Objectives/Goal: To assess prevalence of and correlates to human immunodeficiency virus (HIV) infection among female sex workers in Tashkent, Uzbekistan.
Study Design: Women participating in this cross-sectional study completed a questionnaire and HIV testing between April 2003 and March 2004. Logistic regression analyses determined correlation of variables to HIV infection.
Results: Of 448 women, 10% (45) were HIV infected, which was associated with ever injecting drugs (AOR = 20.20; 95% confidence interval [CI], 7.69–53.07), street-based sex work (AOR = 4.52; 95% CI, 1.84–11.12), exchanging sex for drugs (AOR = 4.74; 95% CI, 1.84–12.18), and more sexually transmitted infection treatments in the preceding 3 months (AOR = 2.43; 95% CI, 1.14–5.17).
Conclusions: Although injection drug use is the strongest correlate to HIV infection, sexual risk behaviors are independently related and should receive focus in prevention efforts targeted to this population.
A study of female sex workers in Tashkent, Uzbekistan, found that 10% were human immunodeficiency virus infected, with greatest likelihood of infection among those having injected drugs or exchanging sex for drugs.
From the *University of California, San Diego, La Jolla, California; †Republic of Uzbekistan Center for HIV/AIDS Control and Prevention, Tashkent, Uzbekistan; ‡U.S. Naval Medical Research Unit No. 3 (NAMRU-3), Cairo, Egypt; and the §U.S. Military HIV Research Program at the Walter Reed Army Institute of Research, and the Henry M. Jackson Foundation for the Advancement Military Medicine, Inc., Rockville, Maryland
We thank Mr. Ilkhom Kasymov and Dr. Adkham Mamatkulov for their assistance with study conduction and Ms. Oksana Subotova for her assistance with translation and data entry. We also thank Steffanie Strathdee, PhD, and Richard Garfein, PhD, for their comments on manuscript preparation. Finally, we are grateful to the women who participated in this study for their time and trust.
This study was supported by the U.S. Military HIV Research Program (USMHRP) at the Walter Reed Army Institute of Research (WRAIR) and by the U.S. Naval Medical Research Center (NMRC), Silver Spring, MD, protocol number CPHS 130, DOD NAMRU-3.2002.013).
Correspondence: Catherine S. Todd, MD, MPH, Division of International Health and Cross-Cultural Medicine, Department of Family and Preventive Medicine, University of California, San Diego, 9500 Gilman Drive, Mailcode 0622, La Jolla, CA 92093-0622. E-mail: email@example.com.
Partial results of this study were previously presented at the 16th Biennial Meeting of the International Society for Sexually Transmitted Diseases Research (ISSTDR), Amsterdam, The Netherlands, July 11, 2005. Abstract No. MP-025.
Disclaimer: The opinions and assertions made by the authors do not reflect the official position or opinion of the U.S. Department of the Navy or Army, of the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., or of the respective in-country National HIV/AIDS Control Programs and other nongovernmental organizations (NGOs).
THE RUSSIAN FEDERATION AND many former Soviet republics are currently experiencing the most rapid increase in number of human immunodeficiency virus (HIV) cases globally.1 The number of adults with HIV increased 29% between 2002 and 2004, which has been largely attributed to injection drug use, the major transmission route in this region.1,2 The poor economic status of these countries has fostered an informal economy based on illicit activities, including drug trafficking and commercial sex work.3–6 Previously published studies have reported HIV risk behavior among female sex workers (FSW); however, there is sparse documentation of HIV prevalence among FSW in these countries.7–9
The FSW population is at increased risk of both becoming infected with and infecting others with HIV and other sexually transmitted infections (STIs). Infected clients may then pass infection to regular partners, resulting in the “bridging” effect.10–12 In Uzbekistan, HIV cases have increased substantially in last 6 years, with 25 cases reported in 1999 to 5612 reported by the end of 2004.2,13 In 2004, 2016 new cases were officially registered, representing a 56% increase over the previous year.13 Although injection drug use currently accounts for 84% of all HIV infections, the possibility of a second epidemic wave related to heterosexual transmission cannot be discounted because the number of cases due to heterosexual contact have occupied a larger percentage of cases in each successive year.2 Previous descriptive studies among FSW in the Russian Federation report a tremendous increase in numbers of women engaging in sex work, changing cultural perceptions about sexual behavior, and inconsistent condom use despite awareness that condoms reduce HIV and STI risk.8,14 Recent outbreaks of syphilis and other STIs among FSWs have been reported in Kyrgyzstan, Kazakhstan, and the Russian Federation, and an increase in HIV infection rates could quickly follow suit.5,15–17
Sex work is illegal in Uzbekistan. In Tashkent, the capital, sex work is principally advertised covertly through print, computers, and taxis, as well as in hotels and restaurants, though street-based and brothel or pimp-controlled work also exists.18,19 FSW are less likely to seek medical care because public clinics may report them to authorities and private clinics are costly. This covert nature and reluctance to seek medical care may result in hidden HIV outbreaks not detected through current screening measures.
Like other Central Asian countries, the first HIV epidemic in Uzbekistan occurred among injection drug users (IDU), with cases exponentially increasing between 2001 and 2002.20 A study conducted in Tashkent among IDU between April 2003 and March 2004 reported an HIV prevalence of 29.8%.21 However, there have been indications that the HIV epidemic has spread beyond IDU into the general population because more than 20 cases of HIV infection have been diagnosed among pregnant women in Tashkent in the last year, with sexual contact with a regular partner the sole risk factor (personal communication, Dr. Ulmaz Tashlanova).
The objectives of this study were to determine the prevalence and correlates of HIV infection in a population of FSW in Tashkent, Uzbekistan.
Materials and Methods
The study was conducted in Tashkent, the capital of Uzbekistan, whose metropolitan population is approximately 2.1 million people. Tashkent is a major trade center in Central Asia and is connected by both rail and air routes to large cities in the Russian Federation, Western Europe, and the Middle East.
Study Design and Enrollment
This cross-sectional study was conducted between April 2003, and March 2004. Potential participants were recruited both by outreach workers in areas of known FSW congregation and among FSW presenting to Istiqbolli Avlod. The mission of Istiqbolli Avlod is to offer counseling and assistance to FSW and victims of human trafficking and, through this work, has established rapport with the sex worker community. A modified approach to snowball sampling was used because participants were encouraged to refer other sex workers within their social network. No additional compensation was provided for such referrals, nor were social networks mapped to determine whether oversampling from a specific group had occurred. Because this population is concealed and difficult to access, no set sample size was designated.
Criteria for inclusion were age 18 years or older, history of sex work within the preceding month, and ability to provide written informed consent. Consented subjects completed a self-administered questionnaire with information regarding demographic and socioeconomic characteristics, sexual risk behaviors in both commercial and noncommercial settings, illegal drug use history, and knowledge of and prior testing history for HIV. A further interview eliciting general and reproductive health history was conducted by the study gynecologist.
Following the questionnaire and examination, participants provided a sample of venous blood for HIV antibody testing. All participants received free pre- and posttest counseling, STI and HIV testing, and STI treatment following local standards of care as compensation for participation. Those participants diagnosed with HIV were provided with referral information to the Republic of Uzbekistan Center for AIDS Prevention and Control (RUCAPC) for further counseling and treatment. Before enrollment, this project was approved by the investigational review boards of the U.S. Naval Medical Research Unit, No. 3, and the Ministry of Public Health of Uzbekistan.
HIV Serologic Testing
Venous blood specimens were tested for HIV antibody by enzyme-linked immunosorbent assay (ELISA; Genscreen HIV 1/2 v. 2, BioRad Diagnostics, Hercules, CA). Confirmation was performed by Western blot assay (WB; HIV BLOT 2.2, GeneLabs Diagnostics, Singapore). Only samples repeatedly reactive by ELISA and confirmed positive by WB testing were coded as HIV-positive. All testing was performed at the RUCAPC in Tashkent, Uzbekistan.
Chi-square and Fisher exact test were applied to evaluate differences in prevalence of HIV infection, and 95% confidence intervals (95% CIs) were based on exact binomial formulas. Logistic regression analysis was used to obtain odds ratios (ORs) and associated 95% CIs. Potential confounders were assessed before analysis; age-adjustment or stratification were used to better characterize some relationships. In order to determine independent risk factors associated with HIV infection, all variables with a P value <0.15 were entered into a multiple logistics regression model. All reported P values were 2-sided; P values <0.05 were considered statistically significant. All statistical analysis was performed using Stata version 8.0 (Stata, College Station, TX).
A total of 448 FSW were enrolled, of whom 45 (10.0%) tested HIV antibody positive. Generally, women entering the study were well educated (81.7% having at least completed secondary school), previously or currently married (73%), and had an identified profession (71.4%), even if not currently employed in that occupation. Women with HIV infection differed little from those without infection by age, marital status, or educational background. The only significant difference detected was that HIV-infected women were more likely to report having been previously diagnosed with hepatitis B or C.
Regarding risky behaviors, drug use was common within the study population, as 20.8% (N = 93) reported any drug use and 9.2% (N = 41) reported ever injecting drugs. Only one third of women reported always using condoms with clients, and just 7.3% reported consistent use with steady partners. No significant association was found between reported inconsistent (less than “always”) condom use with clients or regular partners and HIV infection.
Knowledge of HIV transmission routes and prevention was assessed, with 83% of women stating they knew how HIV was transmitted and 82.8% stating they knew how to prevent HIV infection. However, of those claiming HIV knowledge, only 42.9% correctly identified 3 transmission routes with no incorrect responses, and only 29.2% knew 2 or more ways to prevent transmission, with no incorrect responses. Women correctly identifying condoms as means of prevention (81.3%) were more likely to report condom use with clients (OR = 2.29; 95% CI, 1.26–4.17).
Univariate correlates of HIV infection are displayed in Table 1. Exchanging sex for drugs, any history of drug use, and, particularly, injection drug use were strongly related to HIV infection. Of those with HIV infection, 53.3% had a history of injection drug use. Due to the strong relationship with injection drug use, differences between IDU-FSW and the remaining population were explored. IDU-FSW were more likely to be older, have a longer duration of sex work, have a history of childhood abuse, and a greater number of monthly clients while significantly less likely to know that condoms prevent HIV transmission (Table 2). Of behaviors specific to IDU-FSW, only daily injection use was significantly related to HIV infection (Table 3).
Sex work–related factors significantly related to HIV infection were young age (≤18 years) at initiation of sex work, longer duration of sex work, street-based sex work, and having had more STI treatments in the preceding 3 months, while multiple sex work venues were marginally significant. Median duration of sex work was 5.25 (interquartile range [IQR], 4–10) years, being significantly longer for HIV-infected (8 years; IQR, 5–10 years) compared to uninfected subjects (5 years; IQR, 3–10 years). Only knowledge of condoms as a means of HIV prevention was protective for HIV infection (Table 1).
The number of STI treatments was further assessed for interaction with age, given the established association between young age and STI.22,23 When only those aged 25 years or less were considered (N = 183), number of STI in the previous 3 months was marginally associated with HIV (age-adjusted OR = 1.23; 95% CI, 0.965–1.57). However, for those aged 26 years or more (N = 265), those with HIV infection were more likely to have a greater number of STI treatments in the prior 3 (age-adjusted OR = 3.53; 95% CI, 1.46–8.55) and 12 months (age-adjusted OR = 3.86; 95% CI, 0.938–15.86).
In multiple logistic regression analysis, FSW were 20 times more likely to be HIV infected if they injected drugs (Table 4) Injection drug use, exchanging sex for drugs, increased number of STI treatments in the preceding 3 months, and street-based sex work were also independently related to HIV infection.
To our knowledge, this is the first report of HIV prevalence among FSW exclusively in Central Asia. In a study published in 2002 conducted among high-risk groups in Uzbekistan, including FSW, no HIV cases were detected within this or other risk groups.13,25 However, a 17-fold increase was reported between 2000 and 2003, and a 56% increase in incident cases was recorded for 2004.13,25 Our study detected a 10% HIV prevalence in a Tashkent FSW group, which is lower than that detected in settings with a mature epidemic but higher than the 2.8% prevalence measured in Russian FSW in 2001.26
HIV infection was most strongly associated with injection drug use in all analyses (Tables 1 and 2), similar to the high HIV rates reported among IDU in the Russian Federation and other former Soviet Union countries.27,28 Exchange of sex for drugs was also associated with HIV infection, similar to female IDU in the Russian Federation.7 The significant association between any drug use and HIV (Table 1) appears to be a reflection solely of the strong relationship with injection drug use, as this relationship did not persist in analyses controlled for injection drug use.
Injection drug use confounded relationships between HIV and sex work duration. Possible reasons may be the overlap of the injection drug epidemic with initiation of sex work for those who enter sex work solely to fund drug use or the simultaneous epidemics of injection drug use and sex work following the dissolution of the Soviet Union. Duration of sex work may represent a proxy for establishment of widespread injecting drug use in Tashkent, which, like the general Central Asian region, has emerged in the last decade.29,30 Further evidence to support this theory comes from a study conducted in 2000 among street-based sex workers in Ho Chi Minh City, Vietnam, where an injection drug use epidemic began in 1997, with resultant increases in HIV prevalence in both IDU and FSW.31 Similar to our study, a history of injection drug use (AOR = 101.30) was the strongest risk factor for HIV, though the significant association between young age diverged from our study.32 Nguyen et al.32 also indicated that FSW who injected drugs had a mean number of 3.7 years in sex work, which was significantly lower than for non-IDU FSW (mean 6.1 years; P < 0.001), reflecting the duration of the injection drug epidemic in this setting.
Women with a history of injection drug use also differed by reporting lower knowledge of condoms as protection against HIV, higher numbers of clients monthly, and higher likelihood of childhood abuse. Higher number of clients among IDU-FSW was noted in Viet Nam; however, condom knowledge was not different for IDU-FSW.31 The increased likelihood of women reporting a history of physical or sexual abuse to be IDU has been noted among FSW in other locations.32
Aspects of sex work associated with HIV in our study were number of STI treatments in the preceding 3 months and street-based sex work. The association of number of STI treatments in the preceding 3 months with HIV potentially represents risky behavior or number of exposures. Sex work–associated increases in syphilis and other STI in the Russian Federation have been documented in recent years.5,15 Young age has also been associated with higher rates of STI and risky sexual behavior in previously published studies22,23; this finding is reflected in our study by the significantly higher mean numbers of STI treatments in younger subjects (Table 1). Higher numbers of STI treatments at both 3 and 12 months were independently associated with HIV infection only in older subjects. One possible explanation is that HIV prevalence increases with duration of risky sexual behavior; the presence of higher numbers of STI in HIV-infected older subjects may indicate that risky sexual behaviors have persisted over a longer duration of time. Number of STI treatments in the preceding year was not consistently associated with HIV, in contrast to the relationship observed by others.33–35 Inability to accurately recall the exact number of treatments may account for this lack of association.
In the Russian Federation, FSWs working street and train/bus station venues were at greater risk for STIs, as these women had higher numbers of clients, less access to health care, and earned less for each sexual exchange.14 We found that street-based sex work was significantly associated with HIV infection, potentially due to increased vulnerability of women to clients/authorities and/or having clients (e.g., truck drivers) with multiple high-risk partners. However, client number did not differ significantly between street/truck stop and other sex work venues (2.24 versus 2.11 clients/day respectively; P = 0.357), contradicting the strong association demonstrated in Vietnam, Ghana, Cambodia, and Burkina Faso.36–39 Many women (42%) in Tashkent recruited clients through multiple venues; thus, there was less disparity in sex work environments. The similarity between venues in daily client number makes this theory plausible; future studies should attempt to determine an association.
Our study has limitations that should be taken into account. First, due to cross-sectional design, the associations reported are not causal but may provide a basis for a future cohort study. The sample size also diminishes the certainty with which some associations may be made, particularly for the IDU subgroup (N = 41). Second, our results may have been affected by sampling bias resulting from nonrandom selection, artificially strengthening the associations found in this study, particularly for women using only telephone referral or pimp/brothel venues as they would be theoretically less accessible to outreach workers. Also, limiting enrollment to those 18 years or greater may have created bias as 33.71% of subjects initiated sex work before age 18. Third, this study was conducted solely in urban Tashkent, which may not be representative of the sex work environment in other Uzbek settings. Fourth, condom use, prior HIV testing, and prior STI diagnoses are based on self-report and may not be accurate. Finally, FSW using the services provided by the study may have been at higher risk for HIV as they may have enrolled solely for free, anonymous testing. However, given the threat of prosecution for sex work and the likely universal appeal of free services in a trusted environment provided by this study, we believe that our study population constitutes a representative sample of FSW in Tashkent.
In summary, this study represents a first attempt to assess HIV prevalence and behavioral correlates in a central Asian FSW population. Further studies are needed, though the strong association of HIV infection with injection drug use and sex-work-related risks and the low proportion of consistent condom use among clients suggest directions for HIV prevention programming. Drug treatment and harm reduction programs aimed at decreasing risky IDU behaviors, interventions targeted to FSW and their clients on sexual risk reduction, and development of confidential, accessible reproductive health services for FSW are urgently needed to prevent further HIV transmission within this high-risk group, their clients, and, by extension of the clients, to the general population.
1. UNAIDS. Report on the Global AIDS Epidemic: Joint United National Program on HIV/AIDS. Geneva, Switzerland: UNAIDS, 2004.
2. United Nations Development Programme (UNDP). HIV/AIDS in Eastern Europe and the Commonwealth of Independent States: Reversing the Epidemic: Facts and Policy Options. Bratislava, Slovak Republic: United Nations Development Programme, 2004.
3. International Crisis Group (ICG). Central Asia: Drugs and Conflict: ICG Asia Report No. 25. Osh, Brussels: International Crisis Group, 2001:1–32.
4. Rhodes T, Ball A, Stimson GV, et al. HIV infection associated with drug injecting in the newly independent states, eastern Europe: the social and economic context of epidemics. Addiction 1999; 94:1323–1336.
5. Tichonova L, Borisenko K, Ward H, et al. Epidemics of syphilis in the Russian Federation: trends, origins, and priorities for control. Lancet 1997; 350:210–213.
6. Walberg P, McKee M, Shkolnikov V, et al. Economic change, crime, and mortality crisis in Russia: regional analysis. BMJ 1998; 317:312–318.
7. Benotsch EG, Somlai AM, Pinkerton SD, et al. Drug use and sexual risk behaviours among female Russian IDUs who exchange sex for money or drugs. Int J STD AIDS 2004; 15:343–347.
8. Aral SO, St Lawrence JS. The ecology of sex work and drug use in Saratov Oblast, Russia. Sex Transm Dis 2002; 29:798–805.
9. Rhodes T, Mikhailova L, Sarang A, et al. Situational factors influencing drug injecting, risk reduction and syringe exchange in Togliatti City, Russian Federation: a qualitative study of micro risk environment. Soc Sci Med 2003; 57:39–54.
10. Morris M, Podhisita C, Wawer MJ, et al. Bridge populations in the spread of HIV/AIDS in Thailand. AIDS 1996; 10:1265–1271.
11. Thuy NT, Lindan CP, Phong TH, et al. Predictors of visits to commercial sex workers by male attendees at sexually transmitted disease clinics in southern Vietnam. AIDS 1999; 13:719–725.
12. Ray K. Changing pattern of HIV infection in a tertiary care hospital in New Delhi, India. Int J STD AIDS 2004; 15:753–757.
14. Aral SO, St Lawrence JS, Tikhonova L, et al. The social organization of commercial sex work in Moscow, Russia. Sex Transm Dis 2003; 30:39–45.
15. Nurgazieva J. Kyrgyzstan: STD management during the syphilis outbreak. Entre Nous Cph Den 1999; 42:13.
16. WHO. Task Force for the Urgent Response to the Epidemics of Sexually-Transmitted Infections in Eastern Europe and Central Asia. Available at: http://www.who.dk/document/e69832.pdf
. Accessed February 9, 2005.
17. Borisenko KK, Tichonova LI, Renton AM. Syphilis and other sexually transmitted infections in the Russian Federation. Int J STD AIDS 1999; 10:665–668.
18. Thomas RM. Assessment of the commercial sex work scene in Tashkent, Uzbekistan. UNAIDS Uzbekistan Office Report; 1997:1–7.
19. Oostvogels R. Assessment of female sex workers and clients in Tashkent, Republic of Uzbekistan. UNAIDS Uzbekistan Office Report; 2001:1–10.
20. Kurbanov F, Kondo M, Tanaka Y, et al. Human immunodeficiency virus in Uzbekistan: epidemiological and genetic analyses. AIDS Res Hum Retroviruses 2003; 19:731–738.
21. Sanchez JL, Todd CS, Bautista CT, et al. High HIV prevalence and risk factors among injecting drug users in Tashkent, Uzbekistan, 2003–04. Drug Alcohol Depend. In press.
22. Weinstock H, Berman S, Cates W Jr. Sexually transmitted diseases among American youth: incidence and prevalence estimates, 2000. Perspect Sex Reprod Health 2004; 36:6–10.
23. Panchaud C, Singh S, Feivelson D, et al. Sexually transmitted diseases among adolescents in developed countries. Fam Plann Perspect 2000; 32:24–32, 45.
24. Ruzibakiev R, Kato H, Ueda R, et al. Risk factors and seroprevalence of hepatitis B virus, hepatitis C virus, and human immunodeficiency virus infection in Uzbekistan. Intervirology 2001; 44:327–332.
25. Godinho J, Renton A, Vinogradov V, et al. Reversing the Tide: Priorities for HIV/AIDS Prevention in Central Asia. Washington, DC: The World Bank, 2005. ECSHD/ECCU 8.
26. Steen R, Dallabetta G. Sexually transmitted infection control with sex workers: regular screening and presumptive treatment augment efforts to reduce risk and vulnerability. Reprod Health Matters 2003; 11:74–90.
27. Rhodes T, Lowndes C, Judd A, et al. Explosive spread and high prevalence of HIV infection among injecting drug users in Togliatti City, Russia. AIDS 2002; 16:F25–31.
28. Centers for Disease Control and Prevention. Rapid increase in HIV rates: Orel Oblast, Russian Federation, 1999–2001. MMWR Morb Mortal Wkly Rep 2003; 52:657–660.
29. UNODC/ODCCP. Conference report: regional conference on drug abuse in central Asia: situation assessment and responses. Tashkent, Uzbekistan, December, 2002.
30. Dehne KL, Pokrovskiy V, Kobyshcha Y, et al. Update on the 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.
31. Nguyen AT, Nguyen TH, Pham KC, et al. Intravenous drug use among street-based sex workers: a high-risk behavior for HIV transmission. Sex Transm Dis 2004; 31:15–19.
32. El-Bassel N, Witte SS, Wada T, et al. Correlates of partner violence among female street-based sex workers: substance abuse, history of childhood abuse, and HIV risks. AIDS Patient Care STDS 2001; 15:41–51.
33. Hayes RJ, Schulz KF, Plummer FA. The co-factor effect of genital ulcers on the per-exposure risk of HIV transmission in sub-Saharan Africa. J Trop Med Hyg 1995; 98:1–8.
34. Cameron DW, Simonsen JN, D'Costa LJ, et al. Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet 1989; 2:403–407.
35. O'Farrell N. Genital ulcers, stigma, HIV, and STI control in sub-Saharan Africa. Sex Transm Infect 2002; 78:143–146.
36. Minh TT, Nhan do T, West GR, et al. Sex workers in Vietnam: how many, how risky? AIDS Educ Prev 2004; 16:389–404.
37. Asamoah-Adu C, Khonde N, Avorkliah M, et al. HIV infection among sex workers in Accra: need to target new recruits entering the trade. J Acquir Immun Defic Syndr 2001; 28:358–366.
38. Ohshige K, Morio S, Mizushima S, et al. Cross-sectional study on risk factors of HIV among female commercial sex workers in Cambodia. Epidemiol Infect 2000;124:143–152.
39. Nagot N, Ouangre A, Ouedraogo A, et al. Spectrum of commercial sex activity in Burkina Faso: classification model and risk of exposure to HIV. J Acquir Immune Defic Syndr 2002; 29:517–521.