Summary
The HIV epidemic in Estonia is rapidly expanding, and injection drug users (IDUs) are the major risk group contributing to the expansion. A convenience sample of 159 IDUs visiting syringe-exchange programmes (SEPs) was selected to quantify the association of HIV-risk behaviours and blood-borne infections. A high prevalence of HIV, hepatitis B core antibody (HBVcore), hepatitis B surface antigen (HbsAg) and hepatitis C virus antibodies (56, 85.1, 21.3, and 96.2%, respectively) was associated with high-risk injections, unsafe sexual behaviour and alcohol abuse. These findings emphasize the importance of evidence-based secondary prevention among the HIV-infected, especially given the uncertain sustainability of antiretroviral and substance abuse treatments.
Keywords: injection drug use, HIV, HBV, HCV, high-risk behaviour, Estonia
INTRODUCTION
In the third decade of the AIDS era, new HIV epidemics continue to emerge. The Newly Independent States (NIS) of the former Soviet Union have undergone tremendous socio-political upheaval, and several related health epidemics have emerged. Morbidity and mortality rates have risen1 due to increased violence, high-risk sexual behaviour, substance abuse and infectious diseases including HIV, viral hepatitis, sexually transmitted diseases (STDs) and tuberculosis.2 An epicentre of the health crisis appears to be in Estonia. This Baltic country has a rapidly expanding HIV/AIDS epidemic, and the highest reported incidence and prevalence of HIV (1.3%) in the European Region.3,4 This epidemic is in large part due to injection drug use, a problem common to many countries in the region.5–7
The primary objective of this study was to quantify HIV risk behaviours among injection drug users (IDUs) visiting syringe-exchange programmes (SEPs) and recommend evidence-based strategies for secondary prevention. The second objective was to determine the prevalence of HIV infection among this population of SEP attendees. Additionally, we collected bridging information on self-reported behaviours to determine the risk of transmitting HIV, hepatitis C virus (HCV) and hepatitis B virus (HBV) from IDUs to others.
METHODS
Study settings: SEPs
Study activities took place at two SEPs in Tallinn:
SITE 1, SEP of NGO Convictus, located in the city centre. In 2004, this programme had over 13,500 visits, and over 92,800 sterile syringes were distributed. This SEP is open daily.
SITE 2, SEP of NGO AIDS Support Center, located in the northern part of the city. In 2004, this programme had over 4100 visits, and over 33,000 sterile syringes were distributed. This SEP is open daily except Sunday.
Respondents
Between February and August 2004, 200 SEP clients were approached for study participation, and 159 (80%) were enrolled as study subjects. This convenience sample of current IDUs comprised 134 men and 25 women, all over 18 years of age. All participation was anonymous. When potential participants were approached, the study was briefly explained in Estonian or Russian, and the person was screened for eligibility criteria. Those already participating at the other study site were eliminated by cross-checking birth date, gender and residential zip code.
Procedures
After providing informed consent, all participants were asked to complete an interviewer-administered questionnaire. All participants provided blood samples for HIV, hepatitis B and C, and syphilis testing. At end of study participation, the respondents were offered a direct referral to voluntary HIV counselling and testing (same location at dedicated time). After completing study activities, participants received a food coupon worth approximately €7.
We developed a survey based on the Family Health International Questionnaire for IDUs.8 Items included in the questionnaire were developed in English and modified with the input of Estonian IDU harm reduction service professionals. The questions were translated into Estonian and Russian, and piloted on a sample of 10 SEP customers. The four-item CAGE questionnaire was added to assess the prevalence of alcohol-related problems.9
All study procedures were approved by the Institutional Review Board at SUNY Downstate Medical Center and by the Ethics Board at the University of Tartu.
Laboratory testing
The following commercially available kits were used to test for HIV, hepatitis B and hepatitis C infection markers: HBsAg (V2), CORE (anti-HBc), HCV version 3.0 and HIV-1/HIV-2 III Plus from Abbott Laboratories (Abbott Park, IL, USA). Non-treponemal rapid plasma reagin test (RPR) for syphilis testing was used (RPR, Omega Diagnostics Ltd, Alva, Scotland, United Kingdom).
Statistical analysis
Descriptive statistics, such as mean, median, standard deviation (SD) and/or range, were used for continuous variables in the study. Categorical variables included relative (%) and absolute (n) frequencies. The Mann–Whitney two-sample test for group comparisons was used for continuous variables and Fisher’s exact test was used for categorical variables. Statistical analyses were performed with R 2.0.1 for Windows (www.r-project.org) and each used the conventional 5% significance level.
RESULTS
Participant characteristics
Although almost all respondents (88. 9%) were born in Estonia, 92.6% were non-ethnic Estonians (82.2% ethnic Russians, 10.5% members of other Russian speaking ethnic groups). Two-thirds (67%) were less than 25 years and the majority (70.4%) of participants had never been married. About a quarter of the respondents were students (25.3%), while the remainder were almost evenly split between the 39.5% employed (21.6% full time, 17.9% part time) and 41.4% neither working nor studying. Close to the half of the respondents (42.7%) reported no monthly income and 6.7% earned less than the state-estimated subsistence minimum (approximately €93). Only 3.3% of participants reported income in excess of the average monthly wage (approximately €450).10
Substance abuse
Initiation, frequency and sharing behaviour
The age at initiation of any illicit drug use among the respondents ranged from seven to 30 years. The mean age at initiation of the particular illicit drug and the frequencies of their use are presented in Table 1. Of those who reported marijuana use, 13.4% began to inject drugs within one year after starting smoking marijuana, and the majority (77.7%) did so within two years. At their last injection, the majority (85.8%, n = 139) reported using a sterile needle for injection. The sources for obtaining sterile/new syringes for participants were reported as follows: SEP (59%), pharmacy (28%) and shooting gallery (7.4%). Ethnic Estonians reported significantly more cocaine use (P = 0.01), but such use was not associated with higher income (P =0.9).
Table 1.
The mean age at initiation of the particular illicit drug and the frequencies of their use
| Drug | Marijuana | Cocaine* | Heroin | Street methadone | Home-made opiates | Amphetamines |
|---|---|---|---|---|---|---|
| Used at the last injecting episode No. (%)†,* | NA | 1 (0.6%) | 96 (59.3%) | 1 (0.6%) | 5 (3.1%) | 50 (30.9%) |
| No. (%) of respondents who have used a particular drug | 156 (96.3%) | 64 (39.5%) | 157 (96.6%) | 14 (8.9%) | 118 (72.8%) | 76 (64%) |
| Age at initiation (mean, range) | 15.5 (7–30) | 20 (13–33) | 18.9 (10–47) | 21 (13–35) | 17.8 (10–30) | 19 (10–43) |
| Proportion of frequent users No‡ (%) | 51 (32.5%) | 1 (0.6%) | 107 (68.2%) | 3 (21%) | 41 (35%) | NA |
Cocaine, any means of administration
Twenty-five (15.4%) respondents reported fentanyl use at the last injecting episode
Frequent use: from several times a week to repeated daily use over past three months
Sexual behaviour
Seven respondents were sexually inexperienced, while 13.8% reported having over 50 lifetime sexual partners (12% of men, 19% of women, P =0.44). Within the past year, more than one-third (36.1%) of the IDUs reported one sexual partner, 21.5% two to four partners, and 25.7% five or more partners.
Of the 137 males, 88.9% reported having sex with only female partners, 2.2% reported bisexual behaviour, and 8.9% (n =12) refused to answer this question. Of those sexually active, 52.5% used some method to protect against either pregnancy or STD; male condom was used most frequently (50%), followed by birth control pills (8.6%).
Approximately one-sixth of respondents (16.6%, n =26) reported receiving money for sex (72% of women, 6% of men, P<0.001), and twice as many (34%, all male respondents) paid for sex.
Close to one-third (27%, n =44) of the sample reported having been diagnosed by a doctor with an STD at some point in their lives, and 15.4% (n =25) had an STD diagnosed during the past year.
Bridging behaviour
With regard to sexual networks, 19.6% of sexually active respondents reported having sexual encounters with other IDUs exclusively, while 44.9% reported sex with non-IDUs exclusively. Overall, 77.4% of those sampled had non-injecting sexual partners. The ethnic composition of sexual relationships was mixed insofar as only 52% of Russians reported always having sex with a partner of the same ethnicity. With regard to injecting networks, 70.7% of the Russian IDUs were injecting with Russians only. Estonian IDUs, in contrast, always or almost always injected with Russians.
Travel
Slightly more than half (52%) of the respondents reported travel outside of Estonia, averaging two trips in the past year. The most frequent destinations were the Russian Federation (n =36) and other states of the former Soviet Union (n =5), followed by neighbouring countries in Scandinavia (Finland and Sweden, n =19) and the Baltics (Latvia and Lithuania, n =10). Although the frequency of trips did not differ by gender, the destination did. Women were more likely to travel into Scandinavian countries (P =0.01) while men visited the former Soviet Union (P =0.006).
Health status, access to health care and HIV testing
Of those participating, 67% judged their own health status to be less than good (fair 47.5%, poor 16.7%). While 57.4% (n =93) of the respondents had been covered by state social health insurance one year prior to the date of their interview, only 38.9% (n =63) were health insurance beneficiaries at the time of the interview.
Of those who had ever donated blood (9.2%, n =15), few did so within the past two years (n =2), and most (60%, n =9) did so to find out their HIV status. Of the total sample, 49.3% (n =80) had been previously tested for HIV, and 20.9% (n =34) reported being tested positive for HIV.
HIV prevalence among IDUs
The HIV prevalence among respondents was 56% (n =89). Bivariate relationships between the selected variables and HIV prevalence are presented in Table 2. There was a significant association between HIV prevalence and ethnicity, with non-ethnic Estonian IDUs more likely to be HIV positive (P =0.03). There were no significant associations between HIV infection and injection drug use habits, with one exception, duration of heroin use. HIV-positive respondents generally reported longer heroin injection careers. Neither duration of sexual activity nor number of sexual partners was significantly correlated with HIV infection.
Table 2.
HIV seroprevalence by selected variables
| The whole sample | Group variable | HIV+ (No.=89) | HIV− (No.=70) | P value |
|---|---|---|---|---|
| 23 (SD 6.1) | Age (years, mean, SD) | 23.6 (SD 5.2) | 24.8 (SD 7.2) | NS |
| 15.4% (25) | Gender (%, No. of women) | 15.9 % (14) | 15.7 % (11) | NS |
| 92.6% (147) | Ethnicity (%, No. of non-ethnic Estonians)* | 96.6% (86) | 87.1% (61) | P=0.03 |
| Drug using behaviour | ||||
| 6 (0–30) | Duration of injection home-made opiates use, years (median, range) | 6 (0–29) | 5 (0–30) | NS |
| 5 (0–41) | Duration of injection heroin use, years (median, range) | 5 (0–41) | 4 (0–17) | P=0.02 |
| 39.5% (63) | Syringe sharing within past three months (%, No.) | 42.7% (38) | 35.7% (25) | NS |
| 8.8%, always | ||||
| 46.3%, ohalf of occasions | ||||
| 23.8%, never | ||||
| Alcohol use | ||||
| 59.3% (81) | Clinically significant CAGE score (%, No.) | 52.8% (47) | 48.6% (34) | NS |
| 12.5 | Age at initiation of alcohol use (years, mean) | 13 | 12.5 | NS |
| Sexual behaviour | ||||
| 15 (9–22) | Age at first sexual encounter (years, range) | 15 (9–22) | 15 (11–21) | NS |
| 10 (0–1600) | No of lifetime sexual partners (median, range) | 10 (0–1600) | 12.5 (0–1000) | NS |
| 1 (0–300) | No of last year sexual partners (median, range) | 1 (0–60) | 1 (0–300) | NS |
| 19.7% (32) | Marginalization (not working or studying* without medical insurance either 2004 or 2003) (%, No.) | 21.3% (19) | 18.6% (13) | NS |
| 49.3% (80) | Ever been tested for HIV (%, No.) | 47.2% (42) | 52.9% (37) | NS |
HIV seroprevalence: 25.0% (3/12) among ethnic Estonians, 58.8% (77/131) among ethnic Russians, and 56.3% (9/16) among other ethnic groups (all Russian speaking)
Prevalence of other infections
Hepatitis serologies were as follows: 131 of 154 those tested (85.1%) were positive for antibodies to HBV core antigen, 33 of 155 serums tested (21.3%) were positive for HBV surface antigen and 153 of 159 serums tested (96.2 %) were positive for antibodies to HCV. Four (2.5%, n =159) sera were positive for syphilis by RPR testing. Of the four people with positive syphilis serology, three were HIV infected. HIV and HBV infections were associated: 96.4% of individuals infected with HIV were positive for HBV core antibodies in comparison with 71.4% of the HIV-negative participants (P<0.001).
DISCUSSION
This is the first assessment of correlation between high-risk sexual and injecting behaviours and prevalence of blood-borne infections among Estonian IDUs. This seminal study is particularly relevant given the recent expansion of HIV in this region.11 As has been described in other reports on IDUs from Estonia,10,12 the vast majority of IDUs recruited in Tallinn were non-ethnic Estonians (mostly ethnic Russians). Elevated rates of high-risk behaviour and greater HIV infection among ethnic minority drug users have been noted in multiple settings.13–15 In Estonia, the non-Estonian ethnic minority (95% Russian speaking) comprises 30% of the total population, and the majority (83%) of the Russian speaking populations live in the Northeast or in the capital area of the county.16 The HIV seroprevalence in our sample was 25.0% among ethnic Estonians, 58.8% among ethnic Russians, and 56.3% among other ethnic groups (all Russian speaking). Non-ethnic Estonian ancestry and heroin use were the only statistically significant factors related to HIV seropositivity. Notably, none of the following risk factors appeared significant: age, gender, length of injection drug or alcohol abuse, injection equipment sharing and sexual high-risk behaviours.
Employment for the Russian-speaking population was largely concentrated in segments of the economy that collapsed after the restoration of Estonia’s independence, deepening the problem of unemployment17 and social marginalization. Geographical proximity of the Estonian North-East county (the highest HIV/IDU levels in Estonia) to St Petersburg (Russian Federation) – an area with high rates of HIV/IDU18 – could be an important confounding factor for the observed correlation of HIV/IDU with ethnic origin. Nevertheless, ethnically discrepant patterns have emerged in drugs used (cocaine and amphetamines are preferred to heroin and opiates by ethnic Estonians), modes of drug use (preferably non-injecting drug use by ethnic Estonians) and initiation of drug use (non-ethnic Estonians are more likely to start their drug using career with injecting).19 Additional sociological studies to investigate the association between ethnicity and higher disease prevalence are clearly needed.
Our brief descriptive study does allow us to highlight the temporal hierarchy in substance abuse patterns among Estonian IDUs. Although our study revealed a sequential pattern that begins with tobacco and is followed by marijuana, home-made opiate injection and heroin injection in that order, these data neither support nor refute a gateway drug theory. Since all participants were IDUs, there is no causal evidence that initiation of any one or any particular series of psychoactive or addictive substances led to subsequent injection drug use.
Further study is needed to investigate the relationships among injection drug use, alcohol use and unsafe sex. Over a half of the population had a CAGE score indicating alcohol-related problems. Problem drinking is associated with an increased risk of STDs across a wide variety of populations.20 Our observation of the high level of alcohol abuse suggests that identification and treatment of alcohol abuse may be an important and desirable component of HIV prevention in Estonia. Further studies are needed to ascertain the precise relationships between these variables.
High rates of bacterial sexually transmitted infections among disenfranchised population groups in Moscow were recently reported by Shakarishvili et al.21 While the overall HIV infection prevalence was below 3% among those surveyed, the authors noted that the high prevalence of STDs in these groups may foreshadow a dramatic increase in HIV infections in the Russian Federation.21 In the current study, most participants reported being sexually active, with over one-fourth having more than five sexual partners within past 12 months, and one-third of the men paying for sex. All these factors are associated with high HIV prevalence and frequent symptomatic STDs. The finding that more than three-quarters of current IDUs report non-IDU sexual partners underscores the urgent need for sex risk reduction and prevention research in the IDU community. Further understanding the high-risk networks (sexual, drug using) of IDUs and between IDUs and the general population will be important for developing targeted HIV transmission prevention activities.
There are worrisome signs of ongoing marginalization of the young IDUs, as indicated by the high proportion of young persons neither working nor studying (41.4%) and a decreasing proportion of persons covered by state health insurance (57% in 2003, 39% in 2004). As has been suggested by Wallace, marginalization leads to the changes in lives of individuals and families, and also in the social construction of a neighbourhood. Youth behaviours such as doing well in school, getting a regular job, avoiding substance abuse and maintaining stable relationships become more difficult as the neighbourhood structures that value such attainments dissolve. Negative acts such as violent behaviour, multiple sexual conquests and drug taking are messages that tend to be more easily disseminated in a dissolving community.22,23
Last but not least, we observed a very high prevalence of HIV and other blood-borne infections among our respondents. Similarly, high HIV seroprevalence has been reported from other cities/regions of former Soviet Union.18,24 The high infection levels increase the urgency of secondary prevention. The extremely frequent co-infections among the HIV-infected IDUs will complicate the treatment for HIV, if and when it becomes available. Information on HIV prevalence within the IDU population in the other two Baltic countries (Lithuania and Latvia) is scarce, and warrants further exploration. According to the surveillance data, 71% of HIV cases reported from Latvia and 80% from Lithuania can be attributable to IDU.25
It is worth reviewing the limitations of the study. The nature of a cross-sectional study permits only prevalence measures and does not allow for incidence measures. Further, cross-sectional studies do not permit us to establish causal relationships or directions of causality. Another limitation of the study design is the use of a convenience sample instead of probability or adaptive sampling. However, convenience sampling is particularly well suited to studying populations without a sampling frame and which are hard to target, such as IDUs. We realize that the characteristics of IDUs can differ from site to site, and we do not suggest that our findings are entirely generalizable to other Estonian IDUs or even those IDUs visiting other Estonian SEPs.
Another limitation of the study is lack of confirmation for HIV enzyme immunoassay (EIA) testing. We used one test system (HIV-1/HIV-2 III Plus from Abbott Laboratories) to detect HIV-1/HIV-2 antibodies for study purposes. Given the high number of positive HIV test results and high cost, it was not feasible to perform confirmatory testing. However, the excellent characteristics of the test system used – both sensitivity and specificity close to 100%26 – and the high prevalence of HIV, suggest that the percentage of false-positive or -negative results were likely to be minimal and had no impact on the results of the study.
Despite these limitations, several inferences can be made. First, the problem of higher rates of HIV infection and injection drug use among racial/ethnic minorities may be a general pattern in many areas. Whether such higher rates can be reduced must be considered a critical question in the international epidemiology of HIV among IDUs. Second, as a summary of the current epidemic outlook in Estonia, our work demonstrates the intersection of several epidemics, including drug use, HIV and viral hepatitis. These epidemics are further characterized by high infection rates and the exceptionally young age of those affected. Third, the contribution of heavy alcohol use to amplify the course of epidemics could be substantial. Fourth, complicated questions for international prevention efforts are raised by the degree to which these young people frequently travel both to former Soviet Union states and to Western Europe (especially Scandinavian countries). The appropriate responses to these issues should be focused on scientifically validated, evidence-based approaches to secondary prevention among the HIV infected, especially given uncertainties surrounding the availability, cost and sustainability of antiretroviral and substance abuse treatments. This descriptive study of Estonian IDUs portends an enormous burden on health systems and economic productivity.
Acknowledgments
We thank the staff of NGOs AIDS Support Center and NGO Convictus Eesti for their valuable contribution. This study was supported by a grant no. 5526 by Estonian Science Foundation, and grant no. 3 D43 TW00233 from the Fogarty International Center, NIH, USA.
References
- 1.Leinsalu M, Vagero D, Kunst AE. Increasing ethnic differences in mortality in Estonia after the collapse of the Soviet Union. J Epidemiol Commun Health. 2004;58:583–9. doi: 10.1136/jech.2003.013755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dehne KL, Pokrovskiy V, Kobyshca Y, Schwartländer B. Update on epidemics of HIV and other sexually transmitted infections in the newly independent states of the former Soviet Union. AIDS. 2000;14(Suppl 3):S75–84. [PubMed] [Google Scholar]
- 3.AIDS Foundation East-West (AFEW) 2004 November 25; [ www.afew.org/english/index.php]
- 4.UNAIDS. 2006 report on the global AIDS epidemic. Overview of the global AIDS epidemic. :48. [ www.unaids-org/en/HIV_data/2006GlobalReport/default.asp]
- 5.Aceijas C, Stimson GV, Hickman M, Rhodes T. Global overview of injecting drug use and HIV infection among injecting drug users. AIDS. 2004;18:2295–303. doi: 10.1097/00002030-200411190-00010. [DOI] [PubMed] [Google Scholar]
- 6.Dehne KL, Khodakevich L, Hamers FF, Schwartlander B. The HIV/AIDS epidemicin Eastern Europe: recent patterns and trends and their implications for policy-making. AIDS. 1999;13:741–9. doi: 10.1097/00002030-199905070-00002. [DOI] [PubMed] [Google Scholar]
- 7.Uusküla A, Kalikova N, Zilmer K, Tammai L, DeHovitz J. The role of injecting drug use in the emergence of HIV in Estonia. Int J Infect Dis. 2002;6:23–7. doi: 10.1016/s1201-9712(02)90131-1. [DOI] [PubMed] [Google Scholar]
- 8.Amon J, Brown T, Hogle J, et al. Behavioral surveillance surveys: guidelines for repeated behavioral surveys in populations at risk of HIV. Questionnaire for injecting drug users. Family Health Int. 2000:263–86. [Google Scholar]
- 9.Mayfield D, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry. 1974;131:1121. doi: 10.1176/ajp.131.10.1121. [DOI] [PubMed] [Google Scholar]
- 10.Kalikova N. The HIV epidemic in Estonia. Proceedings of the 3rd Congress of the Estonian Society of Sexually Transmitted Infections (EUSTI); Pärnu, Estonia. 2001. pp. 14–15. [Google Scholar]
- 11.Priimägi L, Kremerman I, Tefanova V, Tallo T, Osadtsaja G. Study on hepatitis C and hepatitis B infected intravenous drug users. Eesti Arst. 1998;6:521–3. [Google Scholar]
- 12.Database of the Estonian Foundation for prevention of drug addiction [narko.sm.ee/Levik/2000/default.asp]
- 13.Craib KJP, Spittal PM, Wood E. Risk factors for elevated HIV incidence among aboriginal injection drug users in Vancouver. Can Med Assoc J. 2003;168:19–24. [PMC free article] [PubMed] [Google Scholar]
- 14.Beyrer C. Hidden epidemic of sexually transmitted diseases in China: crisis and opportunity. JAMA. 2003;289:1265–73. doi: 10.1001/jama.289.10.1303. [DOI] [PubMed] [Google Scholar]
- 15.Grund J, Verbraeck H, Ofner P, Marel O Del, Kamel Kas, Le Romes Durvar. An exploration of drug use and HIV risks among the Roma of Central and Eastern Europe. Paper presented at the 12th International Conference on the Reduction of Drug Related Harm; New Delhi, India. 1–5 April 2001. (God Hits Whom He Chooses; The Roma Gets hit Twice) [Google Scholar]
- 16.Statistical office of Estonia [pub.stat.ee/px-web.2001/I_Databas/Population/Population.asp] (27 December 2004)
- 17.Lauristin M, Vihalemm P. Return to the western world. In: Lauristin M, Vihalemm P, Rosengren E, Weibull, editors. Recent Historical Developments in Estonia: Three Stages of Transition (1987–1997) Estonia: Tartu University Press; 1997. pp. 79–83. [Google Scholar]
- 18.Shaboltas AV, Toussova O, Hoffman I, et al. HIV prevalence socio-demographic, and behavioural correlative and recruitment methods among injection drug users in St Petersburg, Russia. J Acquir Immune Defic Syndr. 2006;41:657–63. doi: 10.1097/01.qai.0000220166.56866.22. [DOI] [PubMed] [Google Scholar]
- 19.Magerova T, Dzalalov S. Narva rehabilitation center for drug addicts and alcoholics. Presented at the US CRDF funded workshop ‘Collaborative response to tackle HIV epidemic in Baltics’; Estonia, Tallinn. 04–06 August 2004. [Google Scholar]
- 20.Cook RL, Duncan CB. Is there an association between alcohol sexually transmitted diseases? A systematic review. Sex Transm Dis. 2005;32:156–64. doi: 10.1097/01.olq.0000151418.03899.97. [DOI] [PubMed] [Google Scholar]
- 21.Shakarishvili A, Dubovskaya LK, Zohrabyan LS, et al. Sex work, drug use, HIV infection, and spread of sexually transmitted infections in Moscow, Russian Federation. Lancet. 2005;366:57–60. doi: 10.1016/S0140-6736(05)66828-6. [DOI] [PubMed] [Google Scholar]
- 22.Wallace R, Wallace D. Socioeconomic determinants of health: community marginalisation and the diffusion of disease and disorder in the United States. BMJ. 1997;314:1341–5. doi: 10.1136/bmj.314.7090.1341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rhodes T, Simic M. Transition and the HIV risk environment. BMJ. 2005;331:220–3. doi: 10.1136/bmj.331.7510.220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.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. [Google Scholar]
- 25.European Centre for the Epidemiological Monitoring of AIDS, WHO and UNAIDS Collaborating Centre. HIV/AIDS surveillance in Europe. End year report 2004. 71 [ www.eurohiv.org]
- 26.Thorstensson R, Andersson S, Lindback S, et al. Evaluation of 14 commercial HIV-1/HIV-2 antibody assays using serum panels of different geographical origin and clinical stage including a unique seroconversion panel. J Virol Methods. 1998;70:139–51. doi: 10.1016/s0166-0934(97)00176-6. [DOI] [PubMed] [Google Scholar]