FINAL REPORT Assessment of the best practices in HIV/AIDS harm reduction programs among civilian population and prisoners in the Russian Federation. Implementing agency: Open Health Institute The project is funded by the World Bank 1 Table of contents EXECUTIVE SUMMARY..................................................................................................................3 INTRODUCTION................................................................................................................................5 METHODS...........................................................................................................................................8 LITERATURE REVIEW...................................................................................................................12 RESULTS: SEMI-STRUCTURED INTERVIEWS..........................................................................33 RESULTS: REPERTORY GRID TECHNIQUES ............................................................................55 CONCLUDING .................................................................................................................................72 KEY POLICY IMPLICATIONS.......................................................................................................75 APPENDIXES ...................................................................................................................................77 2 EXECUTIVE SUMMARY The current report is devoted to the issue of best practice in prison and civil harm reduction projects in the Russian Federation. During the initial stage the comprehensive literature review was conducted in order to investigate the available information on prevention of HIV associated with injecting drug use. Available literature, including publications in biomedical journals, official documents and other relevant information sources was reviewed. Literature review revealed a large scale comprehensive attempt to address the issue of best practice in Russia that was undertaken in 2001. It highlighted the importance of understanding the evolution of harm reduction strategies in Russia and factors determining success and failures. Conceptual framework for exploring the issue of best practices in HR was developed based on the information retrieved. It splits context into factors of different degree of stability ranging from those that can be influenced by the best practices of the HR project to totally non-modifiable. The conceptual framework highlighted that the best practice harm reduction sites are those which achieve good results in a situation of hostile environment and hard-to-raise funding, improving the modifiable context characteristics to less inimical in the course of their functioning, while reaching the objective of being effective, ensuring multi-sectoral collaboration and sustainability. Literature review was followed with semi-structured interviews and repertory grids technique. Repertory grid technique allows quantification of value of success factors based on HR experts' evaluation of the projects which are believed to be successful to different degrees. Content analysis of the interview transcripts included 57 anonymous semi-structured interviews conducted with key informants in six regions of Russia and seven interviews with the federal level decision- makers. Repertory grids analysis was based on 36 repertory grids completed by experts in civil HR projects and 5 repertory grids completed by experts in prison projects. The main results of semi-structured interviews and repertory grids technique indicate importance of external factors. Societal environment and sufficient funding, preferably both from regional and local governments and from external donors was viewed as of highest value in order to achieve sustainability of HR projects. Support of the authorities and law enforcement agencies, were emphasized as additional necessary components for success. The problem of opposition of police on the level of junior street-based staff was mentioned repeatedly. The issue was hard to overcome by trainings, due to high turnover rates of these personnel. The internal factors included experienced staff involved in program realization, wide range of services provided, as well as ensuring that target group is informed on the existence of HR projects. Importance of enrolment of ex-IDUs into outreach work was emphasized. Given the closed nature of the drugs distribution systems now the current IDUs might play important role in linking to clients. The issues of 3 involvement IDUs into outreach work and management of them, unless during stable long term abstinence from drug abuse, were numerous. Data on prison projects indicated the issues of access to services, coverage, length of the projects and availability for all prisoners. The major problem was related to relatively narrower range of provided services, which were not of the same scope as in civil projects. Need to link prison HR projects with external NGOs and other organizations giving continuity of care for clients of harm reduction projects was also emphasized. The main distinction of prison projects was lower exposure to external factors due to a closed nature of the institutions. Influence was mainly limited to the prison administration and the Federal Penitentiary Service. In general, the experts on prison project were more pessimistic about future of the prison harm reduction projects comparing with civil projects. 4 INTRODUCTION HIV/AIDS is a global public health threat all over the world. HIV incidence rates in some Eastern European countries are now among the highest worldwide. Countries of the former Soviet Union are experiencing one of the sharpest increases in HIV incidence ever observed. Until the mid-1990s, the number of HIV cases diagnosed annually in Russia was fewer than 200, and the cumulative nationwide total number of cases in 1995 was only about 10001. From 1996 on HIV incidence in Russia started a rapid accelerating climb. By mid-2001 over 140 thousand infections had been officially registered2 and by April 2006 the number of registered cases has reached over 352 thousand. The incidence has been declining from 2001 to 2004, however in 2005 the incidence exceeded the figure of 2004 possibly marking the threat of the second wave of epidemic, which might be possible given that infection has several routes of transmission. (Figure 1). Still it is not clear whether what we see is the second wave as some experts claim decline in testing coverage in the period of lower rates of HIV in 2002-2004, as well as we can not clearly state whether the latest increase is due to sexual or intravenous injecting transmission as in over 55% of cases the route is not known3. However the real number of HIV cases in the Russian Federation is most likely to be several times higher then what's reported. The national statistics represents only officially registered cases which are picked up through tests demanded by those conscious about being infected and through routine screening procedures institutionalized. At the same time risk of HIV infection is still distributed quite unevenly in the population, with those being under the greatest risk, having much smaller chances of being tested. According to UNAIDS data, about 860 thousand people are currently living with HIV/AIDS in Russia, and the range of experts' assessments is between 420 thousand and 1 million 400 thousand4. HIV tends to spread faster among highest risk groups that often lack routine access to healthcare services and therefore are not in those groups which are likely to be screened or to take test voluntarily. For instance intravenous drug users (IDUs) tend to avoid contact with formal medical care system. Even though the proportion of HIV cases associated with IDU in Russia is 1UNAIDS (1998). HIV infection in Eastern Europe. Geneva, Switzerland: UNAIDS/WHO Joint Program on AIDS. 2 Pokrovsky V. (2001, January). Report of the Russian Federal Center for the Prevention of AIDS. Moscow, Russia: Russian Federal Center on AIDS. 3Federal AIDS centre. 2005 report. 4Human Development report 2005 for the Russian Federation, UNDP, 2005 5 estimated to decline from 90%5 of all cases in 2003 to approximately 70% of cases in 2004-20056, they still represent the majority of known cases. In addition route of transmission is hard to determine exactly, hence this data should be treated with caution. Figure 1. Number of new cases and total number of cases in the Russian Federation, 1996-2005 (National AIDS center). 350000 Newcases 300000 Cumulativecases 250000 200000 150000 100000 50000 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Assessments of injecting drug use prevalence in Russia are also ambiguous. The Russian Ministry of Interior has reported that the number of IDUs could be between 3 and 4 million, while the National Scientific Centre on Addictions in Moscow estimate is around 2 million7. The study with capture-recapture technique conducted in several Russian cities estimated the prevalence of IDUs at 5.4% of the total population8. Given often unclear estimates of the sizes of risk groups it is hard to predict the future of the HIV epidemics in Russia. Different forecasts estimated that in the period from 2000 till 2025 HIV incidence in Russia will reach 4 to 19 millions, while the number of AIDS deaths will constitute from 3 to 12 millions9. The US National Intelligence Council projected that even by 2010 Russia will reach 4 to 8 million HIV/AIDS cases10. Head of the Russian National AIDS Center professor 5 Federal AIDS Centre. HIV Infection: Informational Bulletin (Number 25). Moscow: Federal AIDS Centre. 2003 6Federal AIDS center of Russian Federation, Annual report. 2005. 7 Koshkina E. Prevalence of drug addiction diseases in the Russian Federation in 2000 according to official statistics. Voprosy Narkologii [Addiction Problems], 2001, 3, 61­71. 8 Platt L, Hickman M, Rhodes T The prevalence of injecting drug use in a Russian city: implications for harm reduction and coverage. Addiction. 2004 Nov; 99(11): 1430-8 9Nikolas Eberstadt. AIDS future. Foreign affairs, November/December 2002, Volume 81 Number 6 10National Intelligence Council. The next wave of HIV/AIDS: Nigeria, Ethiopia, Russia, India, and China. ICA 2002- 04D, NEPtember 2002: 11-12 6 Pokrovskiy predicts Russia reaching the toll of 100 000 of deaths from AIDS by 2010 as a very minimum even when estimates are based only on the numbers and dynamics of officially registered HIV cases. The forecast model which based on the assumptions that number of IDUs in Russia is about 2.5 mln., and HIV epidemic in the nearest 3-4 years will continue to concentrate among IDUs, predicts that by 2008 up to 50% of IDUs could be infected with HIV, meaning that number of HIV cases in Russia in 2008 would reach ca. 1.25 mln.11. Other authors have attempted to look beyond the rates of infection. E.g. Ruyl' et al. has modeled the economic consequences of HIV in Russia12. The forecast suggests that in the absence of HIV treatment loss of human resources will be hard hit on economics. Even based on optimistic forecast AIDS related mortality in Russia will reach 21000 persons per month by 2020. The same authors suggested that GNP will decrease by 4.15% in 2010 and if proper measures will not be undertaken the loss may reach 10% in 2020. Investments are expected to decrease even more rapidly than GNP - by 5.5% in 2010 and by 14.5% in 2020 that might have devastating effect on economic development. Given the grim predictions of the trend of epidemic and large role IDUs play in it, it seems quite important to address the issue of drug abuse. However strategies of drug prohibition which to different extents exist in all countries and War on Drugs, that originally was initiatied by the United States government in order to fight against the legal use of certain drugs are facing a number of challenges in the 21st century. For instance, the development and growth of harm reduction movement has pushed drug policies in many countries from the strategy of prohibition to the more regulated and tolerant approaches, while drug prohibition appears to be unable to prevent the increasing cultivation, use, and normalization of cannabis throughout the world13. One of the main reasons why drug prohibition policies fail is that such policies increase IDUs risk for HIV infection14. Another reason calling for measures beyond fighting illegal drugs use relates to problems of overcoming the addiction. Study conducted among IDUs in Moscow in 2002 indicated that 2/3 of Moscow IDUs tried to stop using of injecting drugs. In majority of cases (62.2%) period of abstinence did not exceed six months. Numerous studies conducted in other countries have shown large failure rates among those attempting to quit drug use15. For instance the proportion of those 11V.V. Pokrovskiy HIV infection in Russia: forecast. Issues of virology 3 (2004): 31-34 12K.Ryulf, V.Pokrovskiy, V.Vinogradov. Economic consequences of HIV spreading in Russia. 13Harry G. Levine, Global drug prohibition: its uses and crises. International Journal of Drug Policy 14 (2003): 145-53 14R.N.Bluthenthal, J.Lorvick, A.H.Kral. Collateral damage in the war on drugs: HIV risk behaviors among injection drug users. International Journal of Drug Policy 10 (1999) 25-38 15Amato L, Davoli M, Ferri M, Ali R. Methadone at tapered doses for the management of opioid withdrawal. Cochrane Database Syst Rev. 2004 Oct 18;(4):CD003409. 7 able to abstain form use of intravenous drugs for over one year in a representative sample of IDUs in Saint Petersburg did not exceed 8% 16. Hence no matter how successful anti-drug policies would be in a given country majority of current drug users will not be able to overcome the addiction. This calls for additional policies which will allow working with IDUs in order to prevent spread of HIV among and beyond this population, to allow them to realize minimal human rights and to ultimately try improving their quality of life through better integration in the society. Such approaches are based on harm reduction principles. METHODS During the initial phase of the study a literature review was conducted in order to investigate the available information on prevention of HIV associated with injecting drug use. Search and analysis of literature included periodical scientific publications in biomedical journals, official documents and other relevant information sources. The literature review is structured in following chapters: overall introduction, describing epidemic of HIV and drug use; introduction to the harm reduction concept; history of harm reduction programs and HIV/AIDS prevention programs in penal system; evidence of the effectiveness of harm reduction, including penal system; economic assessment of harm reduction; concept of the best practice and its relation to harm reduction. The review is concluded with a conceptual framework which guides the study. Two meetings with stakeholders were organized in Moscow, where the methodology of the study, conceptual framework and toolkits were discussed and approved. Selection of the regions for the study was also done during the meetings. Representatives of the All Russian Harm Reduction Network, Saint - Petersburg School of Public Health, Federal Service for Surveillance of the Protection of Consumer Rights and Human Wellbeing, Central Public Health Research Institute participated in the meetings. The following regions were selected for the implementation of the study: Pskov, Kazan, Vologda, Tver, Voronezh and Balakovo. Pskov and Kazan were selected as they were recognized as the most successful in implementing harm reduction activities in both civilian sector and prisons. Vologda and Tver where harm reduction activities are also implementing in both sectors were chosen for comparison. Two more regions (Voronezh and Balakovo) with civilian harm reduction projects were included into the study with the additional support provided by Russian Harm Reduction Network. 16Musatov VA. Abstinence from drugs vs. reducing harm. Presentation at the 6th School for journalists writing on health issues. Pushkino, 24.05.06. 8 The study methodology and toolkits were sent via e-mail to the key persons who were unable to attend the meetings with the request to provide comments and suggestions. Feedback was received from the representatives of the Research Institute on Drug Addiction of the Russian Ministry of Health and Social Development and representatives of the International Harm Reduction Development Program (IHRD). All the received comments were taken into consideration and reflected in the final version of the study toolkits. Study toolkits include questionnaire (appendix 1) and repertory grids (appendix 2 and 3). Both toolkits were pre-tested in Moscow and Mytichi before implementation in other selected regions and used for the exploration of factors determining the effectiveness of civilian harm reduction programs and HIV prevention programs in prisons. Pre-tested interviews were also included into the further analysis. The questionnaire was used by the interviewers during semi-structured interviews with key stakeholders. In each region interviews with the following groups of respondents were conducted: providers of harm reduction services (both civilian sector and prison): heads of the civilian harm reduction projects and HIV/AIDS prevention projects in prisons, medical consultants of such projects, managers of outreach work, heads of medical departments in prisons, peer-educators regional policy-makers: representatives of the State Drug Control Service, AIDS Centers, narcological dispensaries users of harm reduction / HIV prevention services: drug users and inmates Standard list of respondents in each region include: - representative of regional AIDS centers - representative of State Drug Control Service - head of harm reduction project (civil and prison) - medical consultant of HR project - manager of outreach workers - outreach workers - IDUs, sex-workers - representative of narcological dispenser - peer educators (prison projects) - head of medical department (prison projects) 9 Number of respondents varies in the regions depending on desire, interest, availability of time and finally consent of potential respondents to participate in the survey (table 1). Number of respondents by groups is presented in table 2. Table 1. Number of interviews per region City Number of interviews Kazan 14 Vologda 7 Pskov 11 Tver 5 Voronezh 11 Balakovo 9 Total 57 Table 2. Number or respondents from six regions by groups Group Number of respondents State Drug Control Service 3 HR project leaders 6 (two of them were leaders of HR projects and chief doctors of AIDS centers at the same time) HR projects managers and 18 outreach workers Trusted doctors of HR 5 projects (medical consultants) AIDS Center 2 (plus two more counted in HR leaders column) Narcological dispensaries 4 HR project clients 8 Peer educators in prison 3 projects Leaders of prison projects 4 Heads of medical 4 departments of prisons TOTAL 57 The average number of respondents in each of the key regions selected for the study was ten. Besides interviews conducted in the key regions, several additional interviews were conducted in Mytichi with outreach manager of local HR project and head of narcological dispenser and in Moscow with representatives of the Research Institute on Drug Addiction of the Russian Ministry of Health and Social Development, Federal Service for Surveillance of the Protection of Consumer Rights and Human Wellbeing, NGOs. 10 Interviews get information on major achievements, challenges and barriers of harm reduction activities in both civilian and prison sectors, main strategies used in the development and expansion of harm reduction programs; main factors influencing the development and daily work of the programs, attitude towards harm reduction strategies from various stakeholders, etc. Each interview lasted about 40 minutes and all interviews were recorded with a dictaphones, transcribed and analyzed. Majority of the respondents interviewed also filled in the repertory grids. The most important advantage of repertory grid is that it is combining open structure of interview with quantitative data of questionnaire. Using this approach it is possible to conduct content-analysis as in interviews and then compare importance of the criteria like it was done in this study. Among disadvantages it should be noted that repertory grid techniques is very time- consuming (comparing to other questionnaire-based methods) and require analysis, which is frequently unavailable in standard statistical software packages. In addition, repertory grid is a idiosyncratic method and is difficult to use for analysis of several grids which frequently limits its applicability outside of clinical psychology where it is used in treatment. It should be noted, though, that situation is changing and contemporary textbooks on repertory grids include chapters on summary analysis of several grids17. Repertory grids were completed by the experts in harm reduction field. It could be representatives of harm reduction team or external experts who periodically visit harm reduction sites with the monitoring purposes and technical assistance. There were two types of repertory grids used in the study. The first type was designed for the experts familiar with one specific harm reduction program, either civilian or prison, for a long period of time. The second type was designed for the experts familiar with three different harm reduction programs, either civilian or prison. All repertory grids were completed by experts in the presence of interviewers in order to avoid possible mistakes during the completion. In total 36 repertory grids on civilian harm reduction projects were completed by the experts and 5 repertory grids were completed by the experts regarding prison harm reduction programs. Lower number of repertory grids on prison projects completed is explained by shorter history of HIV/AIDS prevention projects in Russian penal system and insufficient number of experts well familiar with these programs. 17Jankowicz D., The easy guide to repertory grids. ­ John Wiley and Sons, 2004 11 LITERATURE REVIEW Introduction to harm reduction concept Before discussing the best practices implemented by those attempting to prevent HIV among IDUs, it is necessary to properly define harm reduction. Concept of harm reduction often poorly understood by policy-makers in Russia and many other countries and operationalising the term is important in order to enable constructive discussion. Concept of harm reduction is rather broad and does not relate only to use of injection drugs. The definition of harm reduction from drug abuse is "any policy or program designed to reduce drug-related harm without requiring the cessation of drug use. Interventions may be targeted at the individual, the family, community or society." Examples of harm reduction programs which are supported by evidence of effectiveness include, in addition to needle and syringe exchange programs which prevent the transmission of HIV among IDUs: server intervention programs which decrease public drunkenness; and, environmental controls on tobacco smoking which limit the exposure to second hand smoke or minimizes the harm caused to smokers by consumption of nicotine containing products 18. Thus, in public health "harm reduction" is used to describe a concept aiming to prevent or reduce negative health consequences associated with certain behaviors. In relation to drug injecting, "harm reduction" components of comprehensive interventions aim to prevent transmission of HIV and other infections that occurs through sharing of non-sterile injection equipment and drug preparations. In relation to prisons, harm reduction describes a concept aiming to prevent or reduce negative health effects associated with certain types of behavior (such as drug injecting, tattooing, etc.) and with imprisonment and overcrowding as well as adverse effects on mental health19. The definition of harm reduction varies among those involved in HIV prevention nowadays, emphasizing different aspects of the concept. National Campaign Against Drug Abuse (NCDA) in Australia in 1989 defined harm reduction as a range of strategies, which, while not necessarily doing anything to decrease drug use, do decrease the likelihood of harm resulting from that use20. In 1993 the Ministerial Council on Drug Strategy defined harm reduction (minimization) as an approach that aims to reduce the adverse health, social and economic consequences of alcohol and other drugs by minimizing or 18http://www.doctordeluca.com/Library/AbstinenceHR/CAMH&HR03.htm 19WHO Regional Office for Europe. Status Paper on Prisons, Drugs and Harm Reduction, May 2005 20National Campaign against Drug Abuse. The national campaign against drug abuse 1985-1988. In: Evaluation and future directions. In: National Campaign Against Drug Abuse, vol. 12. Canberra: Australian Government Publishing service, 1989, p.37. 12 limiting the harms and hazards of drug use for both the community and the individual without necessarily eliminating use . 21 In 1993-1997 the Australian National Drug Strategy evaluation recommended the following definition of HR ­ "the middle ground where persons with widely differing views on drug policy can agree with one another regarding practical, immediate ways to reduce drug-related harm". The definition also states that "harm minimization should foster meaningful alliance and support for as wide a variety of potentially effective interventions as possible from all who share the goal of reducing drug-related harm, even though they may disagree about major policy approaches to the prevention of drug abuse per se" . The term `harm minimization' is sometimes used as an alternative to `harm reduction' but carries the clear implication that attempts are being made to reduce harm to the lowest possible level. It is always difficult to be sure that harm has been minimized but easier to establish that harm has been reduced. Also, the term `reduction' implies inclusion of `minimization'. Ministerial Council on Drug Strategy proposed the latest definition that covered a broader conceptualization of harm minimization in 1998. Harm minimization ­ policies and programs aimed at reducing drug related harm22. The United Kingdom Harm Reduction Alliance (UKHRA) defines harm reduction as a concept that includes policies, programmes, services and actions that work to reduce the health, social and economic harms to individuals, communities and society23. The following definition of harm reduction strategies is suggested by Canadian Centre on Substance Abuse (CCSA): "A policy or program directed towards decreasing the adverse health, social, and economic consequences of drug use without requiring abstinence from drug use"24. Although the latest definition is somewhat overly broad and all-inclusive, it is predominantly used throughout the document. There is a hierarchy of risks in harm reduction aimed to avoid HIV infection form drug use: 1. Stop or never start using drugs 2. If you use, do not inject 3. If you inject, use new materials and do not share needles, syringes, spoons, water, drugs 4. If you need to re-use equipment, clean and use your own 21Ministerial Council on Drug Strategy. National drug strategic plan 1993-1997. Canberra: Australian Government Publishing Service, 1993, p.20 22Ministerial Council on Drug Strategy. National drug strategic framework 1998-1999 to 2002-2003. In: Building partnerships: A strategy to reduce the harm caused by drugs in our community. Canberra: Commonwealth of Australia, 1998, p.15 23http://www.ukhra.org/harm_reduction_definition.html 24http://epe.lac-bac.gc.ca/100/200/300/ccsa-cclat/harm_reduction_policy/wgharm.htm 13 5. If you must share, clean or disinfect before use In this document however we shall use both the term Harm reduction (HR) and Harm reduction program (HRP) together with Needle exchange program, which represents a specific part of the HR work. The choice of the term will depend on what the program involved or emphasized. Prior to discussing in-depth harm reduction it's reasonable to describe history of the harm reduction development worldwide and particularly in Russia. History of harm reduction The first harm reduction program (HRP) was introduced in Amsterdam, the Netherlands, in 1984. The program was initiated by a drug-user self organization but then it was adopted by the Municipal Health Department of Amsterdam. Since the middle of 1980s, global expansion of HRP has occurred all over the world. By December 2000, there were at least 46 regions, countries, and territories that reported having at least one HRP25. In 1995, Lithuania was the first former Soviet state that introduced HIV/AIDS prevention measures among IDUs by adopting practices which have shown evidence of effectiveness in a number of developed and developing countries26. Three Lithuanian HIV prevention programs were included into the Joint United Nations Programme on HIV/AIDS (UNAIDS) and United Nations Office on Drugs and Crime (UNDCP) collection of best practice. The first documented harm reduction projects in Russia were established in 1996. By the end of 1997, there were four programs specifically attempting to prevent HIV transmission among drug users. These were: · Moscow outreach program, operated and funded byMédecins Sans Frontières­ Holland(MSF) · St Petersburg needle exchange bus, operated by Russian NGO Renaissance and funded by Médecins du Monde of France (MdM) · Yaroslavl needle exchange and peer-driven intervention, in which drug users were encouraged through the use of coupons redeemed for cash to participate in education, and to educate and recruit their peers, operated by Russian NGO Friends Helping Friends and the University of Connecticut ECHO Project, funded by International Harm Reduction Development program (IHRD) of the Open Society Institute (OSI). · Penza needle exchange and educational activities operated and funded by local authorities. 25Steffanie A. Strathdee, David Vlahov. The effectiveness of needle exchange programs: a review of the science and policy. AIDScience Vol. 1, No. 16, December 13, 2001 26HIV/AIDS Prevention among Injecting Drug Users in Lithuania: Best Practices. Central and Eastern European Harm Reduction Network, October, 2003 14 In addition, an experimental needle exchange program operated in the republic of Chuvashia, east of Moscow, for several months in 1997, funded by local authorities; and a short-lived needle exchange was started by an individual in Moscow in 1996 assisted by AIDS Prevention Action Network (USA) and funded by OSI. While these interventions were significant as pilots, they were not sufficient to prevent or control the HIV epidemics among drug users over a long term because of short time they functioned, their insufficient experience and inability to reach sufficient number of IDUs. In late 1997, OSI (Russian Public Health Program and IHRD) in collaboration with the Russian Federal Ministry of Health, MSF, MdM, a German NGO Interactive Drogenhilfe and the University of Connecticut, USA, started the Russian AIDS Prevention Initiative ­ Drugs (RAPID) to co-ordinate the expansion of HIV prevention activities among IDUs in the Russian Federation. This initiative comprised 3 steps: a) training in harm reduction, b) following rapid situation assessment in participating cities with the technical assistance from international project developers -- MdM, Interactive Drogenhilfe and the University of Connecticut, c) the submission of local project proposals on implementing harm reduction principles for funding by OSI-Russia. Between January 1998 and February 2000, MSF ran a training program to ensure that people working on HIV prevention among IDUs had the skills to conduct rapid situation assessments (RSAs) to determine the extent of drug use and related HIV risk in their city or region, as well as to plan harm reduction interventions. Over 200 participants from 61 Russian cities had completed the training, 62 rapid situation assessments had been conducted. In 1999, in collaboration with the Ministry of Health, OSI-Russia launched a program on "The Prevention of HIV infection among IDUs". The goal of the program was to reduce the spread of HIV/AIDS in the Russian Federation paying particular attention to IDUs. The mechanism to achieve the purpose was by awarding grants to Russian health care organizations, both governmental and NGOs; providing extensive technical assistance to grantees. The projects supported by this program included the following components: Needle exchange (fixed and mobile distribution) Outreach and group education of IDUs The provision of anonymous counselling services and testing for HIV and other infectious diseases Preparation of educational materials for IDUs Referral services Coordination with local municipal administration Coordination with local Ministry of Internal Affairs 15 Coordination with local health authorities Of the 61 cities, which participated in the MSF initial training in 1998-2000, 55 sought funding for a harm reduction program from OSI. By the end of 2000, OSI had been supporting 36 needle exchange projects in 32 regions. In addition to administering grants, OSI and other partners provided extensive technical assistance to the service providers: in 2000-2001, each project participated in 3 international trainings organized by IHRD; IHRD/OSI-Russia Technical Advisers conducted 47 site visits. In 2000, supported by OSI, MSF began a new training program inviting every Russian project to two trainings per year in Moscow, as well as providing short on-site trainings for field workers. OSI had committed to fund the harm reduction projects for a period of 3 years. It was expected that a five-year World Bank loan to Russian Federation, which includes a comprehensive harm reduction component, would begin in late 2001. However, negotiations have been delayed for technical reasons relating to the procurement of TB drugs. In order to maintain the momentum and ensure the sustainability of the projects, Department for International Development (DFID) of United Kingdom has granted support to OSI to continue through Harm Reduction Bridging Project. The project was designed to act as a bridge between OSI activities in harm reduction and the forthcoming World Bank loan. The HR Bridging project had five outputs: 1. scaling-up the existing multi-sectoral harm reduction programme for IDUs and CSWs; 2. developing capacity for monitoring and evaluation of harm reduction; 3. developing in-country capacity for implementing harm reduction; 4. expanding understanding among policymakers of the role of harm reduction in HIV prevention; 5. developing a sustainable roll-out strategy for harm reduction. Thus by the end of 2002 there were 38 regions implementing 51 harm reduction programs, Twenty five of them also covered commercial sex-workers and six of them covered prison inmates. Innovative harm reduction projects were developing in 11 small towns of Russia, taking consideration of peculiarities of small integrated communities often with singe pharmacy and limited number of providers of medical services. In 2003 Russia has accepted a $150 million loan from the World Bank to fight the spread of AIDS and TB, ending a four-year negotiation process. Altogether, 83 regions that expressed interest in the project were selected to implement the AIDS Control Component which includes implementation of prevention programs for high-risk groups In the same year (2003) Open Health Institute (OHI) became a principal recipient of the Global Fund to Fight AIDS, TB and Malaria grant in the third tender round (GLOBUS Project: 16 Global Efforts Against AIDS). GLOBUS is one of the largest HIV/AIDS prevention and treatment projects implemented in the Russian Federation by the consortium of non-governmental organizations: OHI, AIDS Foundation East-West, AIDS Infoshare, Population Services International and FOCUS Media. The project aims at stimulating an effective national response to HIV/AIDS in ten regions of the Russian Federation. One of the major objectives of the project is support for sustainable prevention programs to reduce HIV transmission among vulnerable groups which include further support and expansion of harm reduction programs for IDUs and HIV/AIDS prevention projects in penal system. By 2006 GLOBUS directly supports 25 HR programs in small and big towns of the ten GLOBUS regions and 10 HIV/AIDS prevention programs in penal system. In 2005 via Russian Harm Reduction Network (RHRN) GLOBUS project provides co-funding together with Open Society Institute and the Ford Foundation to twenty more HR programs in 18 regions. In 2006 RHRN continued support for 15 HR projects. The Country Coordinating Mechanism developed and submitted a proposal for the fourth tender round of the Global Fund to Fight AIDS, TB and Malaria for implementing the Program of "Promoting a Strategic Response to HIV/AIDS Treatment and Care for Vulnerable Populations in the Russian Federation". The proposal for a five-year program was approved and the project implementation has started in September, 2005. The Country Coordinating Mechanism approved the Russian Health Care Foundation as a principal recipient of the grant and an executive agency for the Program implementation. Although the program is mainly concentrated on provision of ARV treatment, still during five year period of its implementation 59 harm reduction programs for IDUs and 59 HIV/AIDS prevention programs for sex workers are expected to be supported. In 2005 Russian Harm Reduction Network submitted a proposal for the fifth tender round for the Global Fund to Fight AIDS, TB and Malaria. The proposal was approved by the Global Fund and in the first half of 2006 RHRN expected to complete procedures preceding signing of grant agreement. The main purpose of the fifth round tender project is to expand harm reduction programs in Russia. In total 33 HR projects are expected to be supported in the Global Fund fifth tender round. In addition to internationally supported initiatives, Ministry of Health and Social Development and Federal Service for Surveillance of Consumer Rights Protection and Human Wellbeing also supported activities on HIV prevention among IDUs. Although HIV prevention measures among IDUs have not been included in the official Federal program "Anti HIV/AIDS" 2002-2006, in 2006, President V. Putin pushed allocation of 100 mln. $ towards fighting HIV/AIDS, including HR component (~2 mln$). In 2006 the agreement was signed between Open 17 Health Institute and Federal Service for Surveillance of Consumer Rights Protection and Human Wellbeing envisaging implementation of 15 civil and 15 prison HR programs in 15 cities of Russia. Thus, situation with number of HR programs in Russia is changing very rapidly. In the absence of unified monitoring system of HR projects the additional research is needed in order to accumulate up-to-date information on this issue. Nevertheless crude information on annual number of HR programs in Russia funded by different source is summarized by RHRN in table 3. Table 3. Annual number of HR programs per year. Source of funds 1999 2000 2001 2002 2003 2004 2005 2006 Open Society Institute 23 36 Open Health Institute 42 50 51 45 25 39 Russian Harm Reduction Network 20 15 WHO 3 5 2 2 MDM 1 1 1 Russian Health Care Foundation 14 Local budgets 10 15 5 Others 1 1 3 3 3 3 3 TOTAL 24 38 44 66 74 55 46 63 Average budget of HR project might vary in the range of 10 000 ­ 25 000 USD. Still the main conclusions a study published by Rhodes et al. in 2006 arrives to the conclusion that an urgent scaling-up of harm reduction for IDUs in Russia still is the highest priority27. The study is based on the results of anonymous cross-sectional survey of IDUs recruited from non-treatment settings in Moscow, Volgograd and Barnaul, which were tested for HIV, HCV antibodies and syphilis by means of oral fluid samples. The conclusion repeats the one arrived of an earlier cross-sectional study conducted among IDUs in Toglliaty City in 2002, indicating lack of comprehensive national strategy to promote harm reduction activities in Russia28. Recent revision of evidence of trends in HIV infection, risk behaviour and HIV prevention associated with injecting drug use in the Russian Federation noted the critical importance of policy interventions to maximise syringe distribution coverage among IDU populations29. It should be mentioned that in order to fight HIV/AIDS epidemic in Russia, there should be sufficient expansion of findings due to huge size of the country. Average sum of donors' funding which in majority of cases is the only funding source is varies from 10 to 25 thousand US dollars per HR program. While in small towns this sum 27Rhodes T, Platt L, Maximova S, Koshkina E, et al. Prevalence of HIV, hepatitis C and syphilis among injecting drug users in Russia: a multi-city study. Addiction. 2006 Feb;101(2):252-66 28Rhodes T, Lowndesa C, Judd A. Explosive spread and high prevalence of HIV infection among injecting drug users in Togliatti City, Russia. AIDS 2002, 16:F25­F31 29Rhodes T, Sarang A, Bobrik A HIV transmission and HIV prevention associated with injecting drug use in the Russian Federation. International Journal of Drug Policy 15 (2004) 1­16 18 is adequate to cover sufficient number of IDUs by HR activities, in big towns such as Saint- Petersburg, for example, this amount allows reaching only about 3% of IDUs from their estimated size which is obviously not enough for fighting spread of HIV epidemic. However it should be mentioned that most of the fieldwork was conducted prior to 2005 ­ 2006 breakthroughs in governmental policies and NGO activities, so the situation might have improved significantly by the time of this study. Assuming that the estimates of Russian Ministry of Interior are correct and the number of IDUs is about 3 or 4 million, and given that an average cost per client served by HR program is $23 per year, then the overall amount needed to insure coverage in an ideal situation would be just around $100 mln. per year solely on HR. However this amount is based on outdated costs, excludes training and advocacy expenses, as well as the cost of educational materials and new approaches to reaching IDUs in the current situation, which are discussed below. Hence the amount needed to ensure comprehencive coverage in Russia is probably at least two fold greater. HIV/AIDS prevention programs in Russian penal system HIV/AIDS prevention programs in Russian penal system have developed even more recently in comparison with harm reduction programs for IDUs in civilian sector. Until the late 1990s the imprisonment rate in Russia was the highest in the world, but in 2000 a major legal reform released 200,000 convicts from the correctional colonies. By 2003, there were 1,014 Ministry of Justice facilities, containing 877,000 people, a rate of 670/100,000 population, still very high, but lower then the USA, with the highest in the world rate of 702/ 100,00030,31. In most countries in Europe and central Asia, rates of HIV infection are much higher among prisoners than among general population. There are significant variations in HIV rates among prisoners in different countries. The rates are generally higher in Eastern Europe compared with Western Europe, for example: Estonia (12% in 2002), Ukraine (7% in 200) and Russia (4% in 2002). Drug users are often over-represented in prison populations and usually continue using drugs while incarcerated. Prevention interventions have been introduced in various countries since 1990. Prevention measures in prisons include education on HIV/AIDS, voluntary testing and counseling, 30Abramkin V. [Tyuremnoe naselenie Rossii i drugih stran. Problemy I tendencii]. ROO Centr sodeistviya reforme ugolovnogo pravosudiya; 2003:27. 31Prestupnost' i pravoporyadok v Rossii. Statisticheskii aspekt. 2003. Stat. sb./Goskomstat Rossii M., 2003;85. 19 distribution of condoms, bleach and other disinfectants, needles and syringe exchange, substitution therapy ,32. The 1993 WHO Guidelines on HIV prevention and management of care in prisons states that "Preventive measures for HIV/AIDS in prison should be complementary to and compatible with those in the community. Preventive measures should also be based on risk behaviors actually occurring in prisons, notably needle sharing among injecting drug users and unprotected sexual intercourse"33. The revision of HIV prevention policies of prisons in Moldova, Hungary, Switzerland, Italy and Nizhnii Novgorod region of the Russian Federation showed that in spite of the availability of international guidelines, HIV prevention and management of care in prison is still unsatisfactory. The 1993 WHO Guidelines were fully implemented only in one country out of four reviewed (Switzerland), and partially in two (Italy and Hungary). Authors concluded that greater national and international efforts are needed to stimulate the debate and build consensus on harm reduction activities in prisons34. Using Nizhniy Novgorod as an illustration of how HIV prevention and management of care in prisons was unsatisfactory it should be highlighted that the WHO guidelines state that HIV prevention activities in prisons should be available at the same scope as outside prisons. The main distinguishing feature of Russian prison projects is that syringe exchange is prohibited and condoms are not available to the same degree as in civilian HR programs. Even though harm reduction programs for IDUs are implemented in many regions of Russia, they are not available at the same scope in penal system. Several conducted reviews evaluating prisons syringe exchange programs indicated that these programs are feasible and do provide benefit in the reduction of risk behavior and the transmission of blood-borne infection without any unintended negative consequences35,36. Still needle/syringe exchange is prohibited in the Russian prisons. HIV/AIDS prevention programs in penal system of Russia are limited to such activities as: training of prisoners and prison staff in order to increase their knowledge about HIV infection and ways of its transmission and prevention, support of peer education techniques in order to reach all prisoners with educational programs; provision of prisoners with condoms and disinfectants. 32HIV/AIDS Pervention, Treatment and Care among Injecting Drug Users and in Prisons. Ministrial Meeting on "Urgent response to the HIV/AIDS epidemics in the Commonwealth of Independent States", UNAIDS, 2005 33WHO guidelines on HIV infection and AIDS in prisons. WHO Global Programme on AIDS. 1993: 9 34Bollini P, Laporte JD, Harding TW. HIV prevention in prisons. Do international guidelines matter? Eur J Public Health. 2002 Jun;12(2):83-9 35Dolan K, Rutter S, Wodak AD. Prison-based syringe exchange programmes: a review of international research and development. Addiction. 2003 Feb;98(2):153-8. 36Stover H, Nelles J. Ten years of experience with needle and syringe exchange programes in European prisons. International Journal of Drug Policy. 2003: 14: 437-44 20 Study conducted in Siberian prison for drug dependent males indicated significant increase in knowledge of HIV transmission ways after prisoners' exposure to peer education. At the same time study indicated no difference in use of bleach to clean tattooing or injecting equipment before and after exposure of prisoners to peer education. Authors highlighted that Ministry of Justice should consider implementation of additional harm reduction strategies in prisons, such as methadone treatment and syringe exchange37. International non-governmental organization AIDS Foundation East-West has been implementing HIV/AIDS prevention programs in Russian penal system since 1999. Five-years project (1999-2003) was supported by Social Transformation Programme of the Netherlands Ministry of Foreign Affairs (MATRA), the Elton John AIDS Foundation and Swedish International Development Agency (SIDA). The project aim was to establish sustainable HIV/AIDS prevention and health promotion programs in the prisons of four regions of Russia: Penza, Omsk, Krasnodar and Moscow. Over 600 inmates from four prisons were trained as outreach workers; 500 medical professionals and psychologists were trained on pre- and post-HIV test counseling; 560 security, disciplinary and custodial staff received training on the reduction of risk in the work place; 100 representatives of the regional prison system administrations, heads of various institutions within the Directorate of Corrections, and heads and administrators from the colonies received training on health promotion in prison facilities. Information materials for inmates and prison staff were distributed in the targeted prison facilities, while inmates received access to supplies of bleach and condoms38. In Nizhniy Novgorod and Pskov HIV/AIDS prevention program in prisons has been implementing since 1999-2000. During that time both programs were funded by OSI-Russia with support from local administrations and Penal Reform International. Since 2004 these programs continued to develop with the support of the GLOBUS project. In 2001 when OSI-Russia received above mentioned HR Bridging grant from DFID for continuing and expanding activities on HIV prevention in 2001 ­ 2004 it also included a prison component. Open Society Institute and since 2003 Open Health Institute provided support to eight HIV/AIDS prevention programs in prisons from 2001 till 2004 as part of the Bridging project. Cross-sectional study was conducted in two Russian prisons (with and without HIV prevention programs) in 2003. Five hundred randomly selected inmates in each prison filled in anonymous self-administered questionnaire aimed to determine risky behaviors and to describe the 37Dolan KA, Bijl M, White B. HIV education in a Siberian prison colony for drug dependent males. Int J Equity Health. 2004 Jun 21;3(1):7. 38http://www.afew.org/ 21 knowledge about HIV/AIDS. All inmates in both prisons were offered to proceed through HIV and syphilis blood tests and anonymous urine tests in order to determine presence of opioids. The results indicated higher level of knowledge about HIV among inmates in prisons with HIV prevention program. Prevalence of such risky behaviors as unprotected sex, use of non sterile injecting equipment and failure to appropriately utilize disinfectants was significantly lower in the prison with HIV prevention program. In prison with HR project one new HIV case was revealed which is more likely to be explained by window period prior to incarceration. In both prisons large quantity of new syphilis cases were revealed which is likely to be explained by high levels of transmission with prisons. In the prison with HIV prevention program there were no positive opioid tests results detected while in the prison without HIV prevention program there were five positive tests' results (Table 4). It showed that HIV prevention programs in prisons are unlikely to lead to the expansion of drug use in prisons39. Table 4. Risky behaviors, knowledge and incidence cases in inmates from prison with and without HR activities Prison with HR project Prison without HR project Use of injecting drugs Sex Equal Tattooing Level of knowledge on > HIV/AIDS Needle disinfection > Use of condoms > New HIV cases 1 New syphilis cases 7 14 Positive opiate test - 5 In the framework of the GLOBUS project HIV/AIDS prevention programs in penal system were supported in ten regions of Russia. Over 700 medical and non medical prison staff members were trained on HIV/AIDS prevention methods. Chlorinated substances, vitamins and condoms for prisoners are procured and supplied to penitentiary institutions in ten regions of Russia in the framework of these programs. Several other projects are envisaging working in prison, including 39Bobrik A. Assessment of prevalence of HIV, syphilis and risk factors of its transmission in penal jurisdiction. Medico-social problems of socially conditioned diseases. Scientific works, Ministry of Health, Central Public Health Research Institute, Russia, Moscow, 2004 22 the projects implemented by the Russian Health Care Foundation supported through the World Bank loan and in the fourth tender round of the Global Fund. For instance the later is expected to provide trainings on voluntary counseling and testing (VCT) for the prison staff in 59 regions of Russia. The effectiveness of harm reduction There are various approaches to evaluation of effectiveness of public health interventions, including harm reduction programs and other approaches to HIV prevention. These can be based on more short term easy to measure and long term hard to measure results. As outcomes of harm reduction, such as reduction of new HIV cases are very hard to measure, reduction of risk behaviors are monitored as main indicators of the effectiveness of harm reduction programs, in addition to HIV incidence among IDUs. Even thought most of the studies conducted have shown the reduction of risky behaviors of interest, there are still debates related to the effectiveness of harm reduction activities. Non-longitudinal studies measuring only reported behavioral change and conducted by non- independent researchers can hardly yield firm epidemiological evidence and require cautious interpretation, still these constitute a paramount of research on the issue of harm reduction in the Central and Eastern Europe. For instance cross-sectional study conducted in Central/Eastern European cities: Prague, Budapest, Skopje, Krakow and Poltava, and five Russian cities: Nizhniy Novgorod, Pskov, Rostov-Na-Donu, St. Petersburg, and Volgograd aimed to assess HIV risk behaviors among participants in syringe exchanges40. Interviews with IDUs recruited from syringe exchange program included questions on risky injection risk behaviors for the 30 days prior to the first use of the syringe exchange program and for the 30 days prior to the interview (while using the syringe exchange program). The results suggested that IDUs participating in the exchanges appear to be responding very positively in reducing sharing of needles and syringes: From 1 to 29% of IDUs across the ten programs reported using needles and syringes used by others in the past 30 days which represents significant reduction in comparison with the same risky practice reported in the 30 days prior to use of the syringe exchange (from 7 to 47%). 40Don C. Des Jarlais, P. Friedmann, Jean-Paul Grund, et al. HIV risk behavior among participants of syringe exchange programs in central/eastern Europe and Russia. International Journal of Drug Policy, Vol. 13, Issue 3, NEPtember 2002: 165-70. 23 Harm reduction evaluation in Nigniy Novgorod showed the reduction in needles sharing from 54% to 24%, reduction in containers sharing from 59% to 22%, while flushing syringes with used solution reduced from 47% to 20%.41 In Sverdlovsk Oblast cross-sectional study was conducted to compare behavior change among three groups of IDUs: 1 month before HRPs were established, IDUs attended HRP after establishing and not attended42. The results have shown that HRP attendees are likely to report less HIV risk behavior than those not attending. In Svetlogorsk, Belarus deterministic epidemiological model was utilised to predict that between 1997 and 2000 the needle and syringe exchange intervention averted 414 HIV infections in Svetlogorsk (95% CI, 180­690) and caused a 6.5% decrease in IDU HIV prevalence compared to if there had been no intervention43. Similarly there were behavioural studies aimed at evaluation of harm reduction effectiveness conducted elsewhere. For instance the study by Des Jarlais et al. in New York City showed that needle exchange programs (NEP) decreased risky injection behavior by up to 73 percent, and injectors using NEPs were two-thirds less likely to contract HIV44. In Oakland injectors attending NEPs were 2.5 times more likely than other injectors to stop sharing needles after six months45. Data combined from four cities in US suggest that among NEPs participants the number of injections per syringe was decreased by 44­85 percent and the likelihood that syringes were used only once was significantly raised . 46 Cluster studies which tried to measure effect of harm reduction policies on HIV incidence have produced additional, albeit not strong by strict epidemiological standards evidence. For example, international comparison showed that in 29 cities with established NEPs, HIV prevalence decreased on average by 5,8% per year, but it increased on average by the same percentage in 51 cities without NEPs47. A before-after study of even less strong epidemiological design in Hawaii, 41 G.F.Moshkovich, Realization of Harm Reduction Project in the Nigegorodskaya oblast. The report on the first Russian science conference on HIV infection, AIDS and hepatitis B, C. 13-15 November 2001, Suzdal. 42R. Power, N. Nozhkina. The value of process evaluation in sustaining HIV harm reduction in the Russian Federation. AIDS 2002, 16: 303-4 43 P. Vickerman, C. Watts. The impact of an HIV prevention intervention for injecting drug users in Svetlogorsk, Belarus: model predictions. International Journal of Drug Policy, Vol.13, Issue 3, NEPtember 2002: 149-64 44Des Jarlais, D.C., M. Marmor, D. Paone. HIV Incidence Among Injecting Drug Users in New York City Syringe Exchange Programmes." Lancet, 1996, 348, 987­991. 45Blumenthak R.N., A.H. Kral, L. Gee, E.A. Erringer, B. Edlin The effects of syringe exchange use on high-risk injection drug users. A cohort study. AIDS, 2000, 14: 605-11 46Heimer R, Khoshnood K, Bigg D, Guydish J, Junge B. Syringe use and reuse: effects of syringe exchange programs in four cities. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;18 Suppl 1:S37-44. 47S. Hurley, D. Jolley, J. Kaldor, Effectiveness of needle-exchange programs for prevention of HIV infection. Lancet 349 (1997) 24 where a NEP was set in 1990, HIV infection rates fell from 5 percent in 1989 to 1.1 percent in 199648. Several larger scale reviews have attempted to summarize the evidence available so far on the issue of the effectiveness of harm reduction. The review of 42 studies on the effectiveness of syringe exchange programs in reducing HIV risk behavior and HIV seroconversion among IDUs that was conducted by Gibson et al.49 has shown that 28 of the reviewed studies found positive effects associated with use of syringe, two found negative association, and 14 found either no association or a mix of positive and negative effects. The main disadvantages of the reviewed studies marked by authors are disregard of the source of syringes factor, described in only one of the 42 studies, and of biological outcomes that were used only in 5 studies. There were just few systematic attempts to identify and control for confounders, while secondary use of NEP was not measured in any of the studies. While it seems relatively easy to control for such confounders as age and gender and the type and dose of drugs used, it was seldom undertaken in the studies. Additionally it is very hard to control for such confounders as different attitudes to ones own health and the "degree of marginalization" of IDUs. The main disadvantage with studies based only on self-reports of HRP attendance is that differential misreporting of HRP attendance could bias risk estimates50, .51 J.M.MacNeil and J. Hogle have conducted the review of over 400 abstracts of social, behavioral and evaluation researches conducted in 45 Latin America, Africa and Asia countries over a six-year period52. They also marked failure to track biological outcomes in these studies. Only three of the intervention studies actually measured HIV incidence and other STDs. Thus most of conducted reviews marked the following main disadvantages: the scarcity of research that includes biologic outcomes, disregard of the alternate source of syringes and participation in secondary exchange, insufficient control for confounding factors. They also note that longitudinal studies are ideal for conducting such investigations. Multi-sectoral cohort study that was conducted in three cities of Russia and aimed to assess effectiveness of harm reduction programs indicated statistically significant difference in the rates of 48Vogt, R.L., M.C. Breda, and D.C. Des Jarlais. "Hawaii's Statewide Syringe Exchange Program." American Journal of Public Health. 1998, 88, 1403­04. 49David R. Gibson, Neil M. Flynn, Daniel Perales. Effectiveness of syringe exchange programs in reducing HIV risk behavior and HIV seroconversion among injecting drug users. AIDS 2001, 15: 1329-341 50M. Safaeian, R. Brookmeyer, D.Vlahov et.al Validity of Self-reported Needle Exchange Attendance among Injection Drug Users: Implications for Program Evaluation. American Journal of Epidemiology Vol. 155, No. 2: 169-75 51L. Greenfield, G.E.Bigelow, R.K. Brooner Validity of interavenous drug abusers' self-reported changes in HIV high- risk drug use behaviors. Drug and Alcohol Dependence 39 (1995) 91-98 52 J.M.MacNeil, J.Hogle. Applying social, behavioral and evaluation research to developing country HIV prevention programs. AIDS 1998, 12 (suppl 2): 99-108 25 risky behaviors, HIV prevalence and incidence among IDUs in Russian cities with different level of HR activities as well as between HR participants vs. non-participants. Thus, HIV prevalence among IDUs in the city without harm reduction activities was 14,7% which was 5 times higher than in the city where harm reduction program was two years old (2,1%) and 15 times higher than in the city where harm reduction program has been operating since 1998 (0,3%) (p0.001). By the end of the one year follow-up period incidence rates were 13.2%, 2.5% and 0% accordingly and the difference was statistically significant. All new HIV infections were detected among non-participants of HR projects. Rates of such risky behavior as using non-sterile syringe varied from 19% to 31% among non clients of HR and from 6% to 8% among clients53. There were no statistically significant differences in age, gender, addition and socio-economic characteristics between the HR users and non-users groups. One of the difficulties in making conclusions of HRPs effectiveness is that in many countries there is a non-prescription availability of syringes through pharmacies. For example, study conducted in New Haven showed that 41% of IDUs reported syringes obtained from the pharmacies, 13% - from the NEP, 34% - both and 27% reported neither as their usual source of syringes in the past six months. Nevertheless those IDUs attended NEP were significantly less likely to throw away used syringes54. In countries where syringes are not sold on over the counter bases HR programs provide the only source of injecting equipment, hence their role is even more important. In contrast the study conducted in Baltimore, Maryland aimed to examine the effect of a Needle Exchange Program on the quantity and geographic distribution of discarded needles on the streets suggested that the initiation of NEPs does not result in decrease in the number of discarded needles on the street55. The adverse effect of NEP was shown in cohort study conducted in Montreal56. The cohort study showed that NEP users appear to have higher seroconversion rates then NEP nonusers. One of the possible explanations of the findings relates to confounding, while NEPs tend to attract 53K. Eroshina, et al. Cohort study on the effectiveness of Russian needle exchange programs. Handbook of the 16th Conference on the reduction of drug related harm. March. Belfast, Northern Ireland. 2005 54Khoshhood K. et al. Syringe source, use, and discard among injection-drug users in New Haven, Connecticut. AIDS&Public Policy Journal. 15(3-4): 88-94, 2000 Fall-Winter 55MC Doherty, RS Garfein, D Vlahov et al. Discarded needles do not increase soon after the opening of a needle exchange program. American Journal of Epidemiology, Vol 145, Issue 8: 730-37 56J. Bruneau, F. Lamothe, E. Franco High rates of HIV infection among injection drug users participating in needle exchange programs in Montreal: results of a cohort study. Americam Journal of Epidemiology, Vol. 146, Issue 2: 12994-1002 26 higher risk IDUs who engage in riskier behaviors compared to IDUs who tend to obtain syringes from other sources57, 58, 49. As for HRPs impact on Hepatitis C (HCV) the effectiveness is even more controversial. For example, a random-mixing epidemiological model was used to examine the potential impact of harm reduction interventions on HCV59. The results suggest that HRP is predicted to have little impact on HCV incidence and prevalence within realistic populations of IDUs. In contrast a series of cross-sectional surveys conducted in Glasgow in 1990-1996 showed some albeit unsteady results indicating that HR programs might have decreased the spread of hepatitis C60. The ambiguity of results of evaluations of effectiveness of harm reduction programs may arise from the complexity of the question of interest, however it may also result from the variability of what HRP means in every particular context of each study. For countries with long history of communitarian values it seems quite important that HR programs serve the needs of not only drug users, but they are for the benefit of on non-users. Research into what works in HIV prevention among IDUs or what makes an effective HRP therefore is needed to complement the debate about the policies of combating the AIDS epidemic. Economic assessment of harm reduction The main elements of the cost of HRPs are syringe exchange, outreach work, distribution of prevention (condoms, bleach kits, clean syringes) and education materials (literature and instructions on HIV prevention and safer injection techniques), HIV counseling and testing. The cost effectiveness analysis of HRPs is a very important issue that contributes to future HRPs advocacy. Therefore, many studies have concentrated on the economic analysis of Harm Reduction Programs interventions. All studies have found that HRPs are rather cost effective when compared with other lifesaving interventions. For example, the cost-effectiveness analysis of approved syringe exchange programs and estimation of the cost-saving potential of these programs was conducted in New York State. Based on an estimated 87 HIV infection averted across the seven programs and total program cost of $1,82 57M. T. Schechter et al., Do needle exchange programmes increase the spread of HIV among injection drug users? An investigation of the Vancouver outbreak. AIDS 13, F45-F51 (1999). 58F. I. Bastos, S. A. Strathdee, Evaluating effectiveness of syringe exchange programmes: Current issues and future prospects. Soc. Sci. Medicine 51, 1771-1782 (2000). 59Pollack HA, Cost-effectiveness of harm reduction in preventing hepatitis C among injection drug users. Med Decis Making. 2001 NEP-Oct; 21(5):357-67. 60Taylor A, Goldberg D, Hutchinson S et al. Prevalence of hepatitis C virus infection among injecting drug users in Glasgow 1990-1996: are current harm reduction strategies working? J Infect. 2000 Mar;40(2):176-83. 27 million a cost effectiveness ratio made up at $20,947 per HIV infection averted61. Thus the research arrived to a conclusion that syringe exchange is a cost-effective and even cost-saving intervention when compared to the costs of AIDS treatment strategy for reducing HIV transmission. As the program coverage increases, the marginal cost per HIV infection averted also increases. Although viewed by many experts as arbitrary, UNAIDS suggests that 60% coverage is required in order to stop HIV transmission among IDUs. The study conducted by Holtgar et al.62 showed that at very high coverage (more than 88%), the marginal cost-effectiveness of increased program coverage becomes less favorable but anyway even at coverage level more than 88%, such funding would save society money, i.e. the costs of the program will still be lower then expenditures needed to provide treatment to incremental numbers of infected people. The outcomes of the study conducted in Edmonton based on street-involved IDUs interviews and HIV saliva tests among HRP participants were used in a cost-effectiveness model63. The research estimated that the HRP prevented 20.3 new infections. The cost-effectiveness estimate was $9,537 per HIV infection delayed for one year. Even if HRP activity were to continue for 17 years, which is expected remaining survival time for a person with AIDS the value of $9,537 would increase to $116,024, which is still less than the value used for a case of HIV/AIDS ($150,000). The study aimed to determine whether providing NEP would cost less than the health care consequences of not having such a program was conducted in Hamelton64. The results predicted that NEP would prevent 24 cases of HIV over 5 years, thereby providing cost savings of $1,3 million. It equals to a ratio of cost savings to costs of 4:1. The study on the effectiveness of 16 HRPs was conducted in Russia. The results of the research have shown that total average costs were $40,662 per HRP per year. The financial unit costs were $23,4 per client served and $0,36 per syringe exchanged. A cost-effectiveness ratio of $564 per HIV infection delayed for one year was calculated based on an average program cost and estimated 72,1 HIV infection averted per project65. It might seem relatively costly but in comparison with medical treatment of HIV ($3000-$7000 per year) it proves considerable saving. 61Franklin N. Laufer. Cost-Effectiveness of Syringe Exchange as an HIV Prevention Strategy. JAIDS 2001, 28: 273-78 62D.R.Holtgrave, S.D.Pinkerton, T.S. Jones et al. Cost and cost-effectiveness of increasing access to sterile syringes and needles as an HIV prevention intervention in the United States. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology. 1998, 18 (Suppl 1): 133-38 63P.Jacobs, P.Calder, M.Taylor et al. Cost effectiveness of streetworks' needle exchange program of Edmonton. Canadian journal of public health. May-June 1999, Vol.90, 3: 168-71 64M. Gold, A. Gafni, P. Nelligan et.al., Needle exchange programs: an economic evaluation of a local experience. Canadian Medical Association Journal. 1997; 157: 255-62 65A.V. Bobrik. Cost and cost-effectiveness of HIV prevention among drug users in Russia. Russian Family Doctor, 2002,Vol. 6; 4: 30-36 28 In summary it can be stated that the studies conducted in various places by different groups of authors, including those undertaken in Russia have arrived to anonymous conclusion that Harm Reduction approach is rather cost effective, and that the savings related to aversion of HIV cases are greater then the costs of running the programs. Although the studies reviewed in this section have arrived to different costs of preventing one case of HIV and to different cost- benefit ratios, while costs are relatively easy to estimate the major methodological problem is obviously related to different estimates of effectiveness of programs in varying settings. The degree of effectiveness and even the cost parameter however would be related to the best practice issue, as good management can reduce expenses related to running the program, and coverage and ability to reach the most marginalized parts of IDU community will be incremental to raising the effectiveness. The concept of best practice Before exploring what can be considered best practice in harm reduction it is important to understand the meaning of the concept more generally. Thus United Nations and the international community defined best practice at large as successful initiatives which: - Have a demonstrable and tangible impact on improving people's quality of life; - Best practices are the result of effective partnerships between the public, private and civic sectors of society; - Are socially, culturally, economically and environmentally sustainable. Best Practices are promoted by the United Nations and the international community as means of: - Improving public policy based on what works; - Raising awareness of decision-makers at all levels and of the public about the potential solutions to common social, economic and environmental problems; - Sharing and transferring knowledge, expertise and experience through networking and peer-to- peer learning66. Albeit being very general, the UN definition allows us to concentrate on three issues when approaching the purpose of the current study, which are effectiveness, sustainability and multi- sectorial involvement (mainly as means of achieving effectiveness, but also can be a target in itself). The later directly relates to such issues as involvement of civil society, good governance and coordination, as well as to Three Ones principles promoted by UN agencies. 66http://www.bestpractices.org/bp2004/guide_en.pdf 29 Best practice in the field of HIV/AIDS prevention In recognition of the vital role of HIV prevention, the Joint United Nations Program on HIV/AIDS (UNAIDS) and Family Health International (FHI) have published the Family Health International/UNAIDS Best Practices in HIV/AIDS Prevention Collection67. Encompassing a broad body of knowledge and expertise, this book is centered on HIV/AIDS prevention in the non- industrialized world. This collection is based on the six years of work performed by FHI's international and local partners in the world's largest single international HIV prevention initiative to date: the AIDS Control and Prevention (AIDSCAP) Project. It offers a substantial number of models that may be replicated around the world. Twenty initiatives have been chosen from more than 800 HIV/AIDS and sexually transmitted infection (STI) prevention programs in 50 countries that are considered to be of global importance for dissemination through a case study approach. These initiatives covered different population groups, most vulnerable vis-à-vis HIV infection and were devoted to the successive implementation of different preventive strategies such as education, mass media campaigns, monitoring trends in HIV-risk behaviors, services for women ay high risk, HIV counseling and testing, national HIV/AIDS programs, etc. The selected initiatives were considered to be the most innovative, effective and comprehensive approaches to HIV prevention that have been implemented in non-industrialized countries to date. Bridging the Gap Conference that took place in 2001 with support of San Francisco Department of Public Health aimed to improve standards of care, develop best practice principles for integrating harm reduction approaches into traditional substance abuse services, and increase the accessibility of quality services to people in need of alcohol and drug treatment. One of the conclusions was that harm reduction offers the greatest hope to expand the availability of substance abuse services to people who have not benefited from traditional abstinence-based treatment models68. The issue of best practice in harm reduction in the community and in prisons in Russia was addressed by Dave Burrows in 200169. The following factors that determine best practice of harm reduction in the RF were highlighted: - attitude of HR programs themselves towards strategies of HR (willingness to do something about HIV problem, activism, enthusiasm, etc.) 67http://www2.unescobkk.org/hivaids/FullTextDB/aspUploadFiles/FHIUNAIDSBestPracticesredux.pdf 68 Marlatt GA, Blume AW, Parks GA. Integrating harm reduction therapy and traditional substance abuse treatment. J Psychoactive Drugs. 2001 Jan-Mar;33(1):13-21. 69Burrows D. A Best Practice Model of Harm Reduction in the Community and in Prisons in Russian Federation. IBRD/The World Bank, Washington, DC, 2001 30 - support of local authorities and institutions - international funding and technical support - conduction of rapid situation assessments with the advocacy purposes - increasing secondary exchange and peer education In addition, the following weaknesses of HR programs were highlighted: - low coverage level of IDUs by HR services - absence of sufficient funds for HR programs to do the work required from them - low level of management partly caused by the general underdevelopment of NGO sector in RF - reliance on doctors as HR programs workers rather then on IDUs and ex-IDUs. - lack of democratic structure in HR programs (autocracy and rigid hierarchy) - patchy quality of services, depending on individual managers of HR programs - confusion about the aims of the HR program in some cases and abstinence ideology sometimes hindering HIV prevention objectives - little interest in investing time and money in educating volunteers and outreach workers - lack of understanding about the need for long-term responses to HIV epidemic - poor understanding/experience of monitoring and evaluation - poor networking (cooperation of HR program managers with colleagues in other cities) However it needs to be emphasized that the analysis was undertaken in 2000-2001, when generally harm reduction programs in Russia were very young, with the majority of them operating for a period of less than two years. Little attempts have been made since to understand the evolution of harm reduction strategies in Russia, to further explore successes and failures, and factors which determine these. Thus, results of the current project supported by the World Bank may contribute to the issue of best practice in HIV prevention in the Russia context. CONCEPTUAL FRAMEWORK As we have demonstrated in our literature review best practice is a slippery concept. Often when characteristics of what best practice in harm reduction are listed purposes are mixed with means of achieving these. For the purposes of this study we shall view best practice as combination of achieved characteristics of the HR program and activities, procedures, rules and interventions which can at one hand help reach good results: process, input and outcome indicators. Since harm reduction programs depend on the context they are placed within, it is important to split context into factors of different degree of stability ranging from those fully determined by the best practices of the HR project to totally non-modifiable. 31 Results of the work as we mentioned above can be measured by process (well trained and motivated staff which performs all the duties well, an evaluation and responsive adaptation system is in place, resources are spend wisely, e.g. there is large coverage in terms of total numbers and proportion of IDUs served and all this works within a small budget), by effects on behavior or surrogate outcomes (that includes sharing syringes and sexual practices) and outcomes (incidence of HIV, blood born hepatitis, STI and overdoses). Finally overall impact can be defined (but hardly measured) in terms of degree of keeping HIV and other infections, as well as drug use under control. The very best practice harm reduction sites are hence those which achieve good results in a situation of hostile environment and hard-to-raise funding, improving the modifiable context characteristics to less inimical in the course of their functioning, while reaching the objective of being effective, ensuring multi-sectorial collaboration and sustainability (Figure 2). Figure 2. Conceptual framework. HR program: Processes (formal & informal), actions and achieved characteristics Modifiable Practices (best) Results: outcomes, Context process Non- modifiable 32 RESULTS: SEMI-STRUCTURED INTERVIEWS The best practice review below is based on content-analysis of 57 interviews with HR project staff, clients and other key informants in six regions of the Russian Federation (Kazan, Pskov, Tver, Vologda, Balakovo, Voronezh). Additionally seven interviews conducted with the federal level decision-makers: with representatives of the Research Institute on Drug Addiction of the Russian Ministry of Health and Social Development, Federal Service for Surveillance of the Protection of Consumer Rights and Human well ­ being were also includes, as well as seven interviews conducted while piloting the questionnaire in Mytichi (Moscow region) with staff member of local HR project and head of narcological dispensary. The chapter has the sections devoted to the conceptualization of harm reduction, including substitution therapy and its evolution, as well as sections which look at achievements and failures, as well as success factors and barriers for the civilian and prison HR programs. Understanding of harm reduction concept. Although generally there was common understanding of HR concept, when going into greater level of details various peculiarities in thinking about HR among the respondents became quite clear. The most widespread was "public health oriented" vision of HR which can be split depending on the major focus into two: one aiming at IDUs behavior change in order to prevent HIV transmission inside drug users community (29 out of 67 respondents) and second aiming at keeping HIV epidemic inside this community and not enabling spill out to other risk groups or general population (25 out of 67). As stated by one of the managers of outreach team obviously taking the second broader vision of HR: "It means that HR works for the benefit of the total population but unfortunately the general population often does not understand this". The majority of the respondents from HR programs (12 out of 18) said that main goal of harm reduction is to control HIV epidemic among IDUs and to decrease HIV, hepatitis and STI rates. They also mentioned that: "Although HR started as HIV prevention activity, or even narrower -as needle exchange, this should be seen as only the first step of general "health-oriented strategy". There were other perceptions about HR which went beyond the traditional "public health orientation". Some of HR project activists believed that the role of HR as "the media between government run services and the "Street", which allows the "Street" to get information and care" was more important. It was an unexpected finding for us that the majority of outreach workers (10 33 out of 18) consider HR concept mainly as "an overall health oriented" strategy not emphasizing any specific services. During one of the interviews the words "HIV", "syringe" appeared only after twenty five minutes of the conversation. The respondents highlighted that these programs teach IDUs appropriate attitude to their health in general. To prove this position they indicate that among clients of HR projects there are more people applying for addiction treatment. This point of view was also supported by all four respondents from narcological dispensaries. The interesting philosophic understanding of HR strategy was given by the representative of one of the first Russian HR projects. It was pointed out that HR is an example of humanitarian mission, giving a righteous message to people. It is human protection movement as it is an example of the approach oriented at people ­ IDUs, who are totally unprotected and humbled in the RF. To support this position the representatives of all HR programs visited marked at least to some extend, that currently the government makes all its decisions based on two polar considerations: first precluding that all people are healthy and not taking drugs and second believing that any person who had anything to do with drugs is a potential or present criminal. There seems to be no space for official views that could be placed in between the two. However such moderate position is crucial for provision of relevant support to various groups. Insuring financial and political support is impossible without shifting away from such radicalism in the perceptions of population health among decision makers. Idea raised by at least half respondents was very well formulated by one of them: "HR must be incorporated into the overall prevention structure, it should not be considered as an independent component, it must be coherent part of the general HIV prevention strategy". There was anonymous agreement that unfortunately currently HR remains an isolated and "foreign" service, which can hardly be sustained without external donor support. Several respondents who were among first ones to start needle exchange programs in Russia, HR should have clarified and specifically outlined objectives while a lot of the broad goals should be addressed by governmental institutions. These broad objectives include improvement of access to medical services, social support and legal aid. All these services should be official and properly organized and not be delivered as exceptions based on personal links and patchy donor funding. In this situation HR projects should specifically concentrate on the maintaining and improving good contacts with target groups. Most HR managers (10 out of 18) agreed that the design of HR projects in Russia currently is imperfect, and that they should be primarily (if not totally) considered as "outreach projects" while syringe exchange should be organized and financed by governmental institutions. They believed 34 that HR projects should widely involve IDUs society and improve links with existing services. There is a need for proper management of outreach workers by qualified management staff, adequate resources for mobile services, transport, sufficient premises for psychological support and communication; however most of medical care, such as tests or safe injections should be the function of the existing medical system. Besides specific HR activities program teams in parallel often claim to provide primary prevention of drug use. Psychologists, teachers, medical professionals, and centers for prevention in all six regions received at least some training by HR teams on HIV and drug abuse primary prevention. It was mentioned that narcological services currently oriented only on addiction treatment and can't provide the comprehensive set of the services required for drug users. According to the opinion of one of the managers of HR program, narcological services should be more involved in HR strategy and work in tight collaboration with HR projects, accept patients as individuals and address their complex health problems: "Classic" HR projects as they were set at the onset of the movement lacked active involvement of state medical structures such as narcological and dermato-venerological services and therefore were less successful". At the same time three out of four representatives of narcological service expressed an opinion that "HR program is the only way of dialog with these closed groups. There is no other way to reach them and it works". Representatives of narcological services were convinced that patients coming for treatment through HR projects are more motivated and more oriented on success. They also consider HR as strategy that helps to put IDUs into the social frameworks: "While using certain limited doses of drugs IDUs are still able to perform certain social functions. They are able to control themselves and do not commit any crimes. They are also able to control their sexual life. Harm reduction contributes to quitting or minimising use of drugs. It means that it also motivates IDUs toward rehabilitation". Three out of six leaders of HR projects expressed a view pointing at the need for wider approaches to HR: "Unfortunately, we have to accept that HR programs in Russia are mostly implementing only one element of the whole comprehensive strategy ­syringe exchange. And even that not to full extent. At the same time one of the important elements are something like social centers for IDUs where they can take meal, cloths, can communicate, get social and psychological support and lawyer assistance. The main goal of HR strategy should be achieved through respect of drug-user as personality". 35 The project leaders also were convinced that civil HR projects should be implemented in combination with other social projects such as human rights protection projects, as well as prison projects: "Civil HR programs should be oriented at helping solve conflicts with law-enforcement bodies which will always exist in spite of any success of HR activities. The reason for integration of civil and prison HR projects is that the population of IDUs is constantly exchanging and mix civil and prison subpopulations of IDUs, e.g. about 25% annually according to some estimates and the outbreaks of HIV epidemic in prisons are rather frequent". The views of drug users were relatively unanimous mainly emphasizing the following three aspects of HR project: "the place where you can come without fear of prosecution and get some medical support", "can exchange the needles" and "get information". In conclusion it should be mentioned that every group of respondents had an emphasis on the type of service they were providing, e.g. the narcologists belived that HR should make greater efforts to refer users to addiction treatment. The understanding of the role of HR in protecting those not using drugs from the HIV epidemic was not universal among respondents. The concept of ideal project often diverged from the realities of the HR in Russia today, with most important difference concerning increased emphasis on needle-exchange through the project sites, while believing that the HR should, mainly be the "linking and supporting activity", while the existing, state-run healthcare system should do all the services once the links are there. 36 Attitudes toward substitution therapy. Issue of substitution therapy was not raised by majority of the respondents. When probed most interviewees said the most obvious issue with it is absolute prohibition as methadone is included into the list of hard drugs. Attitude toward substitution therapy was rather controversial among different groups. Some of the heads of the HR projects and HR staff (18 out of 24) consider substitution therapy to be a necessary element of HR strategy: "there is no future of HR without substitution therapy". Approximately half of the respondents not working in the field of HR stated that it is an unacceptable strategy. Thus, all three representative of State Drug Control Service interviewed consider methadone substitution therapy absolutely unacceptable: "We unambiguously against methadone substitution therapy and will never change our attitude toward it. Our point of view is also supported by governmental level". A somewhat less radical opposition to HR largely seems to be related to their point of view that HR can be implemented instead of methadone substitution therapy. In contrast, some representatives of narcological services stated that substitution therapy could be very helpful in specific cases: "Firstly, methadone substitution therapy is very helpful for IDUs with a long history of drug addiction, who are not able to quit drug use. Secondly, in cases when IDUs are hospitalized for any other not directly related to drug use reasons, they are exposed to abstinence syndrome and doctors are not able to help them at all and methadone could be very helpful in this case". In one of the regions a respondent from HR program stated that they have succeeded in conviction of narcological dispensary to provide Tramadol as substitution therapy agent for IDUs. The leader of the program considered this a big victory and great advantage in their region. However to the best of our knowledge this is the only case of semi-legal use of substitution therapy after it was banned in the 1970s in USSR. In the prison sector the project staff quite explicitly pointed out that the substitution therapy is beyond HIV prevention boundaries. They were concerned about implications for meeting overall sanitarian and epidemiological regulations in one hand and for provision of reconciliation and decreasing aggressiveness on the other hand. In summary the attitudes to substitution therapy even among those directly involved in HIV prevention ranged from cautious to negative, besides some activists and outreach staff. It is not clear why this concept seems so hard to promote, however it needs to be understood that even discussions about methadone can be considered illegal "popularization of drug abuse" according to current legislation. 37 HR: evolution of thinking. Prior to discussion of achievements and failures of HR it's important to track the evolution of views of different respondents toward HR. While many of the respondents had positive attitude toward harm reduction from the very beginning, attitudes of others have been significantly reconsidered from the negative at the beginning to positive. At the beginning HR projects in all regions experienced opposition from law enforcement bodies, administration, medical institutes, media and general population. In majority of cases such attitudes have changed at least to some extend. In several regions HR activities went as far as being included into the agenda of the regional multi-sectoral Commissions on HIV/AIDS and Regional Target Programs "Anti-HIV/AIDS". At the federal level the views seem to have changed from negative to "idealistic", and then to "realistic" while issues of implementation became clearer. Indeed at least five decision makers interviewed stated that while being bluntly opposed to HR when they first heard of the concept, they then thought that it is the easy solution to HIV epidemic in Russia. As time went on they realized however that only the best practice programs will produce some impact, while most of the projects will be far from ideal and no single approach will completely solve the issue. Reconsideration of views toward HR by former IDUs who now work as outreach is rather interesting. One of them has indicated that first he was stunned by the idea of syringe exchange: "this is like giving disposable caps to alcoholics". Another former IDU shared her evolution of understanding HR from "HR promotes use of drugs by creating comfortable conditions for drug users" to "if a person want to inject he will do that anyway". This reconsideration took place after quitting. Another outreach just pointed that HR project "attracts the worst users, that do injecting in such awful ways, with very bad equipment!" The outreach said that the contact with HR clients helped him realize that "such marginalized IDUs need help and information ever more". One of the interviewed outreach workers whose acquaintance with HR began when she became client of this program told that in the beginning she had only consumer interest toward HR, "just possibility to take something for free". Then the outreach worker that contacted her told her about possibilities of ARV treatment and that it's possible to live with HIV. ARV access seems to have produced stunning effect on many people who need HR services, and some of them actually began to work as outreaches. The same person described changes of attitude of current HR clients "In the beginning people met me and simultaneously got into my bag and the only interest they had was just "give us syringes". Currently I still use my bag but when I came people first of all want to talk with me, want me to answer their questions about HIV and related issues". 38 Changing of attitude toward HR from the side of HR clients is also interesting: "Many myths among drug use were discredited", "People are changing their attitude. Usually people see IDU as dirt, outcast, that should be isolated from "normal" people. And it leads to ceasing contacts with society. And here us [IDUs] see that we are treated differently and that environment becoming friendlier". Changing attitude of HR clients was also described by one of the medical consultants of HR program: "People started to change... I was very surprised to learn that stationary NEP is visited by IDUs from other district. The money they spend on transportation is enough to buy syringes. But they came to have some contacts; they wanted to see that they are accepted. It was important for them to see that they are not outlaws." Some of the respondents mentioned that that HR- is the only one "air-hole" for IDU. As was mentioned by outreach workers and manager of outreach in one of the regions: "the quality of life and social status of clients is improving" and they start to pay in-kind to society. Considering other members of society "IDUs start to understand what to do with needles, that you should not throw away syringe as somebody could prick himself with needle". This view is supported by the narcologist who said: "several years ago it was impossible to look into mail-boxes or into energy meter because everything was full with used syringes. Under snow there was also great number of syringes". One of the current leaders of HR program in the end of 1990 when HR programs have been introduced in Russia had negative attitude toward these programs and even more was shocked by the idea of HR. "While working as psychologist implementing rehabilitation approach for IDUs, I considered the idea of syringe exchange as shock, something incredible. I was sure that it's not the approach we needed. I worked as family psychologist and provided counseling for families who faced the problem of drug use. I saw adolescence that were best pupils in school, practiced music, sport, etc, and suddenly turned into distracted out of any interest people during very short time period because of drug use. It was awful. And idea of HR sounded just blasphemous. But I have been sensible person and decided first of all to investigate this issue especially taken into account that this approach was used in developed countries... Thus, on one hand there was high degree of distrust from my side, on the other hand I saw well-known people from developed countries promoting the idea of HR, I saw competence of that people and realized that they sincerely want to do something effective in order to fight HIV epidemic. I have experienced internal struggle inside myself. This struggle came to the end when we conducted rapid situation assessment showing that 98% of 232 interviewed IDUs reported risky behaviors permitting easy transmission of HIV. At that 39 moment I realized that HR programs have to be implemented in order to change IDUs behavior modes and prevent further spread of HIV". In summary there seems that almost no one but the drug users were ready to accept the HR idea from the very start. However many turned into sometimes activists of the approach, sometime blindly worshiping HR ideas. Only some respondents have however come to understanding of implementation issues and to realization that best practice is needed in order to make HR program highly effective. 40 Advantages and disadvantages of harm reduction: clients prospective Although in the original questionnaire there were questions concerning "advantages" and "achievements" separately while analyzing the transcripts it became clear that in the majority of cases the two terms were confused with each other. The advantages were more clearly identified by the clients as direct beneficiaries of HR programs. At the same time all of the advantages mentioned by IDUs were mentioned as main achievements by those involved in provision of HR services, specially highlighting the advantage of provision of IDUs with blood tests and counseling. Hence it was decided to limit this section only with advantages as viewed by the group of HR clients. Majority of interviewed IDUs marked the following advantages of HR: - access to medical services - referrals to narcological dispensaries - access to anonymous blood tests and treatment. Confidentiality was crucial for IDUs here: "why would I not trust HRP, if they do not even write down my name" - possibility to get free syringes that prevent use of non-sterile equipment - reduction of chances of HIV transmission - unprejudiced attitude of HR staff toward IDUs Commercial sex workers (CSW) besides the above mentioned advantages also mentioned the possibility to get consultations, condoms, useful information materials. While ranking and prioritizing advantages of HR projects majority of the HR clients (IDUs. CSWs) spoke about availability of information materials, social support and access to medical services, as well as about the supply of syringes and condoms. Among disadvantages IDUs in some regions listed bad quality of materials distributed and fear of prosecution by the police. Indeed in some regions precedents of purchasing leaking syringes was mentioned by IDUs, so they preferred to inject with a used ones not to loose the narcotic solution. CSW had sometimes complained on poor quality of condoms, which tear, slide, decrease satisfaction for their clients or cause allergies. Besides poor quality of materials, the issue of relevance also was raised by HR users, for instance in some cases the change of the drug used required different size of syringe (e.g. home made opiates require 5 ml. syringe, whereas in case of heroin use "diabetes" ones are needed) In the prison sector among disadvantages inmates mentioned absence of lubricants and syringes and lack of sufficiently detailed information on HIV issues. 41 Achievements and failures of harm reduction programs Description of achievements and failures of HR programs was divided into sections devoted to different aspects: to outcome measures such as epidemiological and behavioral measures, to access issues, attitudes of decision-makers and law enforcement bodies, finally to integration into the healthcare system. Epidemiological and behavioral change. This section is devoted to achievements relates to epidemiological and behavioral issues. All the respondents believed that HR helped keeping HIV/AIDS epidemic under control. Part of the respondents stated that the achievement was not the reduction in numbers of new HIV cases but holding it stable. Other highlighted reduction of HIV incidence among IDUs as a main achievement. "If in 2001 there were more than 300 new cases of HIV among IDUs, then in 2005 there were 112 new HIV cases and only 29 of them among IDUs, all other HIV cases were detected in the general population." Both outreaches and HR managers believed that "Rates of HIV transmission through blood went down, currently there is increase of sexual way of HIV transmission", "If before 70-80% of HIV were among IDUs, now its proportion is only 20%". An outreach was convinced that sexual way of HIV transmission is not necessarily related to commercial sex work, "it could be explained by the fact that many people faced with HIV do not want to bear with this diagnosis and try to behave as usual" Although there is lack of solid epidemiological data on the effectiveness of HR programs in the cities included into this survey, the respondents were providing routine statistics on HIV incidence in comparable cities of the same territory with and without HR programs to indicate their achievements. These data suggested lower incidence rates in the cities with HR activities, although interpretation of epidemiological data needs to be cautious. The category of the outcome related achievements also includes increase of safer injecting practices. A head of one of the HR program said "If in 2000 unsafe practice were prevalent in 72% of users (sharing equipment), presently it is 18% only", this claim is repeated by other respondents: "last time we had a survey was in 2005 and 87% of clients practiced safe injection behaviors". Others also claimed that project has significantly influenced behavior of drug users. "I know that people that had a package of used syringes under the bed do not have it anymore", "In the past 5 persons were injecting with one syringe, but not anymore". At the same time at least twelve respondents explicitly said that it's difficult to directly link HR activities with change of HIV incidence and in behavior, as many factors could have had influence. It is not known how the behaviors changed in places where no HR programs were set. 42 Access to target groups Majority of respondents representing HR staff (19 out of 24), epidemiological surveillance service (all 4), narcological dispensaries (all 4) mentioned as one of the main achievements ability of HR programs to get access and work with the most closed groups of population. Finding ways to provide social support to these groups through the programs deemed very important by them. Links to closed societies proved to be one of the main factors that motivate staff of the official medical system to collaborate with HR programs. Main achievement marked by three out of six HR projects leader is trust of both IDUs and general population which probably even more important. However even in regions where the opposition to HR is minor the head of the project mentioned: "Still there is small portion of population that deny idea of HR while majority of general population, including authorities, medical specialists, and police began to understand purposes of HR and openly speak about it". One of the program directors put it as: "Our program was recognized by target groups: IDU, CSWs and prisoners" claiming that "all target groups know about the work of the program". For the majority of the respondents the notion that no other except them was able to gain trust of people in marginalized groups was very important: "The program can reach those circles where they [authority] can not reach... and we can see how the situation in the city is changing". The same achievement was also marked by narcological dispensary and law enforcement representatives, medical specialists and many other respondents. The representative of State Drug Control service in one of regions expressed his rather positive attitude toward HR in comparison to other respondents of the service, emphasizing high coverage levels of IDUs by HR programs as a main achievement. Cessation of drug use among broader groups of IDUs thorough increased information coverage was emphasized as one of the major achievements by representatives of narcological services and AIDS center in several regions. "If person who became harm reduction client quit use of drugs after certain period of time and able to maintain the remission during long period of time, than it's a great advantage of HR project. Even if there will be 2-3 such people per year, it still would be a great achievement". In a number of regions a concerns regarding targeting the risk- behavior groups becoming more difficult at present was expressed. The important factor contributing to issues of maintaining contact is constantly changing "narcotics scene" due to the drug regulation legislation change combined with other economic and social reasons. The "nacro-scene" as it is called by respondents 43 became "closed" in Russia, meaning that stricter legislation led to change of distribution routes, namely by the increased number of "running sellers". The drugs are supplied via "personal contacts" and "home deliveries" rather than through "street or apartment based drug dealing" In order to reach IDUs now the increased resources and additional training is needed, while the outreach management is much harder to organize. This makes the financing issue even more important as the required increase of mobile outreach workers is associated with greater expenditures. In some sites where drug control services have achieved greater success, the haunts previously visited by outreach workers are no longer in place (or not known to HR staff). Now the gathering places do not exist for over 2-3 days and then change, which requires greater flexibility and closer links between HR projects outreaches in order to adjust to these conditions. While "open narcoscene" is easier to work in, as the target groups are easier to reach, "closed narcoscene" requires increased secondary exchange, as well as other creative organizational solutions. Attitude of politicians and decision makers toward harm reduction In the few cities where political support was hard-gained the respondents place it ahead of other achievements and consider it to result from their advocacy work, which also has been mentioned as program major success. In the other sites where political support and governmental co-funding were in place almost from the very beginning of the projects, HR staff are satisfied with the sustainability of this support despite some replacements in regional authorities. They consider the sustainability of the political environment as their success, although they agree that it requires constant thoughtful advocacy work conducted by HR leadership: "Advocacy and promotion in HR requires professionalism not less than any other component". It's interesting to mention an unofficially expressed point of view that HR projects helped authorities for the first time to recognize the real situation with IDUs, e.g. the numbers of IDUs, the practices, etc. Representatives of another region recognized as successful mentioned unification of efforts of different bodies (medical specialists, authorities, etc.) in addressing HIV prevention among their major achievements. "Such global task as fighting HIV can be solved only with united resources of different agencies. Even though HR plays one of the leading roles in this task, without support of other structures HR will not work effectively" Such achievement as low level of HIV among IDUs can very much contribute toward support of HR reduction on the governmental level. Thus, the vice - governor in one of the regions 44 expressed negative attitude toward HR on the personal level, but said that "based on the low rates of HIV in his region he agrees that HR is needed to control HIV epidemic that is why as representative of the executive government he supports HR". The respondent of one of the regions mentioned development of inter-sectoral cooperation. "There was a time period when police took away HR client cards and clean syringes, dirtied the syringes. There was opposition from medical staff because of misunderstanding of HR purposes. The most prevalent opinion was that HR programs distributes syringes and teach people how inject drugs. There are no such cases anymore. Six years have past and situation in 2001, 2002, and 2203 is very much differed from 2006. Inter-sectoral cooperation has been established". Several respondents expressed the view that intersectional collaboration had external effects on improved efficiency of other activities outside HR, e.g. better collaboration between law enforcement bodies and healthcare system helped improve access to narcotic analgesics for patients suffering from chronic pain. Relations with law enforcement structures Good relations with government and advocacy work in many regions led to better interactions with law enforcement bodies: "there is better collaboration now with both state institutions and police, and more recently drug control agencies are crucial in this dialog". Some claimed that such relations has only started (implying there is a significant room for improvement), whereas other respondents claimed that "the program is in tight collaboration with police and other law enforcement structures". The later view is supported by others, explaining nature of this collaboration: "We have wrote handbooks for police, conducted trainings and seminars with police, for those of them who are working with IDUs" The same respondent wording signified that it was very important to work with low-level law enforcement officers, who do the grass level work on the street, while they can do more for harm reduction programs and their clients, for instance provide security for harm reduction personnel. It was often mentioned that relations with private policemen and junior officers are not easy. In spite of a number of trainings for law enforcement officers they still in some regions use any opportunity to prosecute outreach workers and HR clients, extracting money or developing career. It is the grim reality that sometimes HR staff is not experienced enough to stand this pressure. One of the examples provided by outreach worker that often police only formally supports HR through letters and high level informal discussions, but "when a policemen knowing who I am asks me where I have just been, in order to get there and establish haunts..... how does it look like?" On the other hand there are high turnover rates of police staff and additional training of new- comers constantly remains a high priority. It was emphasized by HR representatives of one of the 45 regions that even though common language with Ministry of Internal Affairs and Federal Penitentiary Service was always found, it was rather hard to push these structures to accept HR strategy. Another important key player in the filed of HR is State Drug Control Service. Positive changes in collaboration with it were also considered by many respondents as a big achievement in the work of HR. While the agency was established just in 2003 relations with it is still rather complicated in majority of cities. As it was marked by one of the HR regional leader "Collaboration with State Drug Control Service is always very hard and complicated. I will never believe those who tell that they have no problems with them and that everything is fine". Interestingly in one of the regions representative of State Drug Control Service expressed absolutely positive attitude toward HR "I think that this is needed direction for our population...... The most positive moment of HR is provision of IDUs with preventive and educational measures" and even said that HR programs have to be expanded all over the region. But the same respondent recognized that collaboration of HR with authorities is not very easy "Administration does not provide any financial support but during meetings raise this issue and ask why you did not do this and that, and never ask about the difficulties and never express desire to help". Very prevalent point of view toward State Drug Control Service as well as toward police is that at least they stopped hamper work of HR programs "these structures just have not to hamper. They could be indifferent or interested in HR programs work but more importantly they just must not be our enemies". Often position of State Dug Control Service and police was described by following statement of the HR program head: "if you work and nothing happens - you are simply ignored, but ones something goes wrong then it is your fault". In some regions State Dug Control Service in generally supports HR but still bewares public announcements of their position. Just in one region they made positive statements thought mass media. In contrast it should be mentioned that in some regions in order to avoid the need to express their negative attitude participation in this study was refused. Also in some cases there is an impression that sometimes representatives of Drug Control Service implicitly misinterpret the meaning of HR, considering it as preventive programs with education and information components for the youth or even for general population. It could serve as alarming sign, as in the situation of such misunderstanding, while claiming to support the program, they might actually try to prohibit expanding of typical set of HR activities. Links across health care sector 46 Building links across health care system are considered by the majority of respondents as one of the most important factors of success. Some respondents mentioned that as a result of the advocacy efforts the project has started to make inroads into medical establishment: "even most conservative medical doctors started to doubt their original views". Many narcologists are claimed to have changed their attitude as well. Even though there is still rather large numbers of medical specialist and decision makers that hamper expansion and sustainability of HR. It needs to be noticed that degree of successful cooperation with narcological services is rather different from region to region, with some achieving good level of understanding and effective cooperation. Representative of narcological service mentioned that one of the achievements of HR is the reduction of IDUs number: "for the first time there was reduction of IDUs numbers in our region. Even if we can not say that it totally relates to HR achievement but anyway at least it didn't stimulate increase of their numbers". Another respondent mentioned achievement of HR programs in helping narcological services to reach more patients. Controversially the infection disease doctor from the same region among failures of HR mentioned increase of IDUs population and detection of several new HIV cases among IDUs. Narcologist indicate that criteria of success of HR projects can be added with "narcological" indicators such as the numbers applying for treatment and rehabilitation, the duration of remission. The experience of some respondents was that health care has mostly negative influence on harm reduction programs. It was expressed by one respondent as "direct opposition of governmental institutions, e.g. `AIDS Center" with explanation that from the health care it was "purely negative, bad attitude toward the program". Another respondent mentioned "negative influence of health care authorities... letters, slander, accusations of providing methadone". Representatives of one of the HR project said that he believed reason for misunderstanding often relates to "low competency level of health care administration", "low professional level of health care workers" and that "narcology dispensary is useless". Outreach manager said, "you need to understand that unfortunately time of free care has passed and now it is business" explaining that narcologist interested only in drug users that are capable of paying for addition treatment. This animosity between harm reduction project and health care greatly hampers harm reduction project activities. A head of the HR program stated, that in relation with health care "frequently there is a feeling that we are begging for something from them". This problem also arises while arranging provision of gynecological services for sex-workers. HR leaders even in the 47 advanced regions are complaining that links with this sector is mostly based on personal relations rather than on routine, systemically organized procedures. The support for HR is not universal even in the AIDS centers. In one of the regions the HR project has requested the AIDS center a letter of support for the tender commission of National HIV prevention project leaders were requested. The AIDS center has replied by a letter stating that participation in these projects is an unlawful activity, which concerns promotion of illegal drug use among general population through creating impression of appeasement, lack of responsibility and encouragement of use. Some respondents marked that currently majority of medical specialists have stigmatized and preconceived attitude toward IDUs. Very few doctors would like to work with this group. But it was claimed in the interviews that those doctors who work in close cooperation with HR programs have changed their attitude toward IDUs. Representative of HR program in one of the regions told that there was a time of pressure from narcology and psychiatry services against HR. They claimed that IDUs related issues are their business and HR team is not professional and the whole idea of HR leads to expansion of drug use and drug addiction. But currently many change their views, and one of the syringe exchange points visited in the frames of this study is located on the territory of narcology center and a chief doctor works as a HR consultant. All these positive changes contribute to IDUs feelings that HR program is not bare needle exchange; it provides referrals to medical specialists who are ready to provide support and treatment for them without stigmatized attitudes. Still attitude of medical specialists toward IDUs even in HR programs significantly varies from region to region as well as from case to case. All of the above points of view show importance of establishing and maintaining good relations with health care institutions in the region, while lack of such relations hampers activities and reduce legitimacy of HR work. 48 Factors that influence success and effectiveness of harm reduction programs Internal factors Human resources The following issues were raised during interview in relation to the role of human resources for success of the HR: - HR programs should be implemented by professionally trained people and not in the framework of other daily routine activities. As one of the HR leaders said "These people should be professionally enthusiastic". - Active and enthusiastic HR team: "Devotees with their work being highest priority in life" - Strong leader of harm reduction program who is able to predict local situation with HIV and harm reduction development and deal with it. Role of the leader is important from both internal and external perspectives. Internal role relates to organization of work and functioning of HR project, while external relates to establishment of good relationships with government and other external structures. HR leader must be personally interested in HR "If the leader is indifferent or doesn't personally accept strategy of HR or just makes extra money through work in the project, than project will never work". Majority (37 out of 57) of respondents said the all the projects success is determined by personality and enthusiasm of the leader. - Perfect management including management of outreach work. - Burn-out syndrome of outreach workers which is often prevalent in HR projects, reduce its effectiveness and leads to significant turn over. - Other reasons for turnover is the specificity of the outreach as a social group and a big risk to leave it because of incarceration, uncontrolled drug use, etc. But according to the opinion of some HR leaders outreach workers should be adequately trained: "Highly qualified, well organized work of sufficient number of outreach workers could be even more important than mobile HR points". Organization and regulation. Several comments were related to the organization of work as potential success factor. They concerned with project length, existence of stationary and mobile NEP and project ownership. 49 - Project staff should be aware of the insight of constantly changing type of "narcoscene" and be ready and capable to adjust to these changes. Over formalization was mentioned by HRP staff as a major obstacle to quick and flexible response to changing circumstances. - Project ownership has been discussed in relation to NGO vs. state-owned institution and approximately half of the respondent expressed an opinion that "Best of all if NGO is responsible and state institution is a partner in a project". The state run institutions are seen as ineffective and rigid by NGO staff. - Comfortable location of harm reduction program for IDUs. There were several different opinions on this issue. Representatives of relatively small city marked that when harm reduction program was situated not far from the city center and near main railway station, IDUs visited it more often in comparison with current non-centered location. At the same time it was marked by one of the respondents from HR staff that preferably harm reduction program must be located far from State Drug Control service location and have uncomfortable access road in order to prevent desire of State Drug Control services to conduct additional visits to HR program. Another opinion of one of outreach managers in small town "Location of HR site does not influence its work because outreach work is conducting all over the town, but if we were located closer to the old center we would have more clients and the process of gathering people for training would be also easier". If the town where HR works is rather big, than just one point is not enough and it was suggested that syringe exchange should exist in each policlinic. Around a half of HR staff consider location of HR program as not very important factor at all. - Existence of secondary exchange is very important. Still in some regions its implementation is hampered by street police. - Open hours of harm reduction program most relevant for IDUs. - Outreach workers that have access to IDUs population. There were several controversial opinions on this factor: (1) outreach must be current IDUs or ex-IDUs, (2) Ambiguous attitude toward ex-IDUs doing outreach workers because of burn-out syndrome and high probability of return to drug use, (3) active IDUs must not work in HR programs, even thought attraction of active IDUs as promoters is very important during the program formation because of their access to IDUs environment. - It should be a variety of HR services not limited to bear syringe exchange. Social support and accompanying are not less important than syringe exchange, and socials centers are considered by some respondents as ideal format of HR organization 50 In general the internal success criteria are somewhat limited and are based on project ownership (the best way is a coalition of NGO and state-owned institutions), sustainability (measured by length of time project work), and breadth of provided services combined with accessibility through combination of mobile and stationary NEP "located in 'hot' zones" and existence of effective, well trained and devoted project team. External factors Political support The recognition of the projects and support by the authorities has been repeatedly claimed as the most important success factor. Co-funding from local budgets is hence crucial. As the head of one of the HR projects said "If Russian society will not be involved in HIV problem than we could consider that HIV war has already been lost". Public opinion and mass-media The following factors were mentioned repeatedly by respondents: - The importance of public opinion and mass-media was mostly seen as negative: "[From public side] in case of HIV/AIDS in best cases it is separation, in worst - neglecting and hostility. Different gossips are circulated. Two days ago I have heard that there is a plan of HIV infected to plant infected needles to infect children". The police officer explained, "if mass-media are critical, one almost can not prove anything" and added that "the less discussion of it [harm reduction] the better population attitudes there will be to the project". On the other hand in some places HR staff doesn't consider mass-media as an important player at all. They have no examples of negative influence. And they say that they have a lot to do besides mass- media. In general public opinion is viewed as less important for success as views of authorities or health care providers. Mostly it is the negative public opinion and mass media position that could hamper development of harm reduction. The need for positive changes in public opinion is not emphasized. A success criteria is hence the absence of negative public opinion or explicitly neutral attitude, rather than positive attitude of the public. 51 Financial issues All the respondents mentioned insufficient financing and especially from governmental structures. That is why role of the donors was highlighted "Without donors we will just die because the government is not mature yet in order to recognize us as professionals". The main article in the framework of the project budget that urgently needs to be increased is salary for the staff: outreach workers, medical specialists. As it was underlined the daily work of theses people is highly emotional and associated with contacts with difficult social group. At the same time if at the beginning of the projects in end of 1990s their salary was rather comparable with the average salary in the country now it is less than survival minimum. The respondents even say that if financing issues will not be solved the HR in Russia will be doomed. Sometimes the total budget of the project is sufficient, however it is not allowed to redistribute it between the budget line due to strict donor requirements. Peculiarities of prison HR projects. Achievements and failures of prison harm reduction programs Representative of prison staff of one the prisons where HIV prevention program was implemented illustrated program achievements by the following opinion: "There are two prisons near each other. Prevention program works only in one of them. There is huge difference in knowledge level of both inmates and prison staff of these two prisons with significantly higher knowledge level in the prison where education provided. Before initiation of HIV prevention program levels knowledge were almost equal in both prisons. It proved HIV prevention project in our prison to be very effective". Another respondent mentioned an achievement of absence of HIV transmission within the prison, as well as reduction of hepatitis transmission. There are also educational achievements that are reflected in reduction in numbers of disturbances and improvement of inmate's behaviors. The psychologists mark as the important side-effect result of the projects, that they help to implement and maintain socialization of prisoners and decrease their aggressiveness. As the main failure the majority of the staff and peer educators mentioned absence of social support after release from penitentiary system. It was pointed out that the projects succeeded in teaching people healthy attitude but they are not socially integrated afterwards and are rejected by the society. 52 Among one of the significant failures marked by prison staff was "absence of internal documents on HR strategy, it is reflected only in methodological guidelines" Factors that influence success and effectiveness of prison harm reduction projects Internal factors Volunteers and peer-educators are succeeding to inform majority of inmates about ways of HIV prevention, safer sexual practices. The rates of condom use increase among inmates. Importance of peer education was mentioned by both inmates and prison staff. Representatives of civil narcological centers and State Drug Control Services also marked importance of peer education in prisons. - Availability of condoms and alcohol swabs was emphasized by peer educator as important factor. Importance of condoms availability was also mentioned by prison project coordinator. "First trial with condom distribution in the prison was undertaken in 2001. It was a very interesting experience; some inmates even used condoms as a currency. After that the decision was made to increase quantity of distributed condoms in prisons". - Availability in significant amounts of magazines devoted to HIV/AIDS issues in prisons was also highlighted as priority by peer educators - Conducting of trainings was mentioned as an important factor by both prisoners and prison staff. It contributes to changes in risky behaviors and increased level of knowledge. Trainings provided by external trainers are at higher priority because inmates more willingly contact with them. Additionally the following key success factors were emphasized by respondents: - Positive attitude of Penitentiary Service leaders - Presence of professional psychologists - Education and information of prisoners should be combined with staff education - Inmates involved in peer education activities highlighted that prisons with separate detention of HIV positive patients do not contribute to the development of tolerance toward HIV positive people - Education of prison administration (both internal trainings and educational trips) proved to be very important for changing their attitude toward HR and implementation of HIV 53 prevention programs in prisons. One of the respondents responsible for coordination of such program said: "In the beginning I had doubts about efficiency of HR programs, did not completely understand such strategy, and had some morale difficulties. But after I knew more about it I have changed my attitude and currently I am supporter of HR" - Coordinator of one of the programs also added that prevention programs must be implemented steadily: "If the program was implemented during two years and than stopped, it's not very effective. Permanent functioning is needed" External factors Discussion of the role of external factors affecting prison HR projects is not fully appropriate because the system and prisoners are isolated from the external world. Numerous external factors that were mentioned by the respondents are mainly related to the prison administration and the Federal Penitentiary Service as the main decision-makers. The following citation of prison staff representative confirms this conclusion "In our system people are isolated, we do not have serious external influences. Our administration has the ultimate power". Also the main problem is absence of interdepartmental cooperation on HR issues in Russia. "Police arrest IDUs, in prison sector syringes prohibited and at the same time locally we have to provide IDUs with clean syringes somehow. In order to solve this problem of contradiction it should be reflected in legislation". 54 RESULTS: REPERTORY GRID TECHNIQUES As there are no clearly defined success factors, the expert opinion on this matter is important. The experts themselves may clearly realize what success factors are and what are not. They may consider one HR program as more successful than another, but believe that factors of success are different from what really caused differences. The evaluation of success factors for harm reduction programs should then rely on the comparison of the different projects some of which are more successful than others. To perform this task first of all it is necessary to know what factors experts use for projects evaluation and what factors are responsible for different markings obtained by projects. Materials and methods: the repertory grids analysis One of the methods used for obtaining deeper understanding of how a person (an expert) perceives external objects, such as harm reduction programs is a repertory grid technique described in Appendix 4. This technique is akin to self-generated questionnaire and consists of two stages. On the first stage expert is asked to specify three HR programs and than identify which of them are similar to each other and differ from the third one. After that expert is asked to specify what criteria were used to distinguish two projects from the third. These criteria are noted and further referred to as an effectiveness criteria. On the final stage expert is asked to rate on a scale from 1 to 5 all projects according to the formulated criteria. For the present study each expert was asked to think about three HR programs familiar to him for further evaluation. In order to have information on whether rating was positive or negative, each expert was asked to imagine two reference HR projects: the most successful project (indicated as IP ­ ideal project) and a project unable to fulfill its tasks and achieve goals (indicated as BP ­ bad project). Both ideal and bad projects were virtual projects. Experts were asked to complete table reflected triads of three different projects to be compared. Table consists of eight different triads including real HR projects and reference projects. In the result they analyzed eight groups and generate eight criteria each. (Appendix 2) If an expert was familiar with less than three projects, he was asked to select one project familiar to him during a long period of time. Then expert was asked to recall the project in its initial phase or during the first six months after the expert became acquainted with the project. This state of the project was indicated as P0 (Project, time 0). Then the expert was asked to think about current state of the project. This state of the project was indicated as PP (Project in present time). After that the expert was asked to forecast what will happen with the project within 1.5 ­ 2 years. This vision of the expert was indicated as PF (Project in the future). The expert was asked to think about each of the above-mentioned states of the project as if they were independent from each other individual projects. Then comparison of triads was based on the same logic as was used for the 55 analysis of three different projects (Appendix 3). Analysis of repertory grids Totally experts have completed 36 grids on civil HR programs. Identification of the factors determining effectiveness of civil HR programs with the use of repertory grids was based on 19 different harm reduction programs. Among those most frequently cited programs were those in Balakovo, Kazan, Saint-Petersburg, Vologda and Pskov (table 5) Of the total 36 completed repertory grids, 25 (69%) were based on comparison of one project in three different states (repertory grid type 2) and 11 grids (31%) were based on comparison of three different projects (repertory grid type 1). All experts except one generated 8 comparison criteria in each grid, giving 287 criteria for the effectiveness evaluation. It should be mentioned that that more prevalent use by the experts of the repertory grid type 2 did not affect the results of the study. Theoretically more frequent use of repertory grid type 2 might limit number of elicited constructs and increase `fusiness' of the grid. In order to check this assumption average coefficient of the grid intensivity for grids type 1 and type 2 were calculated. Results indicated no significant differences between two groups. That is why they could be used together and has no implications to the interpretation of the results. Table 5. City location of harm reduction programs selected by the experts in repertory grids City where HR program is # of times program % of all implemented was selected for selections comparison Balakovo 11 18.97 Kazan 7 12.07 St.Petersburg 7 12.07 Vologda 5 8.62 Pskov 5 8.62 Voronezh 4 6.90 Tver 3 5.17 Tomsk 3 5.17 Volgograd 2 3.45 Nyzhnii Novgorod 2 3.45 Barnaul 1 1.72 Novgorod Velykij 1 1.72 Irkutsk 1 1.72 Krasnojarsk 1 1.72 Lipetzk 1 1.72 Mytishi 1 1.72 Omsk 1 1.72 Orenburg 1 1.72 Habarovsk 1 1.72 To analyze criteria that experts used during the repertory grids completion a bootstrap content analysis was used. On the initial stage a theory-based seed classification was used, which 56 divided all criteria by process and outcome criteria. The process criteria were further subdivided into criteria internal and external for the project. All criteria were first divided by those three groups. Then finer grouping was conducted. If a criterion was similar to the previous one, they were put in the same category. If they were differed a new category was created. All subsequent criteria were compared with those categories and if necessary, new criteria were created. If category held several criteria more attempts at creating additional categories were undertaken. If the meaning of a criterion was not entirely clear it was compared with other criteria from the same repertory grid and projects ratings (especially reference) were taken into account. After complete classification was created the criteria were reviewed once again to ensure that they were located in the appropriate categories. Analysis of criteria started with identification of the dominant criteria in the grid. A dominant criterion was a criterion that had highest overall correlation with other criteria in a given grid. The dominance was measured with Bannister coefficient. A criterion with highest Bannister coefficient was considered as dominant. Analysis was appended with calculation of summary Bannister coefficient. In order to avoid confusion and differentiate Summary Bannister coefficient from construct Bannister coefficient it was decided to use the term "intensivity of grid" that has the same meaning as Summary Bannister coefficient. This coefficient shows how `fused' is evaluation system of a given expert (in a given grid). The closer intensivity of grid coefficient to the maximum, the more `fused' is evaluation system of a given expert (the less number of really different evaluation criteria he used internally means that all named criteria are just synonyms). This issue could be more clarified by simple example. If intensivity coefficient is equal to 5054 that reflects 90% of maximum possible value which is 5600 (see notes in Table 7) than almost all constructs have the same meaning. Conversely, if intensivity coefficient is equal to 1703 that reflects 30% of maximum possible value than construct system contains different sets of criteria for project evaluation. In the present analysis only six grids contained systems with more than 70% `fusiness' and only two had `fusiness' more than 80% meaning that majority of experts used different sets of criteria for project evaluation. If two criteria had the same Bannister coefficient, they were included into analysis. If more than three such criteria existed it meant that no dominant criteria exist. In present analysis four grids had no dominant criteria. Further analysis was based on experts' evaluation of the quality of HR project or project states. To make different grids comparable, HR projects/project states distances from ideal project were scaled in the way that zero value had worst (reference) project and value "1" ­ ideal project. 57 Each grid contained information on three different harm reduction projects rated by eight criteria. This formed 861 individual values that were used for evaluation of significance of criteria categories. For this purpose a Spearman correlation coefficient was calculated between ratings of different criteria and assessed value of the project. If criterion was negative (that is absence of important characteristic) the sign of the rating was reversed. Different categories contained different numbers of the individual criteria and so direct comparison of correlation coefficients was unreliable. To overcome this problem the categories were compared based on p-value associated with correlation coefficient. P-value was dependent on both number of criteria (sample size) and absolute value of correlation coefficient. If p-values were similar (or were less than 0.0001) the absolute value of correlation coefficient was used for criteria comparison. The lower confidence interval of correlation criteria by z-method was calculated and used in criteria evaluation. The analysis was done for four levels of aggregation of criteria. All calculations were made in SAS version 8.2 (SAS Institutes, Inc., Cary, NC), according to algorithms described in Plavinski (2006). The distance calculations were made with DISTANCE macro (SAS Institutes). 58 Results The bootstrap content analysis of 287 criteria made by experts during completion of 36 grids led to 36 categories that were combined into nine groups of internal and seven external process variables. Outcome variables were not divided into categories. Two criteria were not included into classification. Detailed division of criteria according to the major classification categories is presented in table 6. Majority (96.8%) of criteria used by experts were process indicators showing that experts do not see outcome indicators as a main factor of HR projects effectiveness. It probably could be explained by many external factors besides HR programs that can influence outcome. Nevertheless six of the above mentioned outcome indicators include decrease in number of new HIV cases among clients of HR programs and decrease in number of viral hepatitis. In one case outcome variable was appeared as dominant criterion for evaluation. In total external variables were used by the experts in 19.2% of all cases, whereas internal variables in 78.1%. The external variables that have maximum number of individual criteria were financing (4.5%) from central and local government and governmental support (4.9%). Usually those criteria were named as `support from local government' or `project recognized on a local and state level', etc. It is possible that the reason why "financing" did not receive the first rating by experts completing repertory grids, is that our sample of HR projects included mostly those which had long-term sustainable funding and this criterion was not appropriate in majority of cases in order to distinguish two HR projects from the third one. Internal variables were divided into variables dealing with interactions with clients (the most important was whether or not project has access to target group), specifics of the employees (whether staff is professional, experienced and has possibility for trainings), coverage of the target group, quality of communication and reporting, quality of management and atmosphere inside the HR program. Among internal variables most frequently used category was atmosphere inside HR program, which included such factors as motivation of the project staff, its salary and other characteristics like capability for innovations, flexibility, good teamwork, organized team, and possibilities to support staff. All of these factors contribute to the project sustainability. 59 Table 6. Categories of the criteria used for evaluation of the effectiveness of harm reduction programs N criteria % of all criteria 1 Outcome variables* 6 2.1% 2 Process Variables 279 97.2% 2.1 Internal processes of the project 224 78.1% 2.1.1 Clients 34 11.9% 2.1.1.1 Project factors influencing contacts 1 0.4% 2.1.1.2 Clients factors influencing contacts 2 0.7% 2.1.1.3 Access to target groups 15 5.2% 2.1.1.4 Trusting relations with members of target groups 6 2.1% 2.1.1.5 Knowledge about harm reduction program among members of target groups 4 1.4% 2.1.1.6 Other 6 2.1% 2.1.2 Employees 47 16.4% 2.1.2.1 Drug users among employees 6 2.1% 2.1.2.2 Experienced staff 2 0.7% 2.1.2.3 Professionalism of the staff 13 4.5% 2.1.2.4 Number of personnel in the project 5 1.7% 2.1.2.5 Training of the staff 5 1.7% 2.1.2.6 Burn-out syndrome among staff 5 1.7% 2.1.2.7 Other 11 3.8% 2.1.3 Organizational climate 41 14.3% 2.1.3.1 Salary of the staff 4 1.4% 2.1.3.2 Motivation of the staff 10 3.5% 2.1.3.3 Other 27 9.4% 2.1.4 Structural and logictical issues 21 7.3% 2.1.4.1 Existence and activity of mobile needle exchange point 8 2.8% 2.1.4.2 Existence and activity of stationary needle exchange point 4 1.4% 2.1.4.3 Existence and quality of printed materials 5 1.7% 2.1.4.4 Other 4 1.4% 2.1.5 Services provided additionally 42 14.6% 2.1.5.1 Social services 3 1.0% 2.1.5.2 Medical services 23 8.0% 2.1.5.3 Other 16 5.6% 2.1.7 Coverage 23 8.0% 2.1.8 Communication and reporting 2 0.7% 2.1.9 Management and governance 14 4.9% 2.1.9.1 Planning according to epidemics stages 2 0.7% 2.1.9.2 Strong leadership 8 2.8% 2.1.9.3 Other 4 1.4% 2.2 External environment 55 19.2% 2.2.1 External factors, influencing access to services 2 0.7% 2.2.2 Societal environment (public and administration opinion) 11 3.8% 2.2.3 Financing 13 4.5% 2.2.3.1 Governmental/budgetary financial support 6 2.1% 2.2.3.2 Source of financing unspecified 7 2.4% 2.2.4 Support from Government (local and central) 14 4.9% 2.2.5 Relationships with police 6 2.1% 2.2.6 Relationships with health care institutions 7 2.4% 2.2.7 Others 2 0.7% * Outcome variables: high effectiveness of HIV prevention; moderate effectiveness of HIV prevention; decrease in HIV/AIDS and drug use, decrease in HIV prevalence among clients of the HR, no cases of risky injection that caused HIV among HR clients; decrease of viral hepatitis incidence 60 Among the most crucial factors existence of mobile NEP was marked. Location and number of stationary NEPs was also cited as evaluation criteria. Much attention was paid by experts to the services provided by HR projects. Most frequently cited criteria were those related to medical services. Among such criteria diagnostic possibilities, access to ARV therapy, existence of physicians to whom IDU can trust, wide range of medical services provided to NEP clients and even existence of substitution therapy were marked as important. On average almost every seventh criterion used by expert (14.6%) related to availability of services. It means that almost each expert has mentioned access to services in the repertory grid. It is interesting that management and governance of the project were at less attention, despite the prevalent opinion of importance of this factor. Strong leadership as evaluation criteria was selected only in 2.8% of all cases, less than in one out of every four grids. Interesting criterion was mentioned by some experts highlighted that planning of HR project should depend on HIV epidemics stage. At the same time because of the large number of criteria used by experts it was important to identify those that are most important for evaluation of harm reduction programs. It was done by two approaches (1) identification of dominant criteria for each repertory grid; (2) identification of correlating criteria categories with implicit value given to each of the projects by the experts. As seen in table 7, there were 41 dominant criteria (four grids had no dominant criteria and nine had two dominant criteria) of which only one was outcome-related. Large group of dominant criteria (one-quarter of all criteria) were from client-relation category (2.1.1 subgroup) stressing importance of good access to IDUs and good relations with target group. Eight criteria were from employee-related category (2.1.2 subgroup), stressing importance of qualified staff not exposed to burn-out syndrome. Six criteria were related to organizational climate in the project team (2.1.3 subgroup) and breadth of services provided by the project (2.1.5 subgroup). As a dominant criteria such variables as existence of good relations in the team, support for the staff, openness to innovations and organized project team were named. Provided services according to the experts should be wide, including different range of diagnostic possibilities, secondary exchange, high quality of provided services. It was also marked that existence of `bare' needle-exchange should be considered as poor quality of project. 61 Table 7. Dominant criteria for each grid (civil HR projects) Grid Main criteria (by Bannister coefficient) Bannister Intensivity Type of number Coefficient* of grid** criteria 13 No cases of injection-related HIV among NEP clients 543 3572 1. 17 High demand by IDU 440 2681 2.1.1.2 23 Interactions with target group 537 3830 2.1.1.2 32 Low number of visits to NEP 553 3764 2.1.1.3 31 Good access to IDU 656 5054 2.1.1.3 27 No contact with target group 370 2325 2.1.1.3 10 Good relations with target group 444 2785 2.1.1.4 24 Good relations with target group 466 2923 2.1.1.4 36 Trust of IDUs to the program 483 3226 2.1.1.4 14 Clients follow safe behavior pattern 475 2896 2.1.1.6 35 Clients flocking to NEP 444 2720 2.1.1.6 8 High professionalism of employee 581 3974 2.1.2.3 22 Good peer consultants 484 3075 2.1.2.3 31 Sufficient number of outreach workers 656 5054 2.1.2.4 25 Lack of training possibilities 587 4050 2.1.2.5 26 High incidence of burn-out syndrome among employees 533 3476 2.1.2.6 20 Adequate outreach 397 2610 2.1.2.7 27 Low staff turn-over 370 2325 2.1.2.7 21 Good relationships with other administrative units 541 3448*** 2.1.2.7 6 Idea of harm reduction supported by the staff 578 3978 2.1.3.2 18 Staff motivated 223 1348 2.1.3.2 5 Good relations in the team 523 3430 2.1.3.3 19 Support for the staff 498 3264 2.1.3.3 23 Openness to innovations 537 3830 2.1.3.3 34 Organized project team 284 1879 2.1.3.3 15 No time-table of employee contacts with clients 211 1303 2.1.4.4 1 Wide range of diagnostic procedures available to the clients 518 3332 2.1.5.2 11 Low quality of medical services for target group 560 3740 2.1.5.2 16 Social and medical support system for clients 390 2265 2.1.5.2 3 "Bare" needle exchange 607 4279 2.1.5.3 7 Good secondary exchange 296 1703 2.1.5.3 4 Additional services provided 626 4609 2.1.5.3 17 Wide IDU coverage 440 2681 2.1.7 4 Wide coverage 626 4609 2.1.7 2 Activity of the leader 444 2907 2.1.9.2 28 State financing 394 2777 2.2.3.1 15 Financing corresponds to services volume 211 1303 2.2.3.2 25 Recognition of project effectiveness on a regional and state 587 4050 2.2.4 level 2 Good contacts with authorities 444 2907 2.2.4 33 Good relations with authorities 378 2103 2.2.4 21 Good outreach work 541 3448*** 2.2.7 * The Bannister coefficient is a sum of squared rank correlation coefficients of a given construct with all other constructs multiplied by 100. The higher is Bannister coefficient the closer given construct associated with other constructs. Maximal value of Bannister coefficient for a construct in a 8-construct grid is 700 (seven correlation coefficients with maximum value 1*100). Correspondingly, the closer Bannister coefficient to this value the more it is `important' in a sense that knowing value of this construct one may predict value of other constructs and total value for the elements. The `important' constructs are more likely influence judgment of quality of the project. ** The intensivity of grid shows how well all construct correlate with each other and whether there are constructs which have low level of correlation with each other (showing that they are measuring different aspect of effectiveness). The maximum value of intensivity for 8-construct grid is 5600 (eight constructs with maximal Bannister coefficient). Such high value means that knowing values for one construct one may precisely predict values for other constructs, implying that those constructs are in fact not different from the first one. Such grids are called `fused' meaning that all elicited constructs are different labels for the same underlying criterion. *** only seven criteria instead of required eight were used by one of the experts 62 In general list presented in table 7 reflects overall structure of criteria used by experts and percentage distribution of different categories that does not significantly differ from table 6. If consider all criteria as a full set of criteria than the question is raised if we could consider dominant criteria as random sample. In this case set of these criteria should be included to the 95% confidence interval (table 8). As seen from table 8 almost all percentage criteria values were included to the 95% confidence interval. The only exception was subgroup 2.1.1 where this value was not included to the confidence interval. There were ten subgroups analyzed and it was expected that in two cases this value would be outside 95% confidence interval by chance alone. For 2.1.1 subgroup the percentage value was included into 99% confidence interval (9.7%-45.2%), meaning that experts did not indicate any subgroup because it contains the most important evaluation criteria. Table 8. Percentage distribution of dominant criteria. Subgroup % of all dominant criteria (95%CI70) % of all criteria71 1 2.4% (0.06%-12.9%) 2.1% 2.1.1 24.4% (12.4%-40.3%) 12.1% 2.1.2 19.5% (8.8%-34.9%) 16.7% 2.1.3 14.6% (5.6%-29.2%) 14.6% 2.1.4 2.4% (0.06%-12.9%) 7.5% 2.1.5 14.6% (5.6%-29.2%) 14.9% 2.1.7 4.9% (0.6%-16.5%) 8.2% 2.1.8 0% (0%-8.6%) 0.7% 2.1.9 2.4% (0.06%-12.9%) 5.0% 2.2 14.6% (5.6%-29.2%) 19.6% Because dominant criteria approach was unable to indicate what category of criteria is the most important for evaluation of harm reduction programs, the second approach aimed to measure influence of criteria categories on the value of the project was used. As was described in material and methods section, this approach is based on use of repertory grid as evaluation instrument. The criteria used by expert in each grid supposed to be used for multidimensional evaluation of the projects. The value of each project was calculated as a simple distance between the investigated project and reference project(s). The distances in this case were calculated between real projects and ideal project. In all grids except one worst (reference) project was furthers away from ideal project. To make project values comparable they were scaled so that distance between ideal and worst reference projects was equal to 1. Because there were 25 grids completed by the experts familiar with the only one HR project it was possible to analyze experts' vision of the future projects development. Results of the analysis 70`Exact' confidence interval, published in Documenta Geigy Scientific Tables, 6th edition (Basel, 1962) 71Percent of the sum of all outcome and process variables 63 indicated that in all cases except two were rather optimistic. They believed that in the future projects would be better than currently and they even will not far from the ideal project. In all 25 projects average value for projects in the past was 0.20 (SD=0.18), indicating that most projects started as relatively poor projects. At present time projects are much better (average value 0.45 with SD=0.21) and will be even better in the future (average value 0.77 with SD=0.26). Exclusion of two pessimistic experts who believed that presently the specific HR project is worse than it was in the past and in the future situation will not improve led to slightly higher values of the projects in the present time (0.46, SD=0.21) and in the future (0.82, SD=0.21). But the value reflected past time did not change (0.20, SD=0.19). One expert believed that nowadays the project is working better than it was in past time but in the future it will work worse. Important to mention that such pessimistic evaluation of the project future was not confirmed by other experts evaluating the same project. Two more experts believed that in the future project situation will not change much. In total 17 experts (68%) believed that projects are moving from poor state in the past to the better situation in present time and to even more better situation in the future, three experts (12%) believed that currently project works worse than it was in the past, but will improve in the future. Five experts were more pessimistic about the future: four of them believed that in the future project will be at the same level or worse despite the fact that currently it works better than in the past. While the fifth expert believed that project is doing worse and worse from the very beginning. For three cities included by the experts into repertory grids it was possible to calculate meaningful average values. It was possible because more than two experts evaluating those three cities. For Balakovo HR project (n=9) the past project state was indicated as poor (average value 0.10, SD=0.07), currently experts considered it much better (average value 0.57, SD= 0.19) and believed in its succeed future (average value 0.87, SD=0.27). Pskov project in the experts view have been also progressing and will do so in the nearest future (3 experts, past value 0.30, SD=0.05, present value 0.56, SD=0.10, future value 0.94, SD=0.05). The same tendency was found for Voronezh HR project (4 experts, past value 0.10, SD=0.08, present value 0.38, SD=0.18, future value 0.83, SD=0.13). Interesting to note that current state of HR project in Voronezh was rated worse in comparison with Balakovo and Pskov HR projects. The Kazan HR project was rated as being without significant changes (3 experts, past value 0.22, SD=0.06, present value 0.31, SD=0.15, future value 0.35, SD=0.24). It is worth mentioning that comparing values of Kazan and Balakovo HR projects it seems that Balakovo is performed better but direct comparison made by one of the experts show that in Kazan the project performs better than in Balakovo (values 0.44 and 0.29, correspondingly). This stresses that experts have different view of the projects they knew and some are just more optimistic than others. 64 In comparison of different cities three experts found no differences between ideal and already existing projects (in two cases they were referring to St.Petersburg project and in one case ­ to Pskov) and in one case between worst possible project and project in N.Novgorod. In all cases different experts gave similar comparative ratings to different programs, though exact values were different because different attitude of the experts to what means to be ideal project. Three cities were rated by three and more experts. The best project was turned to be St.Petersburg (7 experts average value 0.71, SD=0.24). The Kazan project was somewhat worse (3 experts, average value 0.46, SD=0.30) and Vologda project was the worst of those three (3 experts, average value 0.17, SD=0.14). It is interesting to note, that value for Kazan project was close to what was given by experts evaluating progress of project through the time. In both cases project was judged to be halfway between worst and the best possible project (0.46 and 0.31, respectively). The large number of individual values and their dispersion made possible to study answer to the question what criteria are most important predictors of whether experts will recognize project as success or failure. To do so, the correlation between ratings given experts on each criterion and final values of the program were calculated. The correlation coefficients were calculated separately for each category of the criteria. The calculations were repeated for three-level categories and four- level categories. There is no strict rule as to what criteria should be named as important as what as non- important. It is clear, though, that criteria should be based on how well criterion predicts value assigned to the particular project (by the value of correlation coefficient) and how reliable is the estimate of the correlation coefficient. As is noted in Materials and methods, for correlation coefficients p-value is dependent both upon value of correlation coefficient and sample size used for its calculation. The last is also indicator of reliability of estimate of population correlation coefficient. Correspondingly p-value could be used as one of the indicators of predictive ability of criteria. The challenge appears when sample size became relatively large (more than 30 criterion- value pairs) as p becoming less sensitive to the changes in correlation coefficient value. Because of this the second approach has been used, based on calculation of lower 95% confidence limit for correlation coefficient. Width of confidence interval is also dependent upon sample size but its location is much more sensitive to the changes in correlation coefficient value and so should be more informative in cases of relatively large number of criterion-value pairs. The lower confidence limit was calculated according to Fisher's Z method because of necessity to analyze correlation coefficients that could be close to unity. The results of the analysis by three-level categories are presented in table 9. Table 9. Analysis of relative importance of success criteria for civil harm reduction projects. 65 Criteria Categories R LCIR p N Importance p-method Z-method 1 Outcome variables 0.5982 0.1822 0.0087 18 2.1 Internal processes of the project 2.1.1 Clients 0.48619 0.3222 0.0001 102 5 4 2.1.2 Employees 0.50939 0.3757 0.0001 141 3 2 2.1.3 Organizational climate 0.44654 0.292 0.0001 122 2.1.4 Structural and logistical issues 0.45841 0.2376 0.0002 63 2.1.5 Services provided additionally 0.49516 0.3506 0.0001 126 4 3 2.1.7 Coverage 0.51696 0.3193 0.0001 69 2 5 2.1.8 Communication and reporting 0.37187 -0.63 0.4679 6 2.1.9 Management and governance 0.38536 0.0922 0.0117 42 2.2 External environment External factors, influencing access to 2.2.1 services -0.08827 -0.84 0.8679 6 Societal environment (public and 2.2.2 administration opinion) 0.69353 0.4597 0.0001 33 1 1 2.2.3 Financing 0.26054 -0.074 0.1249 36 Support from Government (local and 2.2.4 central) 0.39542 0.104 0.0095 42 Relationships with police and narcological 2.2.5 services 0.18643 -0.307 0.4589 18 2.2.6 Relationships with health care institutions 0.40371 -0.034 0.0695 21 2.2.7 Others 0.77621 -0.096 0.0695 6 As is seen from this table, use of both methods indicated one group of five most important for success criteria. On the first place by both approaches subgroup 2.2.2 ­ good societal environment ­ was placed. It means that all experts believed that good environment with good relations with mass media, interagency interaction, lack of societal stigmatization of IDU, and political support are critical for the success of harm reduction program. At least 25% of variability in program ratings was explained by those factors. The second place according to z-method and 3rd according to p-method was given to good employee. The employee that are experienced, highly trained, have first-hand knowledge of the IDU scene, psychologically in a good shape and exist in sufficient numbers for the needs of the project are key for its success. About 11% of the ratings variability could be ascribed to differences in employee-related factors. The third most important criteria category was breadth of the services provided by the project. It is also interesting to note those factors that did not entered the list of most important. The pure governmental support and financing were judged to be less important than societal environment, stressing the fact that money are not enough for well-functioning harm reduction project and massive campaign directed toward changes in society perception of harm reduction is necessary if one needs to make successful project. Relationship with police and health care institutions also were not included to the list of most important factors. This probably stems from the fact, that in cases when societal environment is good, project will have good relations with 66 police and health care, but if public opinion is against harm reduction the success by good ties with health care and/or police is less probable. From internal factors organization climate inside project, while being important, did not made to the top of the list, probably because some of the factors related to climate has been captured by employee subcategory. Good management and governance was also not among factors that are crucial to the project success. It does not mean that experts believe that good management is irrelevant for success of harm reduction but simply used another approach `Ye shall know them by their fruits72'. The good management will organize process so that the project will have good employee, good relations with clients and so its own `goodness' will be measured by outcome of management work. It is important to not, that good coverage of the target group, though not be in top of the list by both methods, still played a significant role in what experts saw as a good project. More detailed analysis of importance of criteria of success by four-level categories is presented in table 10. As seen from the table good societal environment has retained its first place among criteria of success by both p-method and z-method. Second place went to the `other' subgroup of Employee category. This subgroup included such factors as `mobility of the team', `activity of employee', `understanding of the harm reduction importance among employee', `capability of medics to work with target groups', `low level of staff turn-over', `good outreach team' and `outreach team does not change much' (meaning the same low level of turn-over). Experienced staff was important criteria by z-method, but not made to the top list by p-method (due to small number of experts naming it, but for those experts it was extremely important criterion). The only left criterion that has made to the top list by both methods was availability of medical services to the IDU in harm reduction program. 72Matthew 7:16 67 Table 10. Analysis of relative importance of success criteria for harm reduction projects, detailed categories Criteria Categories R LCIR p N Importance p-method z-method 1 Outcome variables* 0.598 0.182 0.0087 18 2.1 Internal processes of the project 2.1.1 Clients 2.1.1.1 Project factors influencing contacts 1 3 2.1.1.2 Clients factors influencing contacts 0.794 -0.048 0.059 6 2.1.1.3 Access to target groups 0.326 0.036 0.0287 45 2.1.1.4 Trusting relations with members of target groups 0.564 0.131 0.0149 18 Knowledge about harm reduction program 2.1.1.5 among members of target groups 0.764 0.339 0.0038 12 5 2.1.1.6 Other 0.607 0.195 0.0076 18 2.1.2 Employees 2.1.2.1 Drug users among employees 0.401 -0.082 0.0996 18 2.1.2.2 Experienced staff 0.912 0.387 0.0112 6 3 2.1.2.3 Professionalism of the staff 0.339 0.027 0.0345 39 2.1.2.4 Number of personnel in the project 0.661 0.225 0.0073 15 2.1.2.5 Training of the staff 0.618 0.155 0.014 15 2.1.2.6 Burn-out syndrome among staff 0.505 -0.01 0.0548 15 2.1.2.7 Other 0.664 0.415 0.0001 33 2 2 2.1.3 Organizational climate 2.1.3.1 Salary of the staff 0.438 -0.181 0.1541 12 2.1.3.2 Motivation of the staff 0.502 0.174 0.0047 30 2.1.3.3 Other 0.448 0.253 0.0001 80 5 2.1.4 Structural and logistical issues Existence and activity of mobile needle 2.1.4.1 exchange point 0.335 -0.079 0.1098 24 Existence and activity of stationary needle 2.1.4.2 exchange point 0.702 0.215 0.0109 12 2.1.4.3 Existence and quality of printed materials 0.4 -0.142 0.1401 15 2.1.4.4 Other 0.734 0.276 0.0066 12 2.1.5 Services provided additionally 2.1.5.1 Social services 0.04 -0.641 0.9183 9 2.1.5.2 Medical services 0.559 0.372 0.0001 69 3 4 2.1.5.3 Other 0.4 0.131 0.0048 48 2.1.7 Coverage 0.517 0.319 0.0001 69 4 2.1.8 Communication and reporting 0.372 -0.63 0.4679 6 2.1.9 Management and governance 2.1.9.1 Planning according to epidemics stages 0.647 -0.347 0.165 6 2.1.9.2 Strong leadership 0.274 -0.145 0.195 24 2.1.9.3 Other 0.428 -0.193 0.1648 12 2.2 External environment 2.2.1 External factors, influencing access to services -0.09 -0.84 0.8679 6 Societal environment (public and administration 2.2.2 opinion) 0.694 0.46 0.0001 33 1 1 2.2.3 Financing 2.2.3.1 Governmental/budgetary financial support 0.229 -0.321 0.4112 15 2.2.3.2 Source of financing unspecified 0.402 -0.036 0.0712 21 2.2.4 Support from Government (local and central) 0.395 0.104 0.0095 42 2.2.5 Relationships with police and narcological services 0.186 -0.307 0.4589 18 2.2.6 Relationships with health care institutions 0.404 -0.034 0.0695 21 2.2.7 Others 0.776 -0.096 0.0695 6 * Outcome variables: high effectiveness of HIV prevention; moderate effectiveness of HIV prevention; decrease in HIV/AIDS and drug use, decrease in HIV prevalence among clients of the HR, no cases of risky injection that caused HIV among HR clients; decrease of viral hepatitis incidence Other important criteria (by any one of methods) included knowledge about harm reduction program by IDU, organizational climate factors such as good relations in the team, clearly 68 organized work, coordinated work, positive working climate and good relations in the team, shortly all Hertzberg hygienic factors and degree of coverage of target group by the project Repertory grid technique: peculiarities of prison harm reduction programs. The prison harm reduction projects were evaluated with the help of repertory grid by 5 experts. All those experts evaluated one project in its three different states ­ a project in the past, in the present and in the future. The experts has generated 40 success/failure criteria, which, based on bootstrapping approach, could be combined into several major categories (table 11). Table 11. Categories of the criteria used for evaluating effectiveness of prison harm reduction programs N criteria % of all criteria 1.1 Outcome 1 2.50% 2.1 External factors 17 42.50% 2.1.1 Personnel and prisoners 13 32.50% 2.1.1.1 Administration 8 20.00% 2.1.1.2 Personnel 2 5.00% 2.1.1.3 Prisoners 3 7.50% 2.1.2 Non-prison organizations 2 5.00% 2.1.3 Financing 2 5.00% 2.2 Internal factors 22 55.00% 2.2.1 Access to services 3 7.50% 2.2.2 Services provided 7 17.50% 2.2.3 Employee 5 12.50% 2.2.4 Organization 5 12.50% 2.2.5 Other 2 5.00% As seen from the table only one expert has mentioned outcome of the project meaning decrease in recidivism. All other criteria were related to the process of setting up and functioning of the prison harm reduction project. Those criteria were divided into external factors ­ those that are connected with environment in which project is working and internal factors ­ related to project organization, staff and functioning. The external factors were further divided into `human' environment meaning prison administration, prison personnel and prisoners themselves. Every fifth criterion was mentioning the role of prison administration, such as knowledge of harm reduction programs by administration, their understanding of HIV problem and lack of drug in prison problem denial, which are ultimately necessary for support of harm reduction projects which was mentioned by four out of five experts. The positive reaction from prison personnel was mentioned by two experts, one of whom has drawn attention to the necessity of informing prison personnel about safety measures. Three criteria were related to prisoners themselves, stating that for project success it is necessary that prisoners 69 informed about harm reduction programs, they are interested in them and they support harm reduction programs. Additional external criteria dealt with financing issue (necessity to have sufficient financing of harm reduction programs, preferably by state) and role of NGO and non-prison organizations, involvement of which is important, in a view of experts, in prison harm reduction efforts. In the same time large number of criteria were related to functioning of project itself, namely wide range of services provided by the project (involvement of different specialists, access of specialists from civil services to the prison population, social adaptation as a component of harm reduction). In general it should be noted, that though wide range of available services was important set of criteria in both civil and prison harm reduction projects, the experts were more explicit about nature of the services in civil projects comparing with prison ones. The experts on prison projects were mainly speaking about comprehensiveness and, as it follows from negative value attached to isolation of prison population from civil services, for them wide services meant services available to harm reduction program clients outside the prison. Almost every expert has mentioned importance of employee for the project, who should be motivated, have high qualification and, as one expert mentioned, low educational level of outreach workers is hampering effectiveness of the harm reduction projects. The approach to harm reduction, to be successful should be innovative and up-to-date, the project should be well thought-of. The logistic of the project should be better organized for project to be successful. All those factors may but may not play a significant role in the real projects. To answer the question of relative importance of those criteria firstly the dominant constructs in all five grids were calculated (table 12). Table 12. Dominant criteria for each grid interview on prison harm reduction projects Grid Criterion Bannister Intensivity Type of number Coefficient of grid criteria 1 No recidivism 223.8 1186 1 2 No information about harm reduction 602.8 4303 2.1.1.1 programs among prison administration 3 Sufficient length 502.8 3203 2.2.1 1 Positive motivation 223.8 1186 2.2.3 4 Major drug control policies implemented 572 3901 2.2.4 5 New forms and methods of work 341.1 1731 2.2.4 As is seen from the table, in five grids there were six dominant criteria. The only outcome criterion was dominant in one grid together with positive motivation in the project. One dominant criterion was related to information provided to prison administration (in order to be successful, prison administration should have full information about what is harm reduction program and how 70 it works), one was related to access to services ­ the project should have sufficient length and finally two dominant constructs were related to organization of the work of the project, that should base on full implementation of drug control policies and also should be innovative in order to be successful. Unfortunately, due to small number of grids it was impossible to say whether the dominant criteria were really most important, or they just were random sample of criteria population like it was the case for civil projects dominant criteria. To answer this question a second approach in which projects were rated and correlation between project ratings and criteria rating were calculated. In general, the experts on prison project were more pessimistic about future of the prison harm reduction projects comparing with civil projects. Two of the five experts forecasted improvement in the project activities in the future, two believed that situation will be worse, and one though that it would be better but highlighted very poor present level. The median73 value for past of the prison projects was 0.099 meaning that they were in average just a bit better than total failure. Presently they have median value of 0.447. In the future expert believed value would be higher ­ 0.646 but still closer to the halfway between best and worst projects. As seen from table 13 the same criteria are in first places by both p- and z-methods. On the first place there is a set of criteria related to access to services, including high coverage, sufficient length and availability for all prisoners. In general it means that the most important criteria of success for prison project is demonstration that it is `for real', the project is of sufficient length and provides access to all prisoners that are in need. The analysis of grid containing access criteria `wide coverage' was performed via visual focusing and showed that this criterion is tightly associated with such constructs as `support from prison administration', `support from prisoners' and `existing legal framework'. This means that expert believes that existing legal framework leads to acceptance of harm reduction by prison administration that together with prisoners' motivation, results in wide coverage of the prisoners by harm reduction program. Table 13. Analysis of relative importance of success criteria for harm reduction projects z- R LCIR p N p-method method 1.1 Outcome 0.866 0.333 3 73Because of small number of values and significant spread of them it was decided to calculate for prison project not the average, but median values 71 2.1 External factors 0.201 0.201 -0.079 0.158 2.1.1 Personnel and prisoners 0.097 -0.225 0.557 39 2.1.1.1 Administration -0.08 -0.469 0.707 24 2.1.1.2 Personnel 0.393 -0.615 0.441 6 2.1.1.3 Prisoners 0.656 -0.014 0.055 9 4 4 2.1.2 Non-prison organizations 0.828 0.05 0.042 6 3 3 2.1.3 Financing 0.414 -0.599 0.414 6 2.2 Internal factors 0.317 0.317 0.081 0.01 2.2.1 Access to services 0.841 0.424 0.005 9 1 1 2.2.2 Services provided 0.482 0.064 0.027 21 2 2 2.2.3 Employee 0.237 -0.313 0.394 15 2.2.4 Organization 0.21 -0.338 0.452 15 2.2.5 Other 0.406 -0.605 0.425 6 As for the second expert, that has selected availability as a criterion of success, it had less correlation with other criteria, being more connected with positive motivation for harm reduction and outcome criterion ­ absence of recidivism. The second place was occupied by set related to wide range of provided services. It is worth mentioning, that in civil projects array of services was also rated as a top success criteria. This also adds the `reality' to the project, when it strives to provide the same level of services as in civil projects. The final, third place, was given to the links of prison projects with external NGOs and other organizations giving continuity of care for clients of harm reduction projects. Interesting, that despite frequent mentioning of prison administration support as success criteria, in ratings correlation it turned out that that support of prison administration does not discriminate between more and less successful project. Probably this stems from the fact, that support by prison administration is necessary for allowing harm reduction project to start in the first place, but after project has started it does not influence its success. Also financing issues and work atmosphere inside the project seemed to be less important for project success than professional staff of the project. In summary, the prison harm reduction project is successful if it enables to provide access of the prisoners to the services, the services provided in the same volume as outside the prison and clients have continuation of care after release from the prisons. CONCLUDING As we have demonstrated in the literature review best practice is relatively slippery concept, which often consists of a mixture of purposes and activities needed to achieving them. The HR itself is also a controversial approach, and although the majority of literature available on the subject suggests that there are epidemiological and human rights grounds for its implementation, 72 cultural acceptability and implementation issues are often of a concern. Although evidence exists to support the view that HR in Russian Federation had a number of achievements it is quite weak and limited. Additionally HR projects seem to have had different degree of success, which directly leads us to the subject of this document ­ best practices in HR. As indicted in the analysis of the results of the semi-structured interviews in majority of cases leaders of HR projects indicate that distrust and misunderstanding towards the approach was largely overcome at least in the successful regions. The history of HR development in Russia can be viewed as generally successful and public opinion was positively influenced. The decision makers in the regions where the programs are supported and at the federal level have underwent an evolution of attitude towards HR, starting from caution, then shifting to idealistic believe of HR being an absolute solution, and finally some of the respondents were in the realistic state of the development of thinking. The later understood implementation issues, and believed there is a need for further analysis how to ensure that these programs are effective. Funding issue received sufficient attention in both interviews and in repertory grid technique. According to the repertory grids it has received second highest rating. Unfortunately still in the cases when external donors decrease funding there are alarming trends of HR programs closing down as no adequate support and co-financing from local sources are available. At the same time there is clear understanding among most decision-makers who agreed to participate in the study that HR helped at least to keep the epidemic in the country under some control. Barriers preventing them from helping mobilize funding for prevention of HIV among risk groups are not always clear. The influence of the State Drug Control and other law enforcement agencies was quite important. For instance, recent change in most regions led to what's called "closed narcoscene", implying that the sales of drugs are not street-based, but through home deliveries and personal connections, which makes it much harder to enroll drug users into HR. While there seems to be lack of strong opposition at the top level of the police, the junior officers on the street are often hostile to IDUs and outreach workers in many of the regions. There was consensus among respondents that overregulation presents threat to successful implementation of the programs. HR technologies should be flexible and require constant modification. HR should not be viewed as the cook-book and routine practices might not work for a long time. Along with starting new projects in new places there should be provision of more flexibility and autonomy to the existing "old" projects. That could allow to try some innovative techniques and to be more flexible in redistribution of functions and finances. As suggested by respondents HR projects should be properly and constantly evaluated, however it is less than clear 73 how to address this. There was a suggestion that it would be useful to elaborate separate sets of criteria for "new" and "old" projects. The indicators for the first group could be primarily based on syringe exchange indicators, while for the second group measures of success need to orient towards improvement of epidemiological situation mostly. Attitudes of IDUs were quite positive, although it should be acknowledged that only the HR users were interviewed in the frames of the study. They perceived HR programs as nearly the only organizations which were non-hostile to them. There were some issues with relevance and quality of materials provided to HR clients, however this were reported to have been relatively rapidly addressed. While internal success factors mainly concerned with the personnel and leadership characteristics, external ones were generally deemed more important. The external ones mostly concerned with funding, support of the authorities and law enforcement agencies, and the regional situation with drugs. Placement and routing of the exchange-site, as well as type of organization: i.e. NGO versus state owned, were viewed to be of little importance by the respondents. The analysis of repertory grids confirmed that, according to the respondents, the success factors for harm reduction programs mainly lie in societal environment. According to the technique, if public opinion is positive or neutral towards the harm reduction project and to IDUs in general, the project will have higher probability of successful. Explicit lack of such external acceptance will result in difficulties in setting up and maintaining the project. Repertory grids also confirmed that the good project should have professional, well trained and motivated project staff. To be able to involve significant number of IDU and be successful it is very important for the project to provide IDU with wide arrange of services, apart from `bare' needle exchange. The services should be first and foremost health care related ­ from wide array of possible diagnostic capabilities to treatment and even substitution therapy. Involvement of IDUs and former IDUs as outreaches or recruiters was also mentioned as important, despite the issues resulting from the addiction of the personnel. Experienced, understanding goal of harm reduction staff is crucial for the project success. The services should be provided by the team of people linked by good working relations, creating good working conditions for everybody involved and members of target group should have information about existence of harm reduction program. If this is achieved, the coverage of the IDU by the project, the indictors of the project success are also likely to be satisfactory. According to the semi-structured interviews results the retention of the staff became problematic since recent economic growth had raised the average incomes, making salaries of the project staff less attractive. 74 Prison sector HR programs had some peculiarities related to poorer range of services provided: e.g. needle exchange is still illegal in the penitentiary system. The role of external factors was minimized as the programs exist in a very isolated and relatively controlled environment, and there is a lot more power concentrated in the officer's hands. In summary, the best practice project should invest resources in changing societal perception of harm reduction program to more positive, have a well planned recruiting and retention strategy for the staff so, that it will be experienced and with good working relations and be organized so as to provide wide range of services to the IDU, including medical care and information, as well as make sure that target group is informed about services provided. KEY POLICY IMPLICATIONS - HR must be coherent part of the general HIV prevention strategy, and not an independent isolated component (in both civil and prison sector) - It is very important to ensure integration of civil and prison HR projects, while the high exchange rates between civil and prison population of IDUs - Trainings for law enforcement agencies staff should be regular, given high turnover rates among low-level law enforcement officers and the very strong negative influence caused by their opposition in many sites - Donors flexibility is viewed as key by respondents (at least with HR projects effectively functioning during long time period) avoiding sometimes over formalization of HR programs and allowing redistribution of budget between the budget lines - Closing "narcotics scene" poses new challenges for HR projects. - The importance of involvement of active IDUs into outreach work or into recruitment and maintenance of cohorts of clients greatly increases as the work of State Drug Control Committee becomes more effective. - Surprisingly the location of the HR project was not viewed as important, and multiple respondents expressed doubts regarding the mobile exchange sites. Outreach work is perceived as the most important. - The external environment is more important then the internal procedures of the organization implementing HR according to most respondents. However in prisons it is the internal factors which matter, as the projects tend to be isolated from the external influences. - Implementation of substitution therapy and of syringe exchange in prisons seems not feasible in the nearest future, as even some of staff of the HR programs and the majority of decision makers are not ready to consider this possibility. 75 - Combination of HR with other medical and social projects (i.e. human rights protection) can be very effective. - Breadth of services HR programs provide, even if some of the functions do not help control HIV epidemics, attracts the clients and motivates staff, who get burned out and board from doing just needle exchange. - There is a need to legalize and formalize the status of outreach worker, as position, requirements, training, carrier path are all unclear and unofficial. 76 APPENDIXES Appendix 1 Questionnaire Good afternoon, The Open Health Institute is conducting research which aims at improving the understanding of the role, place and factors of harm reduction programs success. The research results will enable to improve measures to prevent HIV epidemic in Russia. We think you may be an important source of information for our research, and would like to have an interview with you. It won't take more than 40 minutes. Your frankness is extremely important for us, therefore the interview is confidential. All received information will be used in the generalized and non-personalized (without indicating names) form, and the project staff, who will analyze the data, won't know anything about you. We would like to have your interview recorded with a dictaphone. The cassette will be sent to Moscow, then stenographers will prepare shorthand notes of the interview, thus analysts will not be able to identify you by your voice even if you have met. You have the right to withdraw your participation in the interview now, or at any time during the research, and to give no answer to any question during the interview. Please confirm your consent to the interview. Do you agree to be recorded with a dictaphone? Respondent's sex Approximate age Position of the interviewed person (department, program officer, client) and organization he/she represents Region Tell the story of your interaction with harm reduction programs (Fishing questions: have you heard of it, worked for it, been its client; how long, in what way?) Has your attitude to the harm reduction strategy been changed since the moment of your acquaintance with the idea and until now? In what way? In your opinion, does (do) the harm reduction program(s) you are acquainted with work effectively? What are the basic achievements of the harm reduction program(s) you are acquainted with? What are the basic failures of harm reduction? How do you think the efficiency of harm reduction programs could be estimated, and how can a successful program be distinguished from an unsuccessful one? How do you think the harm reduction program(s) you are acquainted with has (have) managed to achieve success in preventing HIV epidemic? (ask for copies in case of references to documents) What external factors (state services, population, medical community, drug mafia) influence the work of harm reduction programs? In what way, and who exerts more influence? Who exerts negative, and who, in your opinion, positive influence? Why? 77 Has the attitude of different institutions, population, clients, doctors to the harm reduction program changed? What made for it? What was the role of the harm reduction program in forming public opinion of harm reduction methods and drug users? What do you think were the main factors ensuring success in harm reduction work? - external factors - internal factors What were the main factors entailing failures in harm reduction work? - external factors - internal factors What characteristics should a harm reduction program possess in order to achieve success? (Fishing questions: role of a person, active manager, sociable and charming person promoting harm reduction, well-qualified staff, motivated staff, availability of a team, absence of conflicts, availability of funds, spelled out procedures of activities, good geographical location, long-term work of the program, involvement of users, or ex-users as peer employees, teaching peer consultants in the penal system to work with convicts, a base on the basis of an AIDS centre, or an NGO). What the external conditions should be like to enable the program work effectively? (Fishing question: attitude of the Federal Service for Control of Drugs and Psychotropic Substances Circulation, mayor or vice-mayor, governor, doctors, health care department, militia, narcologists, AIDS centre, the penal system institutions; availability of a strong NGO movement, various sources of funding) How does the situation with drug using and HIV epidemic in the region influence the success of harm reduction programs activities? Which factors, in your opinion, are more important ­ external or internal? What do you think is the role of the following institutions and communities in harm reduction programs activities: NGO AIDS centre Narcological dispensary Federal Service for Control of Drugs and Psychotropic Substances Circulation Militia Penal system institutions Donor organizations Population, mass media Drug dealers Outreach workers (for civil harm reduction programs) Peer consultants (for prison projects) Drug users (clients of harm reduction projects and other people) Convicts Whom would you recommend for talking to on the issue? 78 Appendix 2 Repertory grid type 1. Self ­ administered questionnaire for respondents who have experience with at least three projects Organization___________________________________________ Name (optional)________________________________ SELF-ADMINISTERED QUESTIONNAIRE Code (for the interviewer)__________________ Purpose of the questionnaire ­ find out on what basis you assess the quality of harm reduction projects and/or HIV prevention projects in prisons Mark off with a tick a harm reduction project or HIV prevention project in prison depending on the field of you expertise. Harm reduction project HIV prevention project in prison This poll is conducted in relation to three harm reduction projects and/or HIV prevention projects in prisons familiar to the expert. These projects are indicated as P1, P2 and P3. The expert is asked to number the projects (1, 2, 3). If the expert is a leader or a staff member of one of the projects, he is not allowed to evaluate this project using the proposed scheme. P1______________________________________ P2___________________________________ P3____________________________________ Besides, the expert is asked to imagine the most successful project (indicated as IP ­ ideal project) and a project unable to fulfill its tasks and achieve goals (indicated as BP ­ bad project). IP and BP are virtual projects. Further the expert is asked to imagine three project in accordance with the scheme presented below (for the first row of the table the expert is asked to imagine the ideal project and projects with numbers 1 and 2 ­ P1, P2). Which two projects of the three ones are similar to each other and differ from the third one. Mark of with ticks to those circles which are similar. Now the expert is asked to specify in what characteristic these two projects are similar and differ from the third one. What do they have in common? Indicate the common characteristic in the left- hand column of the table. In the right-hand column indicate the antipode of that characteristic. In other words, how that third project, which does not have such characteristic, can be described. 79 The expert is asked to complete the whole table in the way described above Pole of "ticks" IP P1 BP P2 P3 Pole "without ticks" 80 Now the expert is asked to copy to the below table the contents of the left - and right-hand columns of the table he has just filled (copy characteristics of the projects). In the cells marked with ticks put number "5", in the cells with an empty circle put "1". Remember that "1" does not mean a low grade, as well as "5" does not mean a high grade. "1" and "5" are just used for indication of the grade pole. The expert is asked to imagine those projects which were not used in order to characterize the projects (for instance for the first row of the table it would be Bad Project and Project number 3 (BP and P3). To what extent each of these projects shows the characteristics described in this row? Evaluate it using a five-point scale and taking into account that 1 point is related to the project which has an "antipode" characteristic. Now the expert is asked to complete the whole table in the same way. Pole of "ticks" IP P1 BP P2 P3 Pole "without ticks" Thanks for your cooperation! 81 Appendix 3 Repertory grid type 2. Self ­ administered questionnaire for respondents who have experience with one project only Organization___________________________________________ Name (optional)________________________________ SELF-ADMINISTERED QUESTIONNAIRE Code (for the interviewer)__________________ Purpose of the questionnaire ­ find out on what basis you assess the quality of harm reduction projects and/or HIV prevention projects in prisons Mark off with a tick a harm reduction project or HIV prevention project in prison depending on the field of you expertise. Harm reduction project HIV prevention project in prison This poll is conducted in relation to the harm reduction project familiar to the expert during a long period of time. The expert is asked to recall the project in its initial phase or during the first six months after the expert became acquainted with the project. This state of the project is indicated as P0 in the table (Project, time 0). Then the expert is asked to think about current state of the project. This state of the project is indicated as PP (Project in present time). After that the expert is asked to forecast what will happen with the project within 1.5 ­ 2 years. This vision of the expert will be indicated as PF (Project in the future). The expert is asked to think about each of the above- mentioned states of the project as if they were independent from each other individual projects. Besides, the expert is asked to imagine the most successful project (indicated as IP ­ ideal project) and a project unable to fulfill its tasks and achieve goals (indicated as BP ­ bad project). IP and BP are virtual projects. P0, PP, PF (indicate the name of the project and city where it's implemented)___________________________________________________ Please note that Project in the future (PF) does not reflect how the Project MUST look like, it's the expert's evaluation of how the project will most likely look like taking into account changes in the country, in the region, situation with HIV/AIDS, financing of the project and political situation. Further, the expert is asked to imagine three project in accordance with the scheme presented below (for the first row of the table the expert is asked to imagine the ideal project, a project in its initiation phase and a project at present ­ IP, P0, PP). Which two projects of the three ones are similar to each other and differ from the third one? Mark off with ticks those circles which are similar. Now the expert is asked to specify in what characteristic these two projects are similar and differ from the third one. What do they have in common? Indicate the common characteristic in the left- hand column of the table. In the right-hand column indicate the antipode of that characteristic. In other words, how that third project, which does not have such characteristic, can be described. The expert is asked to complete the whole table in the way described above 82 Pole of "ticks" IP P0 BP PP PF Pole "without ticks" 83 Now the expert is asked to copy to the below table the contents of the left - and right-hand columns of the table he has just filled (copy characteristics of the projects). In the cells marked with ticks put number "5", in the cells with an empty circle put "1". Remember that "1" does not mean a low grade, as well as "5" does not mean a high grade. "1" and "5" are just used for indication of the grade pole. The expert is asked to imagine those projects, which were not used in order to characterize the projects (for instance, for the first row of the table it would be Bad Project and Project in the Future (BP and PF)). To what extent each of these projects shows the characteristics described in this row? Evaluate it using a five-point scale and taking into account that 1 point is related to the project which has an "antipode" characteristic. Now the expert is asked to complete the whole table in the same way. Pole of "ticks" IP P0 BP PP PF Pole "without ticks" Thanks for your cooperation! 84 Appendix 4 Description of repertory grid technique methodology. The Repertory Grid technique is methodology developed in the framework of personal construct theory (PCT). The PCT in turn holds that each person organizes his/her perception of external world according to the number of preset templates that are formed during life of the person. The repertory grid technique was proposed by George Kelly in 1955. George Kelly stated, "the person's processes are psychologically channeled by the ways in which he anticipate events". In other words the person tries to understand flood of sensory perceptions by fitting it to the templates, called constructs. He/she sees what he expects to see. If the results of action undertaken based on the given construct system does not pass reality check the person is changing his/her constructs. Because constructs are formed during whole life of the person the personal construct system is unique, giving situation in which every person sees world differently. This has important consequence for the study of expert opinions with the help of questionnaires and interviews as different experts could describe same criteria of effectiveness by different wordings and, from the other side, they can use different words as synonyms for the same construct. In the same time the larger share of the background between persons the more their construct system will be alike. Kelly himself has made distinctions in understanding reality by different persons: Individuality: "persons differ from each other in their construction of events." Communality: "to the extent one person employs a construction of experience which is similar to that employed by another, his psychological processes are similar to those of the other person." Socialty: "to the extent that one person construes the construction processes of another he may play a role in a social process involving other person." This existence of shared reality (communality) allows the study of expert opinion from the group of experts. Because we cannot directly observe how different persons construe a world there is a need for technique to elicit their constructs. A person is presented with several objects74 and asked what objects are different and what are alike. Then he/she is asked to name what is different between them. He/she is asked to identify two poles for construct, for example `good-bad' or `sweet-sour'. This elicitation of second pole is important because it helps interviewer to understand meaning of the construct. According to the PCT person finds differences between objects because some are fit to one template (construct) and other is not. Continuing presenting different combinations of the objects of one category one may hope to elicit whole construct system of the person. This technique is called repertory grid interview. In the same time a person could use the same construct for judging different combination of the objects and there is a need to see, whether person really have a system of different constructs or there is only one construct (in case of evaluating different projects ­ `a gut feeling') that has different names. To accomplish this a person is asked to rate all objects that were used for elicitation of constructs on a scale from 1 to 5 according to their closeness to one or another pole of the construct. Then the completed grid is subjected to variety of manipulations that help to quantify `alikeness' of elicited constructs75. The one approach is to calculate rank correlation coefficients between all pairs of constructs and then calculate Bannister coefficient (sum of squared correlation coefficients) and intensivity coefficient (sum of all Bannister coefficients). The closer intensivity coefficient to the 74Called in repertory grid techniques `elements' 75There are several other methods to study construct system, namely resistance grid (formed by asking person to compare two constructs and indicate which one is more important) and implication grid (formed by asking person what happened to another construct if one will change its polarity). Those methods were not used in this study because of heavy time consumption (with 8 constructs resistance grid demands 28 comparisons and implication grid ­ 56 comparisons). 85 maximal for a given grid size, the less variable his/her construct system and the higher probability that he/she is just using different names for one discriminative criterion. Main fields of repertory grids application remain clinical and educational psychology but there is an increasing interest in its applications to employee training and development, job analysis, job description and evaluation. The repertory grid is often used in the qualitative phase of market research, to identify the ways in which consumers construe products and services. In health care the repertory grid is used mostly for elicitation patient preferences for different treatment regimens and frequently used in health care education. 86