female children; and children with bathrooms in the house are 157 gms heavier than those without bathrooms. Participation of the child's family in the 5% land scheme is negatively associated with weight. (All of these coefficients are statistically significant at conventional levels.) The 5% land scheme aims at improving household food security by making available 5% of the commune land to individual families for cultivation. Therefore, intuitively, the negative association between this variable and child weights seems contradictory. However, further analysis shows that the distribution of St land is negatively associated with food expenditure. While 48% and 49% of households in the lower two food expenditure quartiles participated in the 5% land scheme, the participation rates for the next two quartiles were 38% and 19 percent, respectively. This implies two things: first, that the 5% land scheme is targeted to the needy, and, second, that the negative association between 5% land and child weights is a reflection of differences in socioeco- nomic status between those who participate and those who do not. These data also suggest that the gap between need and supply may not be adequately bridged by the supplementation (of food security/cash income) from the 5% land. This issue needs further investigation. 3. Addition of household calorie intake per capita and household expenditure on food to the first model increase the explanatory power of the regression model. Both household energy intake and food expenditure are positively associated with weights of sample children. 4. After controlling for the factors listed above, there were signifi- cant differences in c. .ld weights across ethnic groups and ecoregions. Children belonging to the Viet ethnic group had lower weights than other children. Children in the Central Highlands, the Central Coast of Southland, and those in the Red River Delta had lower weights as compared with children in Hanoi and Ho Chi Minh cities. Children in the Mekong River Delta were observed to have the highest weights, followed by children in the North East of Southland, the Central Coast of Southland, and the North Mountain Midlands, in that order. It is useful to note here that this ordering of ecoregions in - 135 - ANNEX B terms of weight-for-age is somewhat different from the unadjusted order presented in the text.3/ 5. When height is used as the dependent variable (Table B.2), age, sex, presence of a bathroom, ownership of a pig, and availability of running water in the house are all positively associated with heights of young children. Participation in the 5% land scheme is again negatively associated with height. Energy intakes and food expenditure do not show an association with height. The latter result is not surprising, and merely reflects the fact that height is influenced cumulatively by past food intakes. Another explanation is that energy intakes (and quite probably food expenditure as well) in a predominantly rural country like Viet Nam vary considerably during different times of the year, and that measurement of energy intakes at a single point in time is unlikely to be related to measures of long-term nutritional status. 6. Significant differences in child heights across ethnic groups and across ecoregions were observed even after controlling for the factors men- tioned above. In particular, children belonging to the Viet ethnic group were likely to have lower heigI.ts than those belonging to non-Viet ethnic groups. Children in the North-east of Northlands were likely to be the tallest, followed by the Central Highlands, North Mountain and Midlands, and the Mekong River Delta, respectively. Children in the Red River delta had the shortest stature, followed by those residing in the Central Coast of Northland and the Central Coast of Southland. 7. Determinants of Child Enerav Intakes. The same data have been used to identify factors associated with the energy intakes of children under 5 years of age. Table B.3 presen.:s results from the analysis of variance using energy intake of children as the dependent variable. Energy intakes are found to be positively associated with the age of a child (but not with gender). The availability of running water in the child's house was the most important factor associated with energy intake. This may, in fact, be a proxy for income or wealth. Food expenditure as a continuous variable was not important in determining energy intakes. Dichotomous variables for food expenditure quartiles were statistically significant, but the size of the coefficients was not biologically significant. Affiliation to the Viet ethnic group was positively associated with energy intakes. But, after controlling for food expenditure (Model 2), ethnic group was no longer significantly associated with energy intake, suggesting that the effect of ethnic group is likely to be mediated through differential food expenditure. 3/ In the data presented in the text, there are no controls for age, gender, participation in the 5% land scheme, etc. - 136 - ANNEX B Table B1: PROXIMAL VARIABLES ASSOCIATED WITH ATTAINED WEIGHTS OF 0-59 MONTHS OLD CHILDREN, 1987-89 (Results from Analyses of Variance) Dependent Variables Weights of children (in 100's of grams) Independent Variables B CoefficLent p-Value Model 1: Constant +61.017 Age (Months) +1.412 .000* Gender (Male) +1 905 .000* Participation in 5% land scheme -1.010 .000* Bathroom in the house +1.570 .000* Model 2 Constant +61.160 Age (Months) +1.3G6 .000* Gender (Male) +2.135 .000* Participation in 5% land scheme -0.948 .007* scheme Bathroom in the house +1.301 .001* Total Calorie intake +0.002 .023* Food Expenditure +0.002 .008* Notes: Viet ethnic group was negatively associated with attained weights. Compared wlth Hanol & Ho Chi Mlnis cities, weLghts were positively associated with the Mekong delta(7) followed by North-east of northlands(6), Central coast of *outhland(4) and North mountain midland(l). WeLghts were lowest (negatively associated) in the Central highlands(5) followed by the Red-river delta(2) and the Central coast of southland(4). * Significant at 5 level. Sourcet NIN, General Nutrition Survey, 1991. - 137 - ANNBX B Table B.2 PROXIMAL VARIABLES ASSOCIATED WITH ATTAINED HEIGHTS O 0-59 MONTHS OLD, 1987-89 (Results from Analyses of Variance) Dependent Varlable: Heights of children (in mm.) Independent Variables B Coefflcient P-Value Model lt Constant +610.2 Age (Months) +6.637 .000* Cender (Male) +4.302 .001* ParticipatLon in 5% land sch -4.806 .000* Source of water: Running water +11.86 .000* Container/well - 6.01 Pond/Lake/Other 0.0 Famlly owns pig(s) +1.531 .014* Family has pond for pisci-culture -1.841 .043* Bathroom ln the house +1.570 .000* Notes: Calorie intake and food expenditure were not statistically significant in determining heights of young children. Zthnic group was significantly associated with height. Viet ethnic groups were likely to have lower heights as compared with non-Viet ethnic groups (p <.006). Eco-region was significantly associated with height. As compared with urban areas, children ln the North-east of northlands(6) were likely to be the tallest, followed by the Central highlands(5), North mountain- midlands(l), and the Mekong river delta(7) respectively. Children in the Red-river delta(2) were the shortest followed by the Central coast of northland(3) and the Central coast of southland(3). * Significant at 5% level. Source: NIN, General Nutrition Survey, 1991. - 138 - ANNEX Tabl- BJ3: PROXIMAL VARIABLES ASSOCIATED WITH CALORIE INTAKE PER CHILD OF 0-59 MONTHS OLD, 1987-89 (Results from Analyses of Variance) DeRendent Variable: Calorie intake oer child Per day Independent Variables B Coefficient P-Value Model lt constant +787.4 Age (Months) +2.468 .000* Family has a veg/kitchen garden -22.965 .000* Family has pond for pisci-culture +12.662 .024* Source of water t Running water +120.58 .003* - Container/well - 62.25 : Pond/Lake/Other 0.0 Model 2: Constant +795.3 Age (Months) +2.346 .000* Family has a veg/kitchen garden -15.478 .013* Family has pond for pisci-culture + 7.498 .255 Source of water t Running water +121.81 .003* : Container/well - 59.76 : Pond/Lake/Other 0.0 Food expendlture quartllet 1 - 0.58 .049* : 2 - 25.06 3 + 6.66 :4 0.0 Notes: The Vlit ethnic group was slgnlficantly associated wlth a higher calorle Lntake. However, the size of thli difference (+20 Calories) was too small to have any biologLcal/practLcal signiflcance. After controlling for food expenditure (Model 2), ethnic group was no longer slgnlflcantly assoclated wlth Calorle lntakes, indicating that the effect of ethnlc group li likely to be mediated through dLfferential food expendLture. Analyses by sco-regLon were not possible because of small sample sizes wlthln cells. * Signiflcant at 5% level. - 139 - ANX C ANNX Cs: HEALTH PROGRAMS INCLUDED IN THE NATIONAL PLAN OF THE COUNCIL OF MINISTERS 1. After a detailed analysis of the country's main health problems and constraints, the MOH selected six national health programs which were included in the National Plan of the Council of Ministers, and further ratified by the National Assembly. Two of these health programs are directed at improving the health services network, and the remaining four are personal health interven- tions directed to target populations, namely, mothers and children and people living in malaria-affected areas. A. Consolidation of villaae. comnune. and district health services 2. This program aims at strengthening commune health centers and intercommunal polyclinics through the following actionst (a) village and commune health centers are entitled to collect fees for services provided, such as delivery of babies, administration of injections and acupuncture, etc., in order to set a welfare fund devoted to improve operations; (b) commune and village health workers will become state employees, and districts will be responsible for paying their salaries, as stated in the decision 123 of the Council of Ministers; (c) the MCH and family planning program (see below) will be the pivotal activity of the village, commune and district health services; (d) all commune health centers will have an assistant physician devoted to treatment with traditional and modern medicine; (s) selected commune health centers with a good performance will be upgraded to local polyclinics by additional qualified staff and equipment and lab facilLties (for malaria, t.b.c., leprosy, family planning, etc.); (f) ward and commune drug stores will be establshed for public sale and to ensure permanent availability of essential drugs; (g) district health centers with adequate buildings and staff will be used as models for improving general management, including effec- tive administrative guidelines, provision of two administrative assistants to the chairman of the district people's committee, and involvement of the district health director and the hospital direc- tor in the mobilizatLon of staff to meet personnel requirements and program integration in support of primary health care; and (h) development of training and refresher courses on community health for directors of vlllage, commune, diLtrict and provincial health services. - 140 - ANNEX C B. Provision of Maternal and Child Health 3. A detailed analysis of this program is in the Family Planning section of the main report. Here, a summary is presented. The Government has assigned top priority to the MCH and family planning program as a mechanism to expand and consolidate primary health care throughout the country. The Government's aim is to encourage small size families for facilitating better health and development. The program objectives are: (a) to reduce the total fertility rate from 3.7 in 1990 to 2.8 by the year 2,000. This implies a reduction of the birth rate by 0.6 per thousand annually, and to increase the contraceptive prevalence rate from 42% in 1990 to 50.5% in 1995; (b) to reduce the annual population growth rate to below 1.8% by the year 2,000; and (c) to strengthen information, education, and communication in family planning as well as in maternal and child health. This program has the widest coverage in the country through the extensive network of ambulatory health facilities (village stations, commune and district health centers, and family planning clinics). The Government has planned the construction of 300 additional family planning clinics in less well served provinces, and the strengthening of contraceptive distribution and supply. Also, the Government has attempted (with little success) integration of MCH and family planning with other health activities such as prevention of Vitamin A and iodine deficiency, control of diarrhea and acute respiratory infection, malaria control and immunizations. 4. The strategy to achieve program goals include: strengthening of program management; manpower training; improved service delivery and quality of care; strengthening of supply and distribution of contraceptives and essential family planning drugs; increased financial allocations; and greater involvement of line ministries and other government agencies. UNFPA has proposed a four-year (1992-95) support program equivalent to US$36 million to implement the Government MCH/FP in eight provinces. It is expected that this strategy would remove the main causes of program inefficiency, i.e. limited availability of contraceptive methods, and less than adequate training and equipment for patient management and counseling. C. Strenathenina of basic hospital services 5. This program aims at expanding and improving the quality of diagnostic and treatment services in hospitals, and combining traditional and modern medicine in hospital therapies. This will be achieved through: (a) promoting the use of laboratory tests such as microbiological and x-rays tests; (b) consolidating key service units such as surgery, obstetrics, pediatrics, hematology, and blood transfusion, and staffing and equipping intensive care units in all hospitals; (c) strengthening nursing cadres with qualified nurses, especially with chief nurses. - 141 - ANNEX C (d) reorganizing the hospital departments of planning, administration, and records; establishing regulations for the functioning outpa- tient and inpatient services; establishing hospital drug stores to serve hospital needs and to sell drugs to patients; (e) combining traditional and modern medicine in hospital practice; (f) coordinating resources among various health facilities to ensure better quality of outpatient and inpatient services. This implies tFs closing of hospital beds where not needed or where they are not well equipped and staffed, and transfer of staff to other facili- ties in support of primary health care; f) providing'training to staff in hospital management. Financial provisions to support these measures will be established: - hospital cost recovery through collection of certain fees - sale of drugs by hospital drug stores - organization of hospital-sponsoring associations - foreign assistance and aid - encouraging after-hours private practice and home care - promoting the opening of private maternity homes D. Malaria Control 6. Although malaria is the leading cause of morbidity and mortality in Vietnam, it has received less attention than warranted. Program objectives are vaguely defined as follows: (a) limiting outbreaks of malaria; and (b) cutting down morbidity and mortality cause by malaria. The proposed strategy involves (a) expansion of detection places through provision of 4,000 micro- scopes; (b) training of more health workers in laboratory testing; c) organiz- ing mobile teams to achieve a target of 2.5 million blood tests per year; (d) ensuring timely supply of antimalarial drugs for an estimated 7.5 million patients and insecticides to protect about 6 million people in malaria infested areas; (e) coordination with the army for the control and care of malaria cases in demobilized soldiers; and e) encouragement of imports of antimalarial drugs by the private sector. E. Expanded Proaram of Immunizations (BPI) 7. The main objectives of this program are: (a) to rise immunization coverage in children under one year of age to 80%. Priority will be given to those areas with a coverage of 50% or less; (b) shift to regular immunization programs in localities where coverage of 70% or more has been achieved; and (c) maintain the campaign option to be used in difficult to reach localities. The main elements of program strategy include (a) ensure adequate supply of vaccines through imports and local production. Local production of DPT and BCG would be perfected with UNICEF assistance; (b) ensure adequate supply of syringes and needles and rational use of the cold chain; (c) strengthening vaccine-control units at local and national levels; and (d) implement material incentives for those involved in vaccinations at the local level. - 142 - ANNZX C F. Essential Druas and Materials 8. This program aims at ensuring adequate supply of essential drugs and equipment for basic health units. Top priority will be given to the supply of drugs against malaria, diarrhea, goiter, tuberculosis, mental diseases, trachoma and leprosy, and for family planning and first aid. Specific measures will be adopted to introduce better management and cost- accounting systems in drug and medical equipment factories. The Government also will promote factory local production through ensuring timely procurement of material supplies and more managerial autonomy, and more efficient distribution outlets by cutting intermediaries, and promoting private drug stores. The Ministry of Health will be fully responsible for international procurement of drugs and medical equipment, and should request tax exemptions on imports. - 143 - ANNEX D IBLX Ds THF 1990-2005 MANPOWER DEVELOPMENT PLAN 1. A 15-year (1990-2005) Manpower Development Plan (MDP) was prepared by the KOH with external assistance from SIDA and the ODA. The rationale for this plan is that there is a need for improving the quality, instead of the quantity, of health workers, since the existing staff does not sBem to be overwhelmed by patient workloads or health activities, and major expansion of health facilities is not expected because current ones are not fully utilized. The following are main features of the MDPs (a) it assumes that the population growth rate will remain around 2% in the next 10 years, and that health policies, resources and the structure of health facilities will remain unchanged; (b) it is an indicative plan, thus requiring periodic reviews; (c) it emphasizes the improvement of quality over quantity of human resources; (d) it proposes levels of health personnel, by type and cadre, over the next 15 years; and (e) it will provide the framework for formulating health manpower policies, including training and employment programs in the health sector. 1. The two main objectives of the MDP ares first, to reduce disparities in staffing across different categories of health facilities by increasing the number of health workers, as well as strengthening their skills, at commune health centers; and, second, to increase the number of qualified health staff more rapidly in the least well-served provinces than in the better-endowed provinces, Strategies to achieve these aims include: (a) adopting policies favoring initial, in-service or upgrading training of students from less well-served provinces; (b) providing facilities for postgraduate training in centers other than Hanoi and Ho Chi Minh City; (c) reducing student intakes in Hanoi and Ho Chi Minh city and increasing intakes elsewhere; (d) making extensive revisions to the medical school curriculum that take into account regional needs; (e) limiting the hiring of new health workers in well-served provinces; and (f) providing pay differentials that favor employees in the least well-served areas. 2. Proposed changes in staffina at commune health centers: The MDP proposes the discontinuation of elementary-level health workers by upgrading them to secondary level. SpecLalization of a selected few categories will be promoted as well as simplification of cadres, which will be more flexible in terms of job descriptions. For instance, it is expected that by the year 2000 all general assistant physicians will be transformed into specialists in public health, traditional medicine, or MCH/FP; also, all secondary level nurses and midwives will be specialized in MCH/FP under the name of MCH/FP nurse-midwife (Table B 1). A career structure for nurses working in hospitals will be established. These manpower changes will be reflected in the cadres of coamune health centers, which will be structured as follows: - 144 -ANNEX D (a) a public health assistant physician, who will be the manager of the commune health center in conjunction with the commune people's committee; (b) an NCH/FP assistant physician or NCH/FP nurse-midwife who will be mainly reuponsible providing all MCH/FP services, including atten- dance of normal deliveries; (c) a traditional medicine assistant doctor who will provide curative care using both traditional and western medicine; (d) other second-level health workers may be employed according to needs and resources of the commune; (e) doctors will only be included in 2,000 selected commune health centers. Table D.sIt PROPOSED MANPOWER FOR COMMUNE HEALTH CENTERS STAIF CATEGORIES CURRENT YEAR 2000 Public Health Doctors 200 2,000 Assistant Doctors General 11,800 0 Public Health 0 8,000 MCH/FP 0 7,000 trad.med. /western 1,000 10,000 Midwives 500 0 MCH/FP Nurse-Midwives 0 3,000 TOTAL 13,500 30,000 3. Proposed staffing for other health facilities: The MDP also calls for strengthening the skills of health personnel in hospitals, specialized health Lnstitutes, administrative health agencies, and other health facilities. The specific proposals of the MDP in this area are: (a) assistant physicians will receive training to become doctors or will be redepioyed as nurses. Some 600 traditional medicine assistant physicians will continue as part of hospital cadres, and 800 NCH/FP assistant physicians and 800 assistant dentists will be assigned to polyclinic.; (b) the number of doctors will be increased by about 50%, mainly with 700 upgraded assistant physicians per year from 1991 to 1999; however the population/doctor ratio will be only slightly altered, since by year 2,000 this ratio will be 2,000 people per doctor, as compared with the present 2,708/por doctor. - 145 -ANNX D (c) the number of specialist doctors will be increased by 60% (from 4,000 to 10,000, including 1,500 pharmacists). Again the ratio of doctors/specialists will be slightly changed from 5.6 to 4.5 spe- cialists per doctor in year 2000. (d) all provincial and district managers will receive training in public health and management of health services; (e) a career development plan for hospital nurses will be established, including new roles as managers of nursing hospital services; and (f) better trained and higher level faculty will be appointed in secondary medical schools. The MDP is a step forward in the analysis and solution of manpower problems of Vietnam. However, its scope is somewhat narrow, since it addresses the issue of composition of cadres but does not address the larger problem of substan- tive improvement in manpower distribution and quality. Table D.2: PROPOSED HEALTH MANPOWER FOR OTHER HEALTH FACILITIES Staff Categories Current Year 2000 lst. Degree Specialists 4,0C0 10,000 University Level Doctors 18,500 27,500 Pharmacists 6,300 7,000 Nurses 40 500 Midwives 30 150 Lab. Technicians 30 150 Secondary Level Assistant Pysicians * general 35,000 0 * traditional medicine 1,000 500 * dentists 800 800 * MCH/FP 0 800 Nurses 14,000 43,000 Midwives 4,000 4,000 Pharmacists 7,300 8,000 Lab. Technicians 6,000 7,000 TOTAL 68,100 64,100 Sourcet Health Manpower Development Plan. Ministry of Health. Hanoi, 1990. - 146 - ANNEX E ANNEX E: DETERMINANTS OF UTILIZATION OF HEALTH SERVICES 1. An attempt is made here to "explain" interprovincial differences in bed occupancy rates for basic health centers with provincial variations in access to health facilities, health staffing, provincial per capita income, and female literacy. The regression results, shown in Table E.1, indicate that the bed occupancy rates of basic health centers are highly responsive to access to health facilities. For example, average distance to health facili- ties has a strong negative impact on bed occupancy (elasticity of -0.23), as do higher ratios of population:facility (elasticity of -0.26). On the other hand, bed occupancy rates are significantly higher (elasticity of 0.29) in provinces that have well-staffed facilities. However, when the average number of personnel per facility is separated into two components -- the number of physicians and the number of other health personnel -- the results indicate that it is the staffing of facilities by nonphysician personnel -- not physicians -- that improves bed occupancy rates. 2. The two surprising findings are that bed occupancy rates are not significantly associated with female literacy, and are inversely related to provincial per-capita income in 1989.4/ The latter result might reflect two phenomena; first, that health status improves with per-capita income, so that there is less demand for health services (and therefore lower occupancy rates) at higher income levels; and, second, individuals utilize the services of higher-level (viz., district and provincial, as opposed to commune) health facilities and private providers as their incomes increase. Table E.1: Regression analysis of bed occupancy rates for commune health centers and inter- communat polyclinics 44 Viet Nam Provinces, 1979 and 1989 Independent Paraueter T- Parameter T- Variable Estfmate Ratio Estimate Ratio Intercept 4.760 3.68* 4.285 3.42* Average distance to facility Ckms.) -0.232 -2.21* -0.285 -2.80* Average population per facilfty -0.262 -3.55* -0.204 -2.43* Average no. of all health personnel per facility 0.293 7.04* Average no. of physicians per facility -0.170 -2.09* Average no. of other health personnel per facility 0.263 2.34* Female literacy aX) -0.317 -1.05 -0.235 -0.80 Dumny variable for 1989 7.620 4.55* 6.670 4.09* Per-capita provincial domestic productLa X dummy variable for 1989 -0.446 -4.64* -0.387 -4.13* F-Ratfo 14.990 8.3 R-Square 0.529 0.4267 No. of obs. 87 87 Nean of dependent variable 37.153 37. 153 Notes: La Since data on per-capita provfncial domestic product are available only for 1989W this varfable Is Interacted iuith a dummy vareable for 1989. Data for 1979 and 1989 are pooled. Equatfon estimated by ordinary least squares. All variables are expressed in natural logarithms. The coefficients can therefore be fnterpreted as elasticitfes. * T statistic significant at 5% level. j/ information on provincial per-capita income is available only for 1989. - 147 - ANBX E 3. Finally, the results also suggest that, after controlling for other factors, there has been a significant increase in utilization rates. Thus, the observed decline in bed occupancy rates between 1979 and 1989 can be attributed to the reduced availability of health facilities per capita and per unit of area and to the reduction in number of health personnel per facility between 1979 and 1989. In the absence of these changes, bed occupancy rates would most likely have increased. 4. Thus, the findings suggest that supply factors -- viz., the availability and proximity of health facilities to the population and adequate staffing of the facilities (but not necessarily by full physicians) -- contribute significantly to higher utilization of health services by the population. - 148 - ANEX P A =IX Fs ESTIMATION OF CHOICE OF PROVIDER WITH DATA FROM THE SURVEY OF HEALTH FACILITY USERS AND PROVIDERS 1. Visits to various providers were aggregated into two type.s public facilities (which included commune health centers and intercommunal polyclin- ics) and private health providers (which included private physicians and traditional healers). Since indLviduals reported price paid per visit only for the provider they selected, the first task was to predict prices paid for public and private providers for each individual in the sample. This was done by regressing price paid per visit on (a) individual attributes, such as age, occupation, educational level,S/ and whether pregnant,6/ (b) household characteristics, such as household income and household size, and (c) communi- ty characteristics, which included a complete set of 190 dichotomous (dummy) variables representing all the districts in the sample. The estimates of the price equation used for predicting provider prices are shown in Table F1.. 2. Since the price paid for health care reflects the quality of care obtained, it is important to purge quality variationo from observed prices (or "unit values") before using them in the health care demand equations.7/ It is assumed here that the variation in unit values across districts (captured by the district-level dummy variables) is the only source of true variation in the price of medical care; all individual and household variation in unit values simply reflects quality variations. Therefore, public and private health care prices were predicted for each household using only the estimated coefficients on the district-level dummy variables. In other words, in predicting prices, all individual and household characteristics shown in Table P.1 were held constant at the sample mean values. 3. Once prices for private and public health care were estimated for each individual in the sample, the second-stage estimation was straightforward. The choice of a provider was defined as a dichotomous variable that assumed a value of one if a public health provider was selected and zero otherwise. As is appropriate for discrete variables, the equation was estimated by the maximum likelihood legit method (Table F.2). To allow for the possibility that response to prices and income may vary with income, interaction terms between prices and household income and a quadratic term in income were included. The price-income interactions were not significant statistically, indicating that the price response of provider choice is invariant with respect to income. However, the income squared term was highly significant, implying that the effect of income on the choice of public providers is nonlinear. 4. There are two limitations of this approach. First, since the survey sampled only individuals utilizing health services (and excluded nonusers), it cannot be used to analyze the factors determining utilization. 5/ For children under 15 years of age, the occupation and educational level of their father was substituted. j/ This variable assumed a value of zero for all males and nonpregnant women. 7/ See Angus Deaton, "Quality, Quantity and Spatial Variation of Price," American Economic Review 78(3), 1988, pp. 418-430. - 149 - ANNEX F It can only point to variable. that influence choice of public or private health provider. Second, there in probably some bias in the results due to the fact that the sample was a facility-based sample -- not a random, repre- sentativo sample of the entire population of the three provinces. The results should, therefore, be treated an illustrative. 5. The results of the empirical application are informative. The characteristics of an individual, much as his or her age, educational level, occupation, and household income, are observed to be significantly associated with the choice of a public or private health provider. For example, both education and household income reduce the probability that an individual will choose a public over a private health provider. The income elasticity of demand for public (as opposed to private) providers is actually negative, indicating that public health service. are viewed as inferior by the public, so that as incomes increase individuals choose private over public health facilities. The finding that the probability of choosing a public health facility is greater for young children and pregnant women reflects the fact that government health facilities are still the providers of choice for (preventive) maternal and child health care. 6. The price elasticity of demand for public health services is estimated to be -0.23. This indicates a sizeable response to increased prices. However, it is important to remember that this number only reflects the switch from public to private health care with an increase in the price of public health care. Since the sample included only individuals who actually used health services, the estimated response does not capture the effect of increased health care prices on individuals who stopped using any health service altogether. 7. Another interesting empirical finding is that, even after controlling for differences in income and education, individuals in occupations such as farming (both cooperative and private) and teaching tend to use public health services more than private services. - 150 - ANNEX F Table F.1: REGRESSION ANALYSIS OF PRICES PAID BY INDIVIDUALS FOR HEALTH-CARE VISITS, MOH SURVEY OF USERS, 1991 Public Health Provider Private Health Provider Independent Variable Estimate Ratio Estimate Ratio Intercept 3449.756 3.06* 2,169.434 0.91 Household size 59.656 0.81 -189.034 -0.98 Total household Income (x 103) 0.511 0.76 8.386 4.00* Whether completed primary school 204.620 0.51 -594.354 -0.48 Whether completed secondary school-677.831 -1.25 -2,021.458 -1.39 Whether completed college 121.407 0.09 2,041.934 0.76 Whether cooperative farmer 833.735 1.43 -1,761.604 -0.83 Whether prIvate farmer -306.830 -0.57 -202.215 -0.15 Whether factory worker 1742.399 2.16* -2,637.957 -1.49 Whether teacher -150.369 -0.16 -631.135 -0.25 Whether professional 2364.193 1.97 -2,7853682 Whether artisan -504.732 -0.49 -9,362.074 -4.00* Whether construction worker 742.158 D.42 -1,508.123 -0.39 Whether housewife -450.598 -0.49 -783.452 -0.44 Whether student -101.523 -0.13 -1,003.945 -0.54 Whether driver -409.000 -0.22 10,189.010 3.08* Whether unweployed -735.979 -0.78 -2,475.669 -1.27 Whether male 728.256 2.21* -123.498 -0.13 Age (years) 6.952 0.29 227.368 3.08* Age squared 0.129 0.38 -2.640 -2.74* Whether pregnant womn -123.618&0.23 6,336.265 2.64* Whether nursing mother 2035.442.68* 1,713.326 0.75 F-RatIo 2.170 7.780 -Squared 0.081 0.236 Number of observations 5,032 1,856 Notes: A complete set of 190 district dummy variables were also included. Their coefficients are not reported here for brevIty. Equations were estimated by ordinary least square. Dependent variable Is price paid for a single visit in dong. e T statistic significant at 5X level. - 151 - hUIX F T ble F.2: Maxima likelihood logIt estimates of the probability of choosing a publfc health provider (viz., commune health canter or polyclinic) over a prIvate health provider (viz., prIvato physician or traditional realer), MON Survey of Users, 1991 Mean of Elasticity Parameter Asynptotic dependent at sample Independent Variable Estimate T Ratio variable mean Intercept 2.110 14.84* Total household Income (x 106) -1.245 -3.46* 203.400 -0.219 Incom squared (x 1012) 0.409 2.46* Price of public provider Cx 106) -60.951 -4.67* 3 726 -0.227 Price of private provider (x 106) -13.912 -2.29* 6,261 -0.087 Income X Price of public provider (x 1012) 51.578 0.93 Income X PrIce of private provider (x 1012) -5.933 -0.26 Household size -0.023 -1.75 5.107 -0.119 Age (years) -0.036 -7.97* 29.719 -0.549 Age squared (x 106) 300.589 4.86* 0.001 Whether pregnant woman 1.485 9.96* 0.082 Whether rwrsing mother 0.226 1.51 0.042 Whether cotpleted primary school -0.325 -4.14* 0.494 Whether completed secondary school -0.718 -7.45* 0.203 Whether corpleted college -1.268 -6.49* 0.020 Whether cooperative farmer 1.199 12.10* 0.222 Whether private farmer 0.706 8.35* 0.297 Whether factory worker -0.099 -0.81 0.062 Whether teacher 0.778 4.59* 0.034 Whether professIonal 0.089 0.51 0.024 Whether artisan 0.144 0.87 0.029 Whether construction worker -0.100 -0.38 0.010 Whether housewife -0.284 -2.09* 0.043 Whether student 0.028 0.22 0.061 Whether driver -0.471 -1.82 0.009 Whether unemployed -0.237 -1.69 0.041 Whether mvle 0.163 2.65* 0.408 Log-Lfkelihood ratio -3813 Mean of dependent variable 0.731 T statistic significant at 5X level. - 152 - ANNEX e ANNLX Os DETERMINANTS OF INTERPROVINCIAL VARIATIONS IN INFANT MORTALITY 1. Census data for 1979 and 1989 on infant mortality rates for each of Vi-t Nam's provinces are used here to study the correlates of health status. The empirieal results indicate the importance of three variables in lowering infant mortality (Table G.1). First, access to health faclilties, as measured by the average distance to a public health facility, has a eignificant association with infant mortality rate. The lower the density of health facilities per unit of area in a province (thereby implying a farther distance on average to each facliLty), the higher is the infant mortality rate in that province. The average size of population served by a health facility is also positively associated with infant mortality, but this relatLonship is barely significant statistically. Second, average hpalth expenditure per health facility has a significant negative effect on the infant mortalLty rate. Sineo higher expenditure per facility is generally synonymous wLth a better supply of consumables, these results reLnforce the importance of drugs, medicines and useable equipment in Lmproving health outcomes. Finally, female literacy has a very strong ameliorating effect on Lnfant mortality, but only ln 1989. The dissimilar effects of female literacy on infant mortality rate ln 1979 and in 1989 may reflect the fact that the relevance of female literacy for lowering infant mortality rate is greater during periods of deteriorating health infrastructure and declining qualLty of services (as Ln 1989) than in other periods. 2. The variables that are not *ignificant Ln explaining inter- provincial variations in Lnfant mortality rate are the average number of health personnel per facLiLty and per capita provincLal domestic product (proxying for per capita Lncomes). The latter result may be surprising, but merely indicates that higher income provines have lower infant mortality rates because (i) they have higher rates of female llteracy, (ii) they provide better access to health facilities for much of the population, and/or (iLi) they allocate more resources on average to each health facility. The lack of slgnificane- of health staffing in influencing infant mortality is also an important flnding, sinee it suggests that resources devoted to increasing the number of health personnel are unlikely to result ln higher rates of infant survival. - 153 - AMNiX a Table G.ls Determinants of Infant Mortality Rate, 44 Provinces of Viet Mam, 1979 and 1989 Elasticity Parameter T- in Independent Variable Estimate Ratio 1989 Intercept 3.719 4.49* Avg distance to health facility 0.199 2.73* Population per health facility 0.081 1.49 No. of health personnel per health facility -0.020 -0.89 Provincial government health expenditure per facility (x1OOO) -0.051 -2.00* Female literacy (%) 0.012 0.06 Dummy variable for 1989 3.675 2.39* Dummy for 1989 interacted witht Avg distance to health facility 0.019 0.17 0.218 Population per health facility -0.095 -0.72 -0.015 Mo. of health personnel per health facility 0.116 0.97 0.096 Provincial government health expenditure per facility (x1OOO) -0.001 -0.02 -0.052 Female literacy (%) -0.603 -1.89 -0.591 Per-capita provincial dometlc product (GDP) (xlOOO) -0.041 -0.64 -0.041 F RatLo 32.070 R Squared 0.839 Number of observations 87.000 Notess RegressLon Ls estimated by OLS. Since all varlables are in natural logs, the coefficients may be Lnter- preted as elastLeities. GDP data are available only for 1989. * T statistic *lgniflcant at 5% level - 154 - STATISTICAL ANNEX STATISTICAL A X Tgble 2.1: AGE-SEX DISTRIBUTION, 1979 AND 1989 CENSUSES 1979 19894 Sex RatIo Age Group Nale Femate Total aote Femate Total 1979 1989 Total 100.00 100.00 100.00 100.00 100.00 100.00 94.2 94.7 0-4 15.43 14.02 14.71 14.83 13.20 13.99 104.8 106.5 5-9 15.36 14.00 14.66 14.06 12.63 13.32 104.4 105.4 10-14 14.20 12.68 13.42 12.37 11.04 11.69 106.6 106.2 15-19 11.55 11.39 11.47 10.78 10.41 10.59 96.5 98.1 20-24 8.92 9.68 9.31 9.19 9.43 9.32 87.7 92.3 25-29 6.81 7.35 7.09 8.60 8.98 8.80 88.2 90.7 30-34 4.60 4.89 4.75 7.23 7.47 7.35 89.6 91.7 35-39 3.78 4.11 3.95 4.95 5.36 5.16 87.5 87.4 40-44 3.59 4.04 3.82 3.32 3.62 3.47 84.7 86.9 45-49 3.89 4.15 4.02 2.81 3.27 3.05 89.3 81.4 50-54 3.23 3.36 3.29 2.76 3.26 3.02 91.5 80.4 55-59 2.66 2.50 2.58 2.94 3.16 3.05 101.3 88.2 60-64 2.11 2.47 2.30 2.28 2.60 2.45 81.5 83.0 65+ 3.87 5.34 4.62 3.88 5.57 4.75 69.1 66.1 Total 100.00 100.00 100.00 100.00 100.00 100.00 95.2 94.7 Medfan Age 17.20 19.30 19.10 21.40 Source: Population Censu 1979 and 1989. Tble 2.2: FMNILY PLANNING EXPENDITURE BY FUNCTIONAL CATEGORY (MILLION DONG), 1984-90 1985 1986 1987 1988 1989 1990 Capital Examnditure Physical Faciltftes (btdgs, vehicles, furn) 5.8 19.1 32.7 235.7 227.2 496.5 Medical Equlpsnt 62.1 217.1 386.2 2050.6 7574.1 9548.9 Staff training 2.8 13.0 27.8 141.4 250.0 850.0 Returrant Examiditure Salartis (a) Adofnistrators 6.9 24.5 62.4 188.6 1100.0 1250.0 (b) Servfce delivery personnwl 33.1 117.4 272.8 1131.4 2750.0 3000.0 Contraceptive Suppies 17.5 62.1 107.6 707.1 950.0 1000.0 Dawnd generatlon (IEC) 6.3 22.5 39.4 117.9 134.0 1080.0 Re serch 3.6 12.7 31.7 141.4 15.5 79.0 TOTAL 138 489 961 4,714 13,001 17,304 Mote: MCH Is fncluded under FP Sources zNON, NCPFP *nd Staff Estimates - 155 - STATISTICAL ANNX TIbtl 2.3: FAMILY PLANNING EXPENDITURE BY FUNCTIONAL CATEGORY (MILLION DONUG 1982 PRICES), 1984-90 1985 1986 1987 1988 1989 1990 Cacftal Exoendlture Physical Facilities (btdgs, vehicles, furn) 1.0 0.6 0.2 0.4 0.2 0.4 Nedical Equipment 10.6 6.3 2.8 3.6 8.1 7.4 Staff training 0.5 0.4 0.2 0.3 0.3 0.7 Recurrent Exoendi ture Salaries (a) Administrators 1.2 0.7 0.5 0.3 1.2 1.0 (b) Service delivery personnel 5.6 3.4 2.0 2.0 2.9 2.3 Contraceptive Supplies 3.0 1.8 0.8 1.3 1.0 0.8 Demand generation (IEC) 1.1 0.7 0.3 0.2 0.1 0.8 Research 0.6 0.4 0.2 0.3 0.0 0.1 TOTAL 21.8 13.1 6.4 7.9 13.7 12.5 Note: MCH is included under FP Sources: MOR, NCPFP and Staff Estimates Table 2.4: Family Pltnnina Exoenditure By Functfonal Category (X), 1984-90 1985 1986 1987 1988 1989 1990 CaoItal Exoenditure Physical Facilitles (bldgs, vehIcles, fum) 4.2 3.9 3.4 5.0 1.7 2.9 Medical Equipment 45.0 44.4 40.2 43.5 58.3 55.2 Staff training 2.0 2.7 2.9 3.0 1.9 4.9 Recurrent Exoendi ture Salaries (a) Administrators 5.0 5.0 6.5 4.0 8.5 7.2 (b) Service delivery personnel 24.0 24.0 28.4 24.0 21.2 17.3 Contraceptive Supplies 12.7 12.7 11.2 15.0 7.3 5.8 Demand generation (IEC) 4.6 4.6 4.1 2.5 1.0 6.2 Research 2.6 2.6 3.3 3.0 0.1 0.5 TOTAL 100.0 100.0 100.0 100.0 100.0 100.0 Note: NCH Is included under FP Sources: MOR, NCPFP and Staff EstImtes - 156 - STATXSTICAL ANNEX Tabl- 2.S: POPULATION AGED 1S YEARS AND ABOVE BY MARITAL STATUS, 1989 ----Males----- ----Females---- ----Total---- No. No. No. Marital Status (000) (000) (000) Single 7465 37.4 6982 31.3 14447 34.2 Married 11899 59.7 12495 56.0 24394 57.7 Widowed 402 2.0 2417 10.8 2819 6.7 Divorced 51 0.3 178 0.8 229 0.5 Separated 70 0.4 201 0.9 271 0.6 Not Stated 59 0.3 52 0.2 111 0.3 Total 19946 100.0 22325 100.0 42271 100.0 Source: PopulatLon Census 1989 Table 2.6: PROPORTION OF POPULATION SINGLE AND EVER MARRIED 1988 and 1989 ------…Malo----- -----FemaleF---- ----eFmales a/--- 3ver- Ever- Ever- Age Group 8ingle Married Single MarrLed Single Married 15-19 95.7 4.3 89.1 10.9 95.8 4.2 20-24 58.4 41.6 42.4 57.6 49.2 50.8 25-29 21.7 78.3 17.7 82.3 20.4 79.6 30-34 6.8 93.2 11.0 89.0 13.1 86.9 35-39 3.2 96.8 8.8 91.2 17.1 82.9 40-44 2.0 98.0 6.0 94.0 18.9 81.1 45-49 1.5 98.5 3.5 96.5 18.0 82.0 EXAXM EYears b/ Mals Female Urban 26.7 24.8 Rural 23.3 22.7 Vietnm 24.5 23.2 */ 1988 VNDH Survey b/ uingulate Mean Age At Marriage Sources Populatlon Census 1989 - 157 - STATI8SICAL ANNiX Table 2.7s Age-SpecifLc Fertility Rates Women's Age Group 1979 a/ 1983-87 1986-87 1988 1989 15-19 0.021 0.007 0.016 0.020 0.026 20-24 0.222 0.151 0.199 0.235 0.192 25-29 0.312 0.264 0.254 0.243 0.221 30-34 0.285 0.237 0.186 0.151 0.167 35-39 0.187 0.160 0.105 0.085 0.110 40-44 0.107 0.084 0.051 0.0S1 0.057 45-49 0.097 0.058 0.019 0.011 0.019 TVR 15-49 5.0 4.80 4.15 3.98 3.98 a/ Staff Bstimates with Indirect Technique Sourcest VNDH Survey 1988 Population Census 1989 Tabl- 2.8t Nean Number of Children Ever Born Among All Women, By Age --------------------------------------------------------------- 1988 1989 Women's Total Rural Urban Total Rural Urban Age Group --------------------------------------------------------------- 15-19 0.0 0.1 0.1 0.1 0.1 0.05 20-24 0.6 0.8 0.5 0.7 0.7 0.4 25-29 1.9 2.2 1.5 1.7 1.9 1.2 30-34 3.0 3.1 2.2 2.8 3.0 2.0 35-39 3.9 4.0 2.9 3.6 3.9 2.7 40-44 4.7 4.8 3.7 4.4 4.6 3.5 45-49 5.8 5.2 4.7 4.9 5.1 4.3 All Ages 2.13 5.59 3.95 2.97 …-------------------------------------------------------------- Sourcess Population Census 1989 VNDH Survey 1988 - 158 - STATISTICAL ANNEX Table 2.9: Mean Preferred Nurber of Children For Ever MarrIed Women, 1988 Current Age of Ever-Married Women ............................................................ Socdo- All Women Economic 15-19 20-24 25-29 30-34 35-39 40-4 45-49 Ages Married Less Background Than 5 Years Residency Urban 2.78 2.38 2.47 2.75 3.08 3.31 3.49 2.84 2.34 Rural 2.45 2.76 2.99 3.46 3.74 3.95 4.12 3.37 2.70 Region North 2.21 2.53 2.72 3.15 3.35 3.57 3.81 3.06 2.43 South 2.79 2.97 3.18 3.47 3.90 4.01 4.27 3.53 2.83 Education Illiterate 2.00 3.20 3.61 4.16 4.38 4.54 4.58 4.16 3.09 Read/Write 2.98 2.99 3.45 3.80 4.08 4.05 4.26 3.86 2.94 Primry 2.51 2.74 2.91 3.28 3.56 3.77 3.77 3.18 2.64 Secondary & More 2.00 2.27 2.44 2.69 2.U 3.02 2.81 2.61 2.32 Occipation Agriculture 2.45 2.77 3.03 3.50 3.76 4.04 4.20 3.39 2.70 Productive Sector 3.56 2.35 2.56 2.95 3.35 3.16 3.55 2.95 2.40 Non-Productive 2.38 2.51 2.63 2.88 3.31 3.59 3.53 3.00 2.47 Sector Total 2.62 Source: VIDN Survey 1988 - 159 - STATISTICAL ANNEX Table 4.1: Infant Mortality Rate, by Province, 1979 and 1989 Infant Mortallty X change rate In IN Province 1979 1989 1979-59 Hountafn and Mid-land 62.6 45.4 -27.5X Ha Tuyen 88.0 52.8 -40.0X Cao Bang 85.0 61.6 -27.5X Lang Son 78.0 56.5 -27.6X Lai Chau 72.0 66.1 -8.2X Hoang Lien Son 67.0 56.3 -16.0X Bac Thai 57.0 45.0 -21.1X Son La 74.6 54.5 -26.9X Quang Minh. 56.1 35.0 -37.6X Vinh Phu 52.3 33.1 -36.7X Ha Bac 45.6 36.3 -20.4X Red River Delte 67.7 37.0 -45.3X Ha Noi 62.0 40.0 -35.5X Hai Phong 52.0 26.0 -50.0X Ha Son Binh 84.0 47.9 -43.0X Hal Hung 71.0 38.0 -46.5X Thai Sinh 66.1 31.6 -52.2X Ha N_ Minh 69.5 34.9 -49.8X C-ntral Coast Northland 83.0 46.5 -44.0X Thanh Hoa 68.8 36.2 -47.4X Nghe Tinh 90.2 53.3 -40.9X Quang Binh 90.1 49.9 -44.6X Qyang Trf 90.1 49.0 -45.6X Thua Thifn Hue 90.1 50.0 -44.5X Central Coast Southland 90.9 47.5 -47.7X tuan Mm Da-Nang 87.2 47.1 -46.0K Quang Ngai 95.5 51.5 -46.1X 8inh Dlnh 95.5 51.5 -46.1X Phu Yen 88.3 44.8 -49.3X Khanh Hoa 88.3 44.8 -49.3X Thuan Hai 90.4 43.9 -51.4X Central Hishiand 103.2 56.4 -45.4X Cia Lal-Kon Tum 115.1 78.5 -31.8% Dfi Lai 97.6 44.9 -54.0X Lam Dong 92.4 43.7 -52.7X North-East Southltos 72.8 33.9 -53.5X TP Ho Chf ifnh 66.0 30.1 -54.4X Song Be 82.0 45.7 -44.3X Tay Nfnh 75.2 39.1 -48.0O Dong Nai 84.5 33.8 -60.OX Vung Tau-Con Dao 71.1 32.4 -54.4X hekorm River Delta 90.0 44.1 -51.0X Long An 93.4 42.3 -54.7X Dong Thap 97.0 48.9 -49.6K An Olang 98.0 50.5 -48.4X Tien Clang 9f.4 30.0 -68.6X Ban Tra 88.0 41.1 -53.3X Cuu Long 81.0 40.5 -50.0X Hau Ciang 83.1 45.4 -45.4X Kien Slang 90.0 53.8 -40.2X MINI Hal 91.0 44.4 -51.2X - 160 - STATISTICAL ANNEX Table 42.2 SEVERITY OF MALARIA, 1985-89 Confirmed Cases Year Falciprum Vivax Severe Cases Deaths 1985 47,317 31,323 4,501 996 1986 40,476 47,261 4,461 954 1987 70,421 60,704 4,561 1,070 1988 97,723 54,174 10,470 2,465 1989 106,371 37,114 13,709 3,439 1989/85 Ratio 2.25 3.05 3.45 Sources Molineaux - Report of Duty Travel to Vietnam, November 1990, WHO STATITIC5JANE Tabt 4.3: PO 1TION CVERAGE OF AM AVERAGE DISTANCE TO HOSPITALS AND POLYCLINICS, BY PROVINCE, 1979 Am 1989 Population per Average distance Population per Average distance hosoftal to hosnital (kms.) Doltvlinic to molyclinic tkms.) 1979 1969 1979 1989 1979 1969 1979 1989 Mountain and Nid-ltn Na Tuysn 43,000 51,300 15.56 14.76 154,800 128,250 29.52 23.33 Coo oang 36,231 40,357 14.38 13.86 29,438 47,083 12.96 14.97 Lug Son 39,833 U,3643 14.69 13.60 68,286 30.550 19.24 11.38 Laf Chou 39,500 54,750 25.61 25.61 24,308 31,286 20.09 19.36 Nong Lien Son 40,579 54,316 15.72 15.72 154,200 41,280 30.64 13.70 Sac That 62,231 73,571 12.61 12.15 134,833 103,000 18.56 14.38 Son La 40,167 56,833 19.42 19.42 482,000 682,000 67.27 67.27 lurng lnh 41,313 47,824 10.87 10.55 73,444 101,625 14.50 15.38 Vinh Phu 65,524 82,136 8.33 8.13 275,200 86,048 17.06 8.33 Na BSc 78,100 103,200 8.57 8.57 781,000 114,667 27.09 9.03 Red River Delta Ha Rol 129,263 160,842 5.99 5.99 163,733 101,867 6.74 4.77 Hat Phong 67,647 85,118 5.30 5.30 143,750 160,778 7.73 7.29 Ha Son Binh 62,000 83,591 8.95 9.15 142,600 102,167 13.57 10.12 H1 Hunw 69,964 87,321 5.39 5.39 653,000 97,800 16.46 5.70 Thai Binh 76,778 90,667 5.24 5.24 230,333 102,000 9.08 5.56 Ha m llinh 103,800 112,750 6.95 6.57 144,167 75,167 8.20 5.37 Central Coast Northland Thanh Hoa 86,963 93,531 11.47 10.54 782,667 136,045 34.41 12.71 Ughe Tinh 84,412 105,382 14.52 14.52 2,870,000 99,528 84.68 14.11 Qung Binh 40,769 107,833 13.81 20.33 15,143 40,438 8.42 12.45 Qyan Tri 33,308 57,375 10.92 13.92 12,371 65,571 6.65 14.88 Thua Thien Hue 61,231 111,375 11.01 14.03 22,743 52,412 6.71 9.63 Central Coast Southland Quen NM Da-Nang 76,737 75,565 14.18 12.88 76,737 48,278 14.18 10.30 Queng Igai 47,368 94,727 9.90 13.01 69,231 148,857 11.97 16.31 Binh Dinh 57,105 83,000 10.07 11.33 88,929 11.73 Phu Yen 47,000 71,333 12.11 13.39 47,000 64,200 12.11 12.70 Khanh Hoa 90,889 12.90 81,800 12.24 Thuun Hal 76,417 90,000 17.44 16.76 76,417 90,000 17.44 16.76 STATISTI Q ANNEX Table 4.3 (continued) Population per Averge distance Population per Average distance hosoital to hosDital (kas.) aolvclinic to golvolinic Cbs.) 1979 1989 1979 1989 1979 1989 1979 1989 Central Hiohland Gia Lea-Kon Tun 48,750 58,400 26.09 23.34 585,000 438,000 90.39 63.92 Doi Lai 53,556 51,368 26.51 18.25 48,200 139,429 25.15 30.06 Lm Dori 55,429 63,900 21.48 17.97 129,333 319,500 32.81 40.18 North-East Southland TP Ho Chi Nlih 102,906 126,581 4.56 4.63 36,589 95,707 2.72 4.03 Song 8e 81,375 104,222 19.53 18.41 117,250 19.53 Tay Virh 75,111 66,083 11.92 10.33 96,571 198,250 13.52 17.88 Dong Naf 129,200 133,800 15.52 12.68 161,500 200,700 17.36 15.52 Vung Tau-Con Dao 46,000 67,500 6.14 6.14 92,000 135,000 8.69 8.69 Mekor. River Delta Long An 94,900 86,154 11.76 10.31 316,333 186,667 21.47 15.18 Dong Thap 130,333 148,556 10.77 10.77 1,173,000 111,417 32.30 9.32 An Guifg 147,500 177,400 10.44 10.44 295,000 147,833 14.76 9.53 Tien Glano 139,778 164,778 9.09 9.09 89,857 211,857 7.29 10.31 Ben Tre 114,889 121,400 8.92 8.46 173,429 10.11 Cuu Long 99,800 106,412 9.04 8.50 Kau Giang 123,667 148,944 10.44 10.44 247,333 141,105 14.77 10.16 Klan Glang 75,846 92,154 12.36 12.36 328,667 171,143 25.74 16.85 Ninh Hea 101,083 129,583 14.25 14.25 404,333 62,200 28.50 9.87 - 163 - STATISTICAL ANNEX TsbL.e 4.4: POPULATION PER BED IN HOSPITALS AND BASIC HEALTH CARE CENTERS, BY PROVINCE, 1979 AND 1989 Population served PopulatIon served by each hosRittl bed by each polyclinic bed Prov1nce 1979 1989 1979 1989 Mountain and Mid-lnd Ha Tuyen 550.89 645.60 489.56 828.49 Coo Sang 739.40 629.29 674.79 m.05 Lang Son 756.33 576.96 1157.38 2332.06 Lai Chau 574.55 658.35 549.57 756.13 Hoang Lien Son 458.93 519.85 375.55 605.22 Bac Thai 685.59 679.87 592.67 847.04 Son La 465.70 497.66 552.75 5126.32 Quang Ninh 357.30 421.19 483.19 706.87 VInh Phu 561.63 681.70 365.86 484.32 Ha Bac 634.70 717.55 501.61 812.76 Red River Delta Ha Nol 649.74 662.93 967.31 1150.64 Haf Phong 337.84 374.51 605.58 864.70 Ha Son Blnh 573.61 699.16 378.15 587.85 Hal Hung 535.54 631.94 601.29 752.49 Thai BInh 558.84 656.94 507.71 570.41 Ha Nam Ninh 690.16 715.85 565.11 847.49 Central Coast Northland Thanh Hoa 686.55 740.89 422.38 504.76 Nghe Tinh 653.76 719.98 391.01 1026.53 suwng BInh 207.84 492.42 166.30 915.49 3yang Trf 169.80 720.09 135.86 758.18 Thus Thfin HNu 312.16 1,100.37 249.76 1040.02 Central Coast Southland Qu n Nm Da-Nang 536.03 448.89 1194.10 1017.03 Quang Nga; 355.31 784.05 469.97 1063.27 Binh Dinh 428.35 474.32 566.58 1071.51 Phu Yen 298.84 548.55 429.76 1492.56 Khanh Hoa 381.50 524.04 548.63 1362.50 Thuan Hai 638.58 608.32 963.24 1343.91 Central Hiahland G*I Lal-Kon Tum 468.37 484.99 443.52 725.27 Dal Lal 471.16 543.45 477.70 1300.67 Lem Dong 406.28 468.62 1127.91 1718.28 North-East Southltnd TP Ho Chl Minh 355.19 362.16 7301.55 6595.63 Song Be 600.55 701.35 877.36 1759.29 Tay Nirih 786.0S 620.42 1365.66 1708.84 Dong Nal 849.44 679.12 1711.26 2403.35 Vung Tau-Con Dmo 368.00 450.00 1533.33 2700.00 Mekona River Delt Long An 774.69 727.88 969.36 1403.76 Dong Thep 1,130.06 979.85 2313.61 1535.59 An Cfang 1,146.97 1,039.07 3248.90 1385.70 Tien CIang 882.81 910.00 1379.39 1239.18 Den Tre 1,050.81 902.83 1509.49 1175.51 Cuu Long 795.43 652.33 1584.13 1102.99 Hau Clang 915.67 863.91 2396.12 2580.08 Kien Clfng 1,202.U 908.19 3015.29 1650.00 Nfnh Hal 817.39 975.11 1619.49 1806.39 - 164 - STATISTICAL ANNEX ITbt 4..: AVERACE SIZE OF POPULATION SERVED BY HEALTH PERSONNEL, BY PROVINCE, 1979 AND 1969 Pomutation Der P03tato Physicaln Pharmacist Nurse Physician Ph. nacist murse and Assat and Asst nd and Ass t and Ass It nd Province Physician Phar acist Mf.dif. Physician Pharmacist Nfdilfo Hountafn nd Mfd-land He Tuyen 863 7,352 923 752 7,036 5U Coo lang 620 5,001 789 689 5,815 746 Lano Son 765 4,526 869 696 4,686 640 Laf Chau 650 4,811 808 581 3,435 576 Hoon Lion Son 760 5,670 1,009 648 3,855 48 Sac Thai 1,050 6,358 1,144 1,213 6,524 1,052 Son La 761 2,952 918 751 2,161 405 ouang Ninh 816 5,806 837 794 3,43 613 Vinh Phu 1,094 7,958 1,326 805 5,077 625 He Sac 1,225 5,404 1,104 1,104 4,959 691 Red Rfv r Deltr Ha Nof 1,053 4,548 577 1,229 4,227 660 HNa Phon 964 4,055 m 1,060 4,307 553 Ha Son Ifih 1.095 4,801 890 1,026 4,133 510 Hal Hung 1,021 5 483 607 1,022 5,883 481 Thai liri 1,005 6,231 1,298 885 6,488 824 Na NM N_nh 1,026 5,617 706 1,060 5,629 580 Centrat Cot Northtan Thanh Hoo 1,004 4,46 1,200 929 5,841 378 Nihe Tfnh 1,270 5,140 698 1,069 4,527 474 uvng linh 4,012 7,647 2,273 357 2,637 224 yano Tr 676 5,527 556 292 2,154 183 Thue Thien Hue 1,397 11,282 1,326 536 3,960 337 Central Coast Southlnd suan NM Do-Nang 1,081 7,242 648 2,730 7,327 495 Quan Ngai 1,194 8,983 704 1,211 4,762 316 Sfri Dinh 1,615 7,324 964 1,460 5,741 381 Phu Yen 1,079 8,335 898 1,124 5,385 332 Khah Hoe 958 10,900 892 1,435 6,875 423 Thuan Hal 1,88 9,825 1,107 2,540 12,392 704 eontratI Hofi Ols Lat-Kon Tun 1,159 14,119 674 2,417 10,636 404 Dai Lal 1,160 13,363 913 1,668 12,359 374 Lm Dong 929 6,030 872 2,639 6,258 1,078 North-East Southlnd TP Ho Chi Ninh 1,160 3,115 669 11 ?? ?1 Song Be 1,267 10,083 803 2 300 9,300 639 Toy Mnh 1,123 9,439 757 4,024 13,796 842 Dong Nal 1,435 13,840 1,208 4,988 17,459 1,125 Vung Tau-Con Dao 1,007 2,935 844 1,353 4,600 5U MakoM River Dolts Long An 1,137 9,574 1,104 5,102 20,630 829 Don Thap 1,264 11,432 623 8,146 22,558 1,602 An Slng 1,208 13,042 1,104 4,917 37,821 1,588 Tfin Clang 1,129 11,237 762 3,574 17,718 1,058 Sen Tro 1,129 13,199 1,208 5,249 22,000 1,499 Cuu Long 1,295 7,865 964 6,238 26,732 1,184 Hau Clang 2,411 9,s74 1,035 4,925 21,000 1,201 Kian Slng 1,011 11,630 1,198 6,847 29,879 1,590 NInh NHa 1,207 14,812 1,051 4,931 20,914 1,385 - 165 - STATISTICAL ANNEX Table 4.6: PRODUCTION AND DISTRIBUTION OF PHARMACEUTICAL PRODUCTS (Million dong) 1980 1985 1986 1987 1988 1989 Current Prices: production 234.5 220.5 1,937 2,171 2,825 3,107 - Central level 139.1 62.3 418 531 678 775 - Local level 95.4 158.2 1,519 1,640 2,148 2,332 Distribution 376.2 1,300 1,979 18,133 20,905 23,792 Constant Prices (1982 dong): Production 214.3 37.5 56.1 15.7 5.0 3.3 - Central level 119.1 10.5 12.0 3.8 1.2 0.8 - Local level 95.2 27.0 44.1 11.9 3.8 2.5 Distributlon 345.1 221.0 57.3 131.2 36.9 25.3 Source: MOH STATISTICAL ANUX Table 4.7: THERAPEUTIC ACTIVITIES, 1980-89 (Absolute numbers in '000) 1980 1985 1986 1987 1989 No. of Consultations 106,903 129,726 130,487 129,718 77,893 No. of In-Patients 7,342 6,288 6,431 6,510 5,105 No. of Out-Patients 1,311 2,864 2,947 2,556 1,287 No. of Radiological Exams 1,533 1,802 1,920 2,153 2,136 No. of Radiographs 670 604 612 653 1,159 No. of Laboratory Exams 36,683 25,369 24,644 31,066 26,858 No. of Surgical Operations 202 294 295 417 357 Total No. of Treatment Days 39,856 51,423 52,616 53,444 37,591 Average Length of Stay (Days) 9.1 8.2 8.1 8.2 7.1 Average No. of Days per Month each bed is used 27.0 28.2 29.0 28.5 21.3 Average Consultations per Inhabitant 2.0 2.2 2.13 2.07 1.2 Source: Ministry of Health, 1990. - 167 - STATISTICAL ANNX T ble 4.8: OCCUPANCY RATES OF HOSPITALS AND BASIC HEALTH CARE CENTERS, BY PROVINCE, 1979 AND 1989 Occupancy rate Hospital of CHCs and occu.ancv rate oolvelnics_ Provinee 1979 1989 1979 1989 Mountafn wnd Mid-lnd 44.68 50.47 50.31 41.96 Hn Tuyn 44.52 52.49 50.09 40.90 Cao Bang 42.07 48.91 46.10 39.81 Long Son 55.68 67.84 36.39 16.78 Loa Chau 41.95 43.15 43.86 37.57 Hoang Lien Son 43.94 50.57 53.70 43.43 Bac Thai 45.04 47.62 52.10 38.22 Son La 51.75 59.35 43.60 36.09 Cuang iunh 54.08 58.68 39.99 34.84 Vinh Phu 38.36 39.94 58.89 56.21 Ha Bac 42.32 50.81 53.55 44.86 ReldivygrJftsjf 68.25 51.61 45.62 40.70 Ha Noi 57.19 59.36 38.42 34.20 Hai Phong 57.73 65.93 32.21 28.55 HN Son Binh 37.87 42.84 57.45 50.96 Hal Hung 51.28 49.74 45.68 41.77 Thai Binh 45.59 64.09 50.17 50.78 Ho Nan MNinh 40.46 47.98 49.39 40.53 Centrat Coast Northians 35.59 50.18 54.74 41.94 Thwnh Noa 31.25 77.30 50.80 57.51 Hsho Tinh 36.28 55.25 60.66 38.75 Ouan Binh 40.27 57.26 50.33 30.80 Gyang Trl 40.27 46.79 50.33 42.54 Thua Thien Hue 40.27 40.76 50.33 43.13 Central Coast Southlond 50.77 62.04 32.42 28.53 ouan Ham Da-Nang 62.51 60.33 28.06 26.62 Ouang mgaI 53.69 53.39 40.59 39.37 Ilnh Dinh 53.69 66.17 40.59 29.29 Phu Yen 53.18 67.43 36.93 24.78 Khanh Hoa 53.18 65.68 36.93 25.26 Thuan Hal 55.10 62.40 36.53 25.24 Central HIahtond 52.04 65.41 41.37 29.66 Cie Lal-Kon Tum 46.88 57.08 49.51 38.17 Dal Lal 45.83 68.12 45.21 28.46 Lms Dong 67.54 72.70 24.33 19.82 North-East Southland 75.76 75.48 16.69 13.45 TP Ho Chi Minh 86.93 79.89 4.22 4.38 Song be 58.09 63.36 30.76 25.26 Toy NInh 59.72 69.19 34.38 25.12 Done Hat 64.34 74.68 31.94 21.04 Vurg Tau-Con Dao 79.36 83.33 19.04 13.88 MNlkno River Delto 63.44 58.28 31.52 33.32 Long An 54.03 63.17 43.18 32.75 Dong Thep 64.47 56.85 31.49 36.27 An Cang 67.51 52.28 23.83 39.20 Tin CIang 54.24 52.87 34.72 38.82 Ban Tre 58.96 53.84 41.04 41.35 Cuu Long 66.34 62.13 33.31 36.74 Mau Claon 67.27 65.12 25.70 21.80 Kin Clang 66.83 58.33 26.65 32.10 Minh Hal 64.86 55.15 32.74 29.77 - 168 - STATISTIC ANNEX Table 4.9: PUBLIC EXPENDITURE ON HEALTH (MILLION DONG), 1984-90 1984 1985 1986 1987 1988 1989 1990 Recurrent Exoenditure 3,478 940 3,996 16,021 57,503 201,060 314,000 Heatth Services 2,445 675 2,881 11,920 43,529 142,753 226,080 Curative 2,054 547 2,276 9,894 35,694 97,072 146,952 Preventive 391 128 605 2,026 7,835 45,681 79,128 Salaries 699 172 799 3,124 8,970 39,810 66,568 Research 52 19 68 577 2,128 8,445 13,502 Others */ 282 74 248 401 2,875 10,053 7,850 Investment Exoenditure 211 400 275 833 4,405 25,500 50,803 Civil Works 91 177 98 307 1,559 7,472 15,851 Equipment 101 182 161 500 2,687 16,320 30,990 Training 19 41 15 27 159 1,709 3,963 Total Exoenditure 3,689 1,340 4,271 16,854 61,908 226,560 364,803 Ai X of Total Exoenditure Recurrent 94.3 70.2 93.6 95.1 92.9 88.7 86.1 Investment 5.7 29.8 6.4 4.9 7.1 11.3 13.9 aI Includes maintenance, repairs of equinpmnt and buildings Notes Includes all public expenditures (central, provincial, district) except communes Source: NON and Staff Estimates - 169 - STATISTICAkL ANKX TIbtI 4.10: PUBLIC EXPENDITURE ON HEALTH (MILLION DONG, 1982 PRICES), 1964-90 1964 1965 1986 1967 1966 1969 1990 Recurrent Exoenditure 113 160 116 116 102 214 243 Heatth Service 80 115 83 86 77 152 175 Curative 67 93 66 71 63 103 113 Preventive 13 22 la 15 14 49 61 Salerfes 23 29 23 23 16 42 51 Research 2 3 2 4 4 9 10 otherse / 9 13 7 3 5 11 6 Investemnt Expenditur 7 68 a 6 a 27 39 Civil Works 3 30 3 2 3 8 12 Equlpment 3 31 5 4 5 17 24 Training 1 7 0 0 0 2 3 Total Exonnditure 120 228 124 122 110 241 282 As K of Total Exoenditure Recurrent 94.3 70.2 93.6 95.1 92.9 88.7 86.1 Investment 5.7 29.8 6.4 4.9 7.1 11.3 13.9 of Includes mintnnce, repaire of equipment nd bultdings Note: Includes all public expnditures (central, provincial, district) except coonmes Source: NON and Staff Estimates - 170 - STA'BTXCAL ANX Tale 4.11: DISTRIBWTION OF PUBLIC HEALIN EXPENDITURE (C), 1964-90 1984 1965 196 197 1988 1989 1990 Recurrent Exownditure 100.0 100.0 100.0 100.0 100.0 100.0 100.0 HNelth Services 70.3 71.8 72.1 74.4 75.7 71.0 72.0 Curatfve 84.0 81.0 79.0 83.0 82.0 68.0 65.0 Preventive 16.0 19.0 21.0 17.0 18.0 32.0 35.0 Slaories 20.1 18.3 20.0 19.5 15.6 19.8 21.2 Reserch 1.5 2.0 1.7 3.6 3.7 4.2 4.3 others */ 8.1 7.9 6.2 2.5 5.0 5 2.5 IMrtmnt Expenditure 100.0 100.0 100.0 100.0 100.0 100.0 200.0 Civil Works 43.1 44.2 35.8 36.8 35.4 29.3 31.2 Equilnt 48.0 45.6 58.7 60.0 61.0 64.0 61.0 Training 8.9 10.2 5.5 3.2 3.6 6.7 7.8 Total Exndltur As X of Total Exoenditur Curatfve 55.7 40.8 53.3 58.7 57.7 42.8 40.3 Proventive 10.6 9.6 14.2 12.0 12.7 20.2 21.7 Salary 19.0 12.8 18.7 18.5 14.5 17.6 18.2 Civil Works 2.5 13.2 2.3 1.8 2.5 3.3 4.3 Equipmnt 2.7 13.6 3.8 3.0 4.3 7.2 8.5 af Includes mintenane ropairs of equfpewnt and bultdings Mote: Includes all pLhiLc *xpenditures (centralt provincial, district) except comes Soure: NON nd Staff Eat mtes - 171 - STATISTICAL ANNEX Table 4,12: Provincial Health Expenditure, by Province, 1989 Provincial Government Health ExDenditure as % of as % of total Total Per Capita provincial provincial Province (dong) (dong) GDP govt. exp Ha Tuyen 162,663 158 0.07 0.55 Cao Bang 987,431 1,747 0.74 4.48 Lang Son 1,672,786 2,738 1.20 6.64 Lai Chau 2,863,187 6,540 2.62 13.79 Hoang Lien Son 1,031,237 999 0.37 3.10 Bac Thai 137,407 133 0.04 0.56 Son La 2,863,090 4,199 2.09 13.54 Quang Ninh 2,517,089 3,096 0.47 4.63 Vinh Phu 2,976,882 1,648 0.45 7.68 Ha Bac 2,361,353 1,144 0.36 6.69 Ha Noi 13,741,334 4,496 0.80 8.34 Hai Phong 9,940,643 6,867 1.63 22.25 Ha Son Binh 2,065,151 1,123 0.28 4.21 Hai Hung 3,490,379 1,427 0.37 5.97 Thai Binh 2,591,443 1,587 0.37 7.86 Ha Nam Ninh 2,847,717 902 0.24 4.49 Thanh Hoa 3,818,475 1,276 0.47 5.81 Nghe Tinh 1,921,486 536 0.27 1.92 Quang Binh 592,716 912 0.47 2.67 Qyang Tri 709,987 1,548 0.85 3.53 Thua Thien Hue 494,376 555 0.14 2.49 Quan Nam Da-Nang 8,131,637 4,678 1.23 16.06 Quang Ngal 2,542,241 2,440 0.81 11.97 Binh Dinh 874,783 703 0.21 3.29 Phu Yen 1,017,554 1,585 0.32 4.43 Khanh Hoc 1,382,196 1,691 0.44 5.20 Thuan Hai 926,666 793 0.22 2.91 aia Lai-Kon Tum 61,903 71 0.03 0.19 Dai Lal 708,603 726 0.28 1.74 Lam Dong 1,832,157 2,866 1.20 8.36 TP Ho Chi Minh 60,846,455 15,505 0.75 27.72 Song Be 3,650,609 3,893 1.52 13.25 Tay Ninh 2,014,064 2,540 0.87 6.42 Dong Nai 4,229,046 2,107 0.47 8.37 Vung Tau-Con Dao 860,111 6,371 0.20 5.38 Long An 1,312,519 1,172 0.17 2.39 Dong Thap 2,338,625 1,749 0.24 3.75 An Giang 3,727,890 2,102 0.36 5.35 Tien Giang 1,881,212 1,268 0.20 3.37 Ben Tre 1,637,553 1,349 0.36 6.60 Cuu Long 2,508,635 1,387 0.24 6.41 Hau Giang 5,534,044 2,064 0.36 8.60 KLen Giang 2,554,211 2,132 0.29 5.55 i4nh Hal 2,093,106 1,346 0.26 5.29