Russia has experienced one of the fastest growing HIV epidemics in the world, with an estimated 90% of the country's HIV cases registered after 2000 [1,2]. In 2005, the official number of registered HIV cases in Russia exceeded 300 000; however, experts believe the number of HIV-infected individuals is closer to 860 000 (range: 420 000–1.4 million) [2–4]. The HIV epidemic in Russia has been fuelled largely by a dramatic rise in injection drug use over the last 15 years. The estimated 1.5–3 million injection drug users (IDUs) in Russia account for 70–85% of all registered HIV cases [1,3,5].
Virtually no active IDUs and only 5000 non-IDUs in Russia received highly active antiretroviral therapy (HAART) in 2005 [1,6]. By the end of 2005, approximately 139 000 individuals in Russia are expected to need HAART, although the number may be as high as 250 000 [1,7]. HAART dramatically reduces mortality from HIV infection. HAART also reduces HIV infectivity and may therefore provide substantial benefits from reduced transmission of HIV [7–10].
Although the HIV epidemic in Russia has been driven by injection drug use, current HIV treatment resources are targeted almost exclusively to non-IDUs, in part because of concerns about adherence with HAART among IDUs . To understand whether such a strategy could be successful in slowing the HIV epidemic in Russia, we evaluated the effectiveness and cost-effectiveness of treatment strategies that target non-IDUs. We also evaluated treatment strategies that target IDUs, and untargeted strategies that provide HAART without regard to injection drug use status. Previous studies have evaluated the cost-effectiveness of HAART programs in the US [12,13] but no studies of the cost-effectiveness of HAART programs in Russia have appeared.
We developed a dynamic compartmental model of HIV transmission and progression in a population of IDUs and non-IDUs. Compartmental models have been used to evaluate the cost-effectiveness of a variety of HIV prevention and treatment programs [14–16]. Individuals transitioned between compartments according to rates defined by the dynamics of disease transmission and progression . Additionally, individuals who began HAART transitioned from an untreated to a treated compartment. We implemented the model in Microsoft Excel 2003. Full details of the model are provided in the Appendix, available at www.stanford.edu/∼brandeau/RussiaAIDSAppendix, or from the authors upon request.
We implemented the model using data for St. Petersburg, which has relatively high HIV prevalence, and conducted sensitivity analysis using data for Barnaul, a city with lower HIV prevalence [3,17–20]. Key parameter values and sources are summarized in Table 1; the Appendix (Table A2) shows additional data and sources, and values considered in sensitivity analysis.
We considered the adult population of men and women aged 15 to 49 in St. Petersburg. The population was subdivided into 12 groups, based on injecting behavior (IDU or non-IDU), HIV status (uninfected, asymptomatic HIV, symptomatic HIV, or AIDS), and treatment status if HIV-infected (receiving HAART or not receiving HAART). We defined the stage of HIV infection (asymptomatic HIV, symptomatic HIV, or AIDS) according to CD4 cell count (> 350, 200–350, or < 200 cells/μl). We defined compartments to capture variations in key parameters (e.g., viral transmissibility and quality of life) across different health states. For these analyses, we assumed HIV-infected individuals have already been identified, either through screening programs or evaluation of symptoms.
In the base case, approximately 4% of adults aged 15 to 49 years were IDUs, and initial HIV prevalence was 35% among IDUs and 0.63% among non-IDUs, based on data from St. Petersburg (Table 1 and Appendix Table A2) [3–5,17–19]. In sensitivity analysis, we implemented the model using data from Barnaul, a smaller city (population, 600 000) in south-western Siberia where approximately 6.4% of adults were IDUs, and initial HIV prevalence was 1.7% among IDUs and 0.06% among non-IDUs (Table 1 and Appendix Table A2) [3,20].
Individuals entered the population as a result of maturation (14-year-olds turn 15). All new entrants were assumed to be uninfected. Individuals exited the population as a result of maturation (49-year-olds turn 50) or death from HIV or other causes. We considered the adult population aged 15 to 49 years because the majority of HIV infections occur in this age group. The annual non-AIDS-related death rate for IDUs was significantly higher than for non-IDUs, due to the increased risk of mortality from drug overdose and other complications of injection drug use (Appendix Table A2). Based on data from St. Petersburg, the estimated total annual death rate exceeded the birth rate; thus, the population decreased over time (Appendix Table A2).
HIV transmission and progression
The model incorporated HIV transmission via needle-sharing and sexual partnerships. We estimated the average number of injections per year, the fraction of injections that were shared, and the probability of infection transmission to an uninfected individual per risky injection (i.e., an injection involving a needle previously used by an HIV-infected individual) (Table 1 and Appendix Table A2) [15–17,21–24]. The latter parameter varied across IDU compartments because injection transmissibility depended on viral load and whether the individual was undergoing HAART . Estimates of needle-sharing behavior reflect common practice in Russia, which includes some needle-cleaning prior to injecting .
We modeled HIV transmission via sexual contact on a per partnership basis. All sexual partnerships resulted from random mixing, although we allowed for preferential mixing between IDUs and non-IDUs. We estimated the average number of new sexual partnerships per year, the average condom usage rate, condom effectiveness, and the annual probability of infection transmission to an uninfected individual per unprotected sexual partnership (Table 1 and Appendix Table A2) [15,16,25–29]. The data reflect average sexual behavior practices in the non-IDU population, which may include high-risk individuals such as commercial sex workers.
We estimated the rate at which HIV-infected individuals progressed through disease states based on a detailed model of the natural history of HIV infection and the effect of HAART on disease progression (Appendix Table A2). We assumed that IDUs and non-IDUs experienced similar disease progression rates. For each compartment, we estimated the rate of death from HIV and other causes (Appendix Table A2).
Highly active antiretroviral therapy
Antiretroviral therapy for HIV includes the use of multiple antiretroviral drugs in combination, known as HAART, with the goal of completely suppressing HIV viral replication. We assumed that HAART would be given according to current treatment guidelines, as specified by the US Department of Health and Human Services [7,30–32]. In our model, HIV-infected individuals with a CD4 cell count below 350 cells/μl (i.e., symptomatic HIV or AIDS) were eligible to receive HAART [7,30–32]. The primary advantages of HAART include a reduction in viral load – and thus, infectivity – and a delay in the progression of HIV infection [7,32]. We estimated that suppressive HAART reduced the annual probability of HIV transmission per sexual partnership by 90% (Table 1) [7–10]. The degree to which HAART reduces transmission from shared needle use is unknown; in our base-case analyses, we assumed the reduction was 50%, substantially less than for sexual transmission due to the probable higher efficiency of transmission from shared needles (Table 1) . Based on a model of the natural history of HIV and the effectiveness of HAART, we estimated that HAART reduced the annual AIDS-related death rate by 20%, and led to a six-fold reduction in the progression rate from symptomatic HIV to AIDS (Appendix Table A2).
In 2005, virtually no HIV-infected IDUs and approximately 1% of HIV-infected non-IDUs received HAART in Russia. We considered four incremental HAART programs: 80% of infected, treatment-eligible IDUs and 1% of infected, treatment-eligible non-IDUs receive HAART (IDU targeted treatment strategy); no infected IDUs and 80% of infected, treatment-eligible non-IDUs receive HAART (non-IDU targeted treatment strategy); 50% of all infected, treatment-eligible IDUs and non-IDUs receive HAART (untargeted treatment strategy); 80% of all infected, treatment-eligible IDUs and non-IDUs receive HAART (optimistic untargeted treatment strategy). In order to account for incomplete accessibility and imperfect adherence to HAART programs, we estimated that at most 80% of infected individuals could be treated with HAART.
Health outcomes and costs
For each intervention, we estimated quality-adjusted life years (QALYs) gained and costs incurred over a 20-year time horizon, as well as all future lifetime healthcare costs and QALYs, discounted to the present at 3% annually. For infected individuals, quality of life decreased with HIV disease progression (Appendix Table A2). We assumed that IDUs have a lower quality of life than non-IDUs by a (multiplicative) factor of 0.9 (Appendix Table A2).
We estimated the annual per person cost of HAART in St. Petersburg to be US$ 1950, based on prices that have been recently negotiated in Russia [33,34]. For IDUs who received HAART, we included an additional annual cost of US$ 500 for outreach, counseling, and adherence interventions. For all individuals with HIV, we estimated the annual non-HAART healthcare costs to be US$ 570 . All individuals, whether HIV infected or not, incurred an annual healthcare cost of US$ 115 .
HIV infections prevented
We estimated the resource requirements for each treatment strategy, measured by number of people treated over the 20-year time horizon, and calculated the (undiscounted) number of HIV infections prevented over the 20-year time horizon (Table 2; Fig. 1). The optimistic untargeted treatment strategy treated the most people (84 100) and prevented the most infections (11 000 among IDUs and 36 500 among non-IDUs). The IDU targeted treatment strategy treated 50 200 people and prevented 40 000 infections, including 30 000 infections among non-IDUs, due to reduced sexual transmission from IDUs to non-IDUs. The untargeted treatment strategy treated 57 300 people and prevented 30 800 infections (6800 among IDUs and 24 000 among non-IDUs). The non-IDU targeted treatment strategy treated 47 000 people and prevented the fewest infections (9500), almost all among non-IDUs.
With no incremental HAART programs, we estimated that HIV prevalence among IDUs would reach 63.6% after 20 years (Fig. 2a). Strategies that included treatment of IDUs had only a modest effect on prevalence among IDUs: even though these strategies prevented many infections among IDUs, HIV-infected IDUs lived longer, which attenuated reductions in prevalence.
With no incremental HAART programs, we estimated HIV prevalence among non-IDUs to be 1.7% after 20 years (Fig. 2b). All treatment programs that included IDUs reduced HIV prevalence among non-IDUs (to 1.1% with the IDU targeted treatment strategy, to 1.6% with the untargeted treatment strategy, and to 1.4% with the optimistic untargeted treatment strategy) (Fig. 2b). If the non-IDU targeted treatment strategy were implemented, the prevalence of HIV after 20 years increased to 2.1%. This seemingly paradoxical result occurred because relatively few infections were prevented and each infected individual lived longer; the net effect was an increase in prevalence. Strategies that included treatment of IDUs reduced prevalence among non-IDUs because the reduction in prevalence from infections averted outweighed increases in prevalence due to infected individuals living longer.
We calibrated the results of our model by comparing the model's estimates of HIV prevalence with observed HIV prevalence in St. Petersburg. The number of registered HIV cases in Russia is thought to underestimate actual HIV prevalence, and estimates of prevalence vary substantially [2–4]. In addition, empirical evidence about longitudinal changes in prevalence in specific population groups is sparse. However, HIV prevalence as calculated by the model is consistent with estimated HIV prevalence in Russia and trends observed in IDUs and non-IDUs.
We evaluated the cost, effectiveness, and cost-effectiveness of each treatment strategy (Table 2 and Fig. 3). We report the cost-effectiveness ratios of each strategy relative to the status quo, rather than to the next best alternative, because some of the strategies may not be feasible or considered appropriate (e.g., targeting resources exclusively to IDUs).
The least effective strategy was to target antiretroviral treatment only to non-IDUs. The most effective strategy was to treat as many HIV-infected people as possible (optimistic untargeted treatment strategy); this strategy was also the most expensive. The IDU targeted treatment strategy and the untargeted treatment strategy had similar effectiveness, but the IDU targeted treatment strategy cost substantially less. Of note, the IDU targeted treatment strategy was both more effective and less expensive than the non-IDU targeted treatment strategy.
The cost-effectiveness of expanded antiretroviral treatment relative to the current level of treatment ranged from US$ 1500 to US$ 2600 per QALY gained, depending on the strategy (Table 2, Fig. 3). Allocation of treatment resources to IDUs was the most cost-effective, and allocation of resources exclusively to non-IDUs was the least cost-effective. After eliminating the IDU targeted treatment strategy, which may be considered inappropriate, the incremental cost-effectiveness ratio of the best strategy (optimistic untargeted treatment strategy) to the next best strategy (untargeted treatment strategy) is US$ 1729.
We performed sensitivity analysis over estimated data ranges for all parameters (Appendix Tables A2 and A3). The number of HIV infections averted over 20 years was sensitive to four key parameters: sexual transmission infectivity, HAART-related reductions in infectivity, risky injecting behavior, and risky sexual behavior (Appendix Table A3).
In our model, non-IDUs can only be infected via sexual contact. We evaluated the effect on infections averted if the annual probability of sexual transmission per partnership was moderately higher or lower. As in the base case, the IDU targeted treatment strategy prevented more HIV infections and was more cost-effective than was the non-IDU targeted treatment strategy.
We conducted sensitivity analysis on the effectiveness of HAART at reducing HIV transmission from injection drug use and sexual contact. If HAART reduced injection infectivity by only 10% (instead of 50% as assumed in the base case), the IDU targeted treatment strategy prevented 30 114 infections, whereas the non-IDU targeted treatment strategy prevented 9463 infections. Conversely, if HAART reduced sexual infectivity by only 50% (instead of 90% as assumed in the base case), the IDU targeted treatment strategy prevented 18 691 infections, in comparison with 2732 infections prevented by the non-IDU targeted treatment strategy. If HAART was less effective at reducing both injection infectivity (10% reduction) and sexual infectivity (50% reduction), the IDU targeted treatment strategy prevented 9021 infections, more than three times the number of infections prevented by the non-IDU targeted treatment strategy. In all cases, the overall pattern of our findings remained unchanged: the IDU targeted treatment strategy was substantially more effective at preventing infections among both IDUs and non-IDUs than the non-IDU targeted treatment strategy. Additionally, the IDU targeted treatment strategy was more cost-effective than the non-IDU targeted treatment strategy, regardless of the effectiveness of HAART in reducing infectivity.
Comprehensive data on risky injecting and sexual behavior in Russia is very limited. We performed two-way sensitivity analysis on the annual number of injections per year and the fraction of injections that involve a shared needle. We also conducted sensitivity analysis on the annual number of sexual partnerships for IDUs and non-IDUs. Variations in risky injection behavior led to greater variation in infections prevented among IDUs; variations in risky sexual behavior led to greater variation in infections prevented among non-IDUs.
The cost of HAART is expected to decrease three-fold in Russia over the next 5 years, as Russian officials have negotiated lower prices from several drug manufacturers [1,34]. We estimated the incremental cost-effectiveness ratio if the annual cost of HAART, including counseling and adherence services, was significantly greater or less than assumed in the base case. In both cases, the IDU targeted treatment strategy was almost twice as cost-effective as the non-IDU targeted treatment strategy (Appendix Table A3).
Because St. Petersburg has relatively high HIV prevalence, we also estimated the effect of HAART programs in a low-prevalence region, using epidemiologic and behavioral data for the city of Barnaul. To ensure comparability with prior analyses, we assumed treatment levels similar to St. Petersburg, although actual treatment rates may be higher in Barnaul. The results for Barnaul (Appendix Table A4) are qualitatively similar to those for St. Petersburg and emphasize the benefits of providing IDUs and non-IDUs with access to HAART. The IDU targeted treatment strategy averted more infections and was more cost-effective than the other treatment strategies, and the IDU targeted treatment strategy generated more QALYs at less total cost than did the non-IDU targeted treatment strategy.
We developed an epidemic model of HIV in Russia to assess the health outcomes and costs associated with strategies to expand antiretroviral use. Our analysis indicated that expanded use of antiretroviral therapy, if appropriately implemented, could dramatically reduce HIV incidence among the general population in Russia, would result in enormous population-wide health benefits, and would be economically efficient. Our analysis also showed that the strategy of focusing treatment resources almost exclusively on non-IDUs, which is the current strategy in Russia, provided the least health benefit, was the least economically efficient, and will likely fail to slow the spread of HIV among IDUs or the general population.
We assessed strategies that focus on the IDU population to understand the significance of this mode of transmission in the overall epidemic. The importance of treatment for IDUs is highlighted by our finding that targeting HAART exclusively to IDUs prevented more HIV infections among non-IDUs than did targeting HAART exclusively to non-IDUs. Targeting treatment to IDUs prevented four times as many infections as treating only non-IDUs, and three-quarters of the prevented infections were among non-IDUs. Nonetheless, we neither advocate, nor consider appropriate, strategies that exclusively treat IDUs. Rather, our findings emphasize the critical need to include plans to treat both IDUs and non-IDUs as antiretroviral treatment is expanded in Russia.
A key question is whether providing antiretroviral treatment to IDUs in Russia is feasible, and whether adherence with treatment in these groups would be sufficient to obtain the benefits estimated in our analysis. Our analyses assumed that delivery of HAART to IDUs would require more intensive ancillary support, and we included an additional annual cost of US$ 500 per IDU for services such as treatment for substance abuse, interventions to improve adherence, and social services. Treatment for substance abuse and harm reduction programs are important adjuncts to provision of HAART, and can include substitution therapy, needle exchange, and safer injection sites. Rcent studies in other settings have found that both IDUs and non-IDUs who utilize counseling and support services exhibit similar adherence and drug resistance to all major antiretroviral treatment regimens [11,37]. In addition, substitution therapy can improve adherence and therapeutic response to HAART . Experience with treatment and harm reduction programs in Russia is limited, so empirical evidence about their feasibility in Russia is lacking. Successful implementation of such programs would require changes in national law regarding substitution therapy, and consistent regional implementation of recent changes in laws intended to increase access to harm reduction and treatment.
The cost-effectiveness of expanded use of antiretroviral therapy in Russia is favorable when judged by criteria developed by the World Health Organization (WHO) . WHO criteria consider a health intervention cost-effective if it costs less than three times per capita gross domestic product (GDP). GDP in Russia was US$ 9800 in 2004, and thus treatment with antiretroviral therapy meets WHO guidelines. However, interpretation of the cost-effectiveness of interventions in Russia is challenging because very few interventions have been evaluated with cost-effectiveness analysis . The cost-effectiveness of all treatment strategies will improve substantially if the cost of antiretroviral therapy decreases further. Finally, the cost-effectiveness of a strategy should be viewed in the broader context of other important factors such as equity. Although therapy targeted to IDUs is the most cost-effective strategy, we consider it to fail the criterion of equity, leaving untargeted therapy as the most efficient approach.
The effect of expanded antiretroviral use on the course of the HIV epidemic depends on the degree to which HIV transmission by sex and injection drug use is reduced. This reduction depends on the reduction in viral load from HAART and on baseline sexual and needle-sharing behaviors. Reductions in sexual transmission from use of HAART have been well documented but a reduction from decreased transmission by needle sharing, although reasonable, is speculative [7–10]. However, even if we assumed that HAART reduced transmission via needle sharing by only 10%, the overall conclusions of our analysis remained unchanged.
The epidemiology of HIV varies substantially in different geographic locations within Russia. Our primary analyses used data from St. Petersburg, which has a relatively high HIV prevalence among IDUs and commercial sex workers. To understand whether our results could be generalized to other regions, we also evaluated the costs and benefits of expanded antiretroviral therapy in Barnaul. Although Barnaul has a substantial population of IDUs, HIV prevalence among IDUs remains less than 2%, for reasons that are not fully understood. The general pattern of our findings remained the same in Barnaul, with the exception that the IDU targeted treatment strategy averted the most infections.
In summary, our analyses demonstrated quantitatively that slowing the spread of HIV in St. Petersburg, Russia is inextricably linked to antiretroviral treatment for the IDU population. Successful provision of antiretroviral therapy will depend on concomitant treatment of substance abuse. Treatment of substance abuse in Russia presents many challenges. Health care resources are highly constrained, harm reduction programs are uncommon, and substitution therapy with methadone is illegal. Plans to expand antiretroviral therapy in St. Petersburg provide the potential for enormous health benefits. However, if the full potential of this investment in health is to be realized, antiretroviral treatment must reach both the IDU and non-IDU populations.
We thank the investigators and clinicians in Russia who provided data for our analyses.
Sponsorship: This work was supported by a grant from the National Institute on Drug Abuse, National Institutes of Health (R-01-DA-15612), and by the Department of Veterans Affairs.
1. World Health Organization (WHO). Summary Country Profile for HIV/AIDS Treatment Scale-up: Russian Federation, June 2005
. Geneva: World Health Organization; 2005.
2. AIDS Foundation East West. Officially Registered HIV Cases in the Russian Federation: 1 January 1987 through 2 May 2005
. Moscow, Russia: AIDS Foundation East West; 2005.
3. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2004 Report on the Global AIDS Epidemic
. Geneva, Switzerland: United Nations; 2004.
4. Feshbach M, Galvin C. HIV/AIDS in Russia – an Analysis of Statistics. Washington, DC: Woodrow Wilson International Center for Scholars; 2005.
5. Rhodes T, Sarang A, Bobrik A, Bobkov E, Platt L. HIV transmission and HIV prevention associated with injecting drug use in the Russian Federation. Int J Drug Policy 2004; 15:1–16.
6. Samoilov D. Double discrimination: drug users living with HIV/AIDS. HIV AIDS Policy Law Rev 2004; 9:83–85.
7. Sanders GD, Bayoumi AM, Sundaram V, Bilir SP, Neukermans CP, Rydzak CE, et al
. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. N Engl J Med 2005; 352:570–585.
8. Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al
. Viral load and heterosexual transmission of human immunodeficiency virus type 1: Rakai Project Study Group. N Engl J Med 2000; 342:921–929.
9. The National Institute of Mental Health (NIMH). The NIMH Multisite HIV Prevention Trial: reducing HIV sexual risk behavior. Science
10. Kamb ML, Fishbein M, Douglas JM, Rhodes F, Rogers J, Bolan G, et al
. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases: a randomized controlled trial. JAMA 1998; 280:1161–1167.
11. Open Society Institute. Breaking Down Barriers: Lessons on Providing HIV Treatment to Injection Drug Users
. New York, NY: Open Society Institute; 2004.
12. Beck EJ, Mandalia S, Gaudreault M, Brewer C, Zowall H, Gilmore N, et al
. The cost-effectiveness of highly active antiretroviral therapy, Canada 1991–2001. AIDS 2004; 18:2411–2418.
13. Schackman BR, Freedberg KA, Weinstein MC, Sax PE, Losina E, Zhang H, et al
. Cost-effectiveness implications of the timing of antiretroviral therapy in HIV-infected adults. Arch Intern Med 2002; 162:2478–2486.
14. May RM, Anderson RM. Transmission dynamics of HIV infection. Nature 1987; 326:137–142.
15. Zaric GS, Barnett PG, Brandeau ML. HIV transmission and the cost-effectiveness of methadone maintenance. Am J Public Health 2000; 90:1100–1111.
16. Zaric GS, Brandeau ML, Barnett PG. Methadone maintenance and HIV prevention: a cost-effectiveness analysis. Manage Sci 2000; 46:1013–1031.
17. Shaboltas AV, Verevochkin SV, Hoffman I, Raider R, Perdyu TZ, Kozlov AP. Organisatsiia kogortnogo issledovaniia i otsenka zarazhennosti i zarazhaemosti VICH sredi in'ektsionnykh narkopotrebitelei v Sankt-Peterburge [Development of cohort research and assessment of HIV infection rate and susceptibility to HIV infection among IDUs in St Petersburg]. Russkii Zhurnal VICH/SPID i Rodstvennye Problemy [Russian Journal HIV/AIDS and Related Problems] 2004; 8:20–28.
18. Smolskaya T, Tretiakova V, Ostrovski D, Suvorova S, Ogurtcova S. Three years of experience on HIV control in St. Petersburg (1998–2001) among IDUs. International AIDS Conference
, Barcelona, Spain, 2002 [abstract WePeC6119].
19. Rakhmanova A, Volkova G, Vinogradova E, Alsterlund R, Moestrup T, Andersson M, et al. Harm Reduction (Needle Exchange) Program Drop-in Center & Bus for HIV/AIDS prevention. Epidemiology situation on HIV/AIDS in St. Petersburg. The project Malmo, Sweden – St. Petersburg, Russia for drug users (1999–2002)
. AIDS, Statistics, Health
20. Sultanov LV. HIV statistics in Altai Krai. Barnaul, Russia: AIDS Center, Barnaul, Altai Krai; 2005.
21. Smolskaya T, Rusakova M, Tsekhanovich A, Yakovleva A, Tretyakova V, Piskarev I. Sentinel seroepidemiological and behavioural surveillance among female commercial sex workers in Saint Petersburg (Russian Federation) in 2003
. XV International AIDS Conference
. Bangkok, Thailand, 2004 [abstract ThOrC1371].
22. Kaplan EH, Heimer R. A model-based estimate of HIV infectivity via needle sharing. J Acquir Immune Defic Syndr 1992; 5:1116–1118.
23. Somlai AM, Kelly JA, Benotsch E, Gore-Felton C, Ostrovski D, McAuliffe T, et al
. Characteristics and predictors of HIV risk behaviours among injection drug using men and women in St. Petersburg, Russia. AIDS Educ Prev 2002; 14:295–305.
24. Altai Krai AIDS Center, Altai State University, DAR Research Center (Russian Federation Ministry of Education), UNICEF. Povedenie naseleniia svyazannoe s riskom VICH-infitsirovaniia: Otchet ob issledovanii i metodicheskie rekomendatsii
. [HIV Risk Behaviour of Population: Research Report and Methodological Recommendations.
]. Barnaul, Russia: Altai Krai AIDS Center; 2003.
25. Grassly NC, Lowndes CM, Rhodes T, Judd A, Renton A, Garnett GP. Modelling emerging HIV epidemics: the role of injecting drug use and sexual transmission in the Russian Federation. China and India Int J Drug Policy 2003; 14:25–43.
26. Amirkhanian YA, Kelly JA, Issayev DD. AIDS knowledge, attitudes and behaviour in Russia: results of a population-based, random-digit telephone survey in St Petersburg. Int J STD AIDS 2001; 12:50–57.
27. Watters JK. Trends in risk behavior and HIV seroprevalence in heterosexual injection drug users in San Francisco. J Acquir Immune Defic Syndr 1994; 7:1276–1281.
28. Laumann EO, Gagnon JH, Michael RT, Michaels S. The Social Organization of Sexuality: Sexual Practices in the United States. Chicago, IL: University of Chicago Press; 1994.
29. Pinkerton SD, Abramson PR. Effectiveness of condoms in preventing HIV transmission. Soc Sci Med 1997; 44:1303–1312.
30. Panel on Clinical Practices for Treatment of HIV Infection. Guidelines for the Use of Antiretroviral Agents in HIV-1-infected Adults and Adolescents
. Bethesda, MD: United States Department of Health and Human Services; 2005.
31. Yeni PG, Hammer SM, Hirsch MS, Saag MS, Schechter M, Carpenter CC, et al
. Treatment for adult HIV infection: 2004 recommendations of the International AIDS Society-USA Panel. JAMA 2004; 292:266–268.
32. Erb P, Battegay M, Zimmerli W, Rickenbach M, Egger M. Effect of antiretroviral therapy on viral load, CD4 cell count, and progression to acquired immunodeficiency syndrome in a community human immunodeficiency virus-infected cohort Swiss HIV Cohort Study. Arch Intern Med 2000; 160:1134–1140.
33. Bobrik A. Expected Price of the Cheapest ARV Regimens. Moscow, Russia: Open Health Institute; 2005.
34. Boykewich S. Supply of anti-HIV drugs falls short
. Moscow Times
2005: Sect. 1.
35. Adeyi O, Baris E, Chakraborty S, Novotny T, Pavis R. Averting AIDS crises in Eastern Europe and Central Asia. Washington, DC: The World Bank; 2003.
36. World Health Organization (WHO). Per Capita Total Expenditure on Health at Average Exchange Rate (US$)
. Geneva, Switzerland: World Health Organization; 2001.
37. Wood E, Hogg RS, Yip B, Dong WW, Wynhoven B, Mo T, et al
. Rates of antiretroviral resistance among HIV-infected patients with and without a history of injection drug use. AIDS 2005; 19:1189–1195.
38. Palepu A, Tyndall MW, Joy R, Kerr T, Wood E, Press N, et al
. Antiretroviral adherence and HIV treatment outcomes among HIV/HCV co-infected injection drug users: The role of methadone maintenance therapy. Drug Alcohol Depend 2006; 84:188–194.
39. World Health Organization (WHO). The World Health Report 2002 – Reducing Risks, Promoting Healthy Life
. Geneva, Switzerland: WHO; 2002.
40. Bobrik A. HIV prevention among drug users in Russian cities: cost-effectiveness analysis. Rossiskii Semenii Vrach [Russian Family Doctor] 2002; 6:30–36.
© 2006 Lippincott Williams & Wilkins, Inc.