Secondary Logo

Journal Logo


Risk Factors and Prevalence of HIV and Sexually Transmitted Infections Among Low-Income Female Commercial Sex Workers in Mongolia

Hagan, José E. MS*; Dulmaa, Nyamkhuu MD, PhD

Author Information
Sexually Transmitted Diseases: February 2007 - Volume 34 - Issue 2 - p 83-87
doi: 10.1097/01.olq.0000237670.59140.0f
  • Free

MONGOLIA IS AN INDEPENDENT LANDLOCKED central Asian country bordered by Russia to the north and China to the south with a population of predominantly ethnic Mongol descent. Mongolia has had few reported cases of HIV despite high prevalence of other sexually transmitted infections (STIs). Mongolia's ties to the former Soviet Union ensured strong support for governmental public health initiatives in the past. Since 1990, with the dissolution of the Soviet Union and loss of this support, there has been a steady increase in the prevalence of STIs and other communicable diseases1–11 (Mongolian National Center for Communicable Diseases [NCCD], unpublished data). Two recent studies have documented the occurrence of drug-resistant strains of gonorrhea in Mongolia.12,13 Current surveillance efforts focus on antenatal and sexually transmitted clinic attendees, and the most recent reports have documented syphilis prevalence rates among female commercial sex workers (FSWs) to range from 4.7% to 18.8%11 (NCCD, unpublished data).

Despite this increasing STI burden, only 19 HIV cases have been reported to date in Mongolia. At the time of this study, only 3 cases had been detected. All of the cases have been sexually transmitted. Ten cases were in men who have sex with men (MSM) and 5 were heterosexually transmitted in people who had lived abroad. The route of transmission for one patient is unknown. Only 3 of the detected cases were in commercial sex workers (Robert Hagan, World Health Organization representative for Mongolia, personal communication, April 2006).

Commercial sex is illegal in Mongolia, but it is a common industry nonetheless. Recently, the growth of international tourism has led to an expanding supply of sex workers in urban centers. This has resulted in lower monetary charges, forcing some FSWs to work for 1,000 tugrug or less per sexual encounter (1,000 tugrug was equivalent to $1 U.S. during the time of this study). In addition, homeless youth and impoverished single mothers in Mongolia are known to regularly engage in sex work for minimal money to meet basic needs, behavior which is often referred to as “survival” sex work. Such activity is associated with increased risk for STI acquisition. These women have little negotiating power with their clients and may be less able to insist on condom use.14,15 The unique needs and behaviors of the lowest-income FSWs may place them at high risk of contracting sexually transmitted infections. Additionally, their marginalized status in society may make them more likely to have unreported illness as a result of their limited access to the healthcare system.

It is known that both ulcerative and nonulcerative genital infections increase HIV transmission. Commercial sex workers and their sexual contacts, with high rates of STIs, are therefore important populations contributing to the transmission of HIV and other STIs in developing countries.16–18 Other authors have noted that FSWs of different income strata have different sexual risk behaviors.19,20 Our study was designed to assess the prevalence and risk factors for HIV, syphilis, gonorrhea, and trichomoniasis among low-income sex workers in Mongolia. We also evaluated low-income sex workers as a potential reservoir for undiagnosed HIV and other STIs in Mongolia.

Materials and Methods

Population and Recruitment

Study participants were recruited with the assistance of 2 former FSWs in June and July 2002. As part of public health outreach and case-finding activities, women self-identifying as FSWs were approached in known locations of low-income commercial sex work. Study participants were also identified through social networks by the snowball sampling technique. Inclusion criteria were females aged 16 or older who charged less than 25,000 tugrug per sexual contact (“low-income FSW”).

Low-income FSWs in the capital city Ulaanbaatar gather in front of movies, post offices, bars, rent-per-hour hotels, and at the train and bus stations. Locations in Ulaanbaatar frequented by Chinese or Russian laborers or traders were emphasized. In Darkhan and Selenge, 2 cities in northern Mongolia connected by the trans-Siberian railroad line, we targeted the train stations and housing projects because these areas were known centers of commercial sex work. Participants were given a brief description of the study in lay Mongolian language, including expected benefits and inconveniences. A free medical examination, including testing for syphilis, gonorrhea, trichomoniasis, and HIV infection, was offered to all prospective participants along with counseling, condoms, and educational materials. Free treatment was offered for syphilis, gonorrhea, and trichomoniasis.

This study received approval from the Mongolia NCCD. Exemption from review was granted by Vanderbilt University Medical Center Institutional Review Board. Informed consent was obtained in written Mongolian language. The consent form was read to participants if needed. Participants were free to selectively participate in either the physical examination or the questionnaire. All participants were recruited anonymously. A card with an identification number was provided to participants and used to anonymously link test results with the participants.

Data Collection

Participants underwent pretest counseling, and biologic specimens were obtained by a physician and nurse at the NCCD in Ulaanbaatar or in temporary facilities in Darkhan and Selenge. All laboratory procedures followed the standard testing protocols of the Central Reference Laboratory of the NCCD.

Venous blood was collected for HIV testing with HIV enzyme-linked immunosorbent assay (Vironostika HIV 1/2; Organon Technika, Boxtel, Holland). Qualitative rapid plasma reagin (RPR) screening test was used to detect Treponema pallidum serum markers. Confirmation by positive T. pallidum hemagglutination assay (TPHA) was considered evidence of active syphilis infection. Speculum examinations were performed and cervical swabs collected from each participant by a physician. Cervical swab samples not obtained at the NCCD were added to glass slides and culture media at the temporary testing site and transported to the district hospital for incubation and final laboratory diagnosis. To test for Trichomonas vaginalis, slides were prepared for wet mount and immediate microscopy. Neisseria gonorrhoeae was detected by Gram stain and culture with a positive result on either test considered diagnostic. Participants were also screened for pregnancy by urine hCG test.

Those who tested positive for syphilis, gonorrhea, and/or trichomoniasis were counseled and provided with free treatment according to the standard syndromic treatment protocols of the NCCD. Each participant was given a package containing condoms, STI/HIV educational materials specifically targeting FSWs in Mongolian language, as well as contact information to learn their test results and receive treatment.

Sociobehavioral data were collected using a questionnaire that contained 43 multiple-choice, graded-response and written-response questions in Mongolian language. In addition to socioeconomic and demographic characteristics, sexual behavior, drug and alcohol use, sexual and general health, and STI/HIV knowledge and beliefs were assessed. The questionnaire was first pilot-tested and modified before collecting final data. Study personnel were present to provide assistance with the questionnaires if necessary, including verbal administration of the questionnaire to illiterate participants.

Statistical Analysis

Data were analyzed using Stata 6.0 (Stata Corp.). Univariate logistic regression was used to determine crude odds ratios. Adjusted odds ratios for both STI infection and for condom use (comparing “never use condoms” against all other responses) were calculated from a logistic regression model. Variables for multivariate analysis were selected from those significantly associated with condom use using the results of backward elimination stepwise regression. Ranked categorical variables (e.g., education level) were analyzed using the first category as the reference group. Bivariate correlation of behavioral factors was performed using the χ2 test.


Baseline Characteristics

A total of 200 commercial sex workers were contacted, of whom 179 (90%) agreed to an HIV test and 147 (74%) agreed to complete the questionnaire. STI examination results are available for 132 women (66%). The median age of the participants was 26.0 years (Table 1). Almost all had at least 8 years of schooling, 64% had at least one child, and 74% were unemployed. Half (53%) reported having no long-term partner. The median age of first intercourse was 18 years, and the median age of initiating commercial sex work was 21 years. Seventy-three percent had greater than 2 years of commercial sex experience. The frequency of commercial sex encounters ranged from one to 5 per night with a median of 2. Fees ranged from food or household items to 25,000 tugrug per client, which was the upper limit for inclusion in the study. Most participants charged fees between 5,000 and 15,000 tugrug (approximately $5–15 U.S.) per client. Over three fourths of participants made less than 15,000 tugrug a night in total commercial sex earnings. None of the women reported working in association with a brothel, whereas 25% worked with a pimp, 19% worked as a group of women, and the remainder worked alone (50%) or provided no answer (6%).

Baseline Sociodemographic Characteristics

Social Needs, Drug Use, Sexual Behavior, and HIV/Sexually Transmitted Infection Awareness

Questionnaire responses indicated that the participants were most concerned with acquiring food or money to buy food (64%), their perceived risk of contracting diseases (36%), their physical safety (35%), and finding a warm and safe place to sleep (21%). Drug and alcohol use was rare. Only 14% reported being concerned about excessive alcohol use, whereas only 8% reported that they were concerned about acquiring money to purchase alcohol, drugs, or tobacco. No women reported intravenous drug use or having previous sexual contact with somebody who used intravenous drugs. More frequent alcohol use was associated with higher fee per sexual encounter (χ2, 27.6; P = 0.04).

Almost 60% of participants reported that they “always use condoms.” The most commonly reported barriers to consistent condom use were client refusal (27%) and the high cost of purchasing condoms (11%). Eleven participants (8%) reported that their clients refuse to pay full price if asked to use condoms, and each of these 11 participants was infected with one or more STI.

Disease Prevalence

No cases of HIV were diagnosed in the 179 FSWs screened. Of 132 women, 57 (43%) tested positive for T. pallidum through nontreponemal and treponemal testing (RPR and TPHA, respectively), 18 (14%) tested positive for N. gonorrhoeae, and 37 (28%) women tested positive for active T. vaginalis infection. A total of 88 (67%) were diagnosed with one or more STIs and 22 (17%) were infected with more than one pathogen. Sixty-seven percent of the women reported symptoms of STI or pelvic inflammatory disease, including genital ulcers, vaginal discharge, and dyspareunia. Two women were pregnant, and both tested positive for active syphilis infection.

Factors Associated With Regular Condom Use and Sexually Transmitted Infection

A logistic model was created to estimate the adjusted odds ratios for variables associated with condom use or with STI. Variables associated with condom use in unadjusted analysis included younger age, higher level of education, more alcohol use, fewer years of commercial sex experience, greater fee per client, and more clients per night. However, in multivariate analysis of condom use, there were no clear trends observed. Total nightly income of between 5,000 and 15,000 tugrug was an independent predictor of condom use (odds ratio [OR] = 21.0; 95% confidence interval [CI] = 1.4–314.0), whereas a total income below 5,000 or above 15,000 was not. Having greater than 2 years of commercial sex experience was associated with STI in both univariate and multivariate analysis (OR = 8.2; 95% CI = 1.4–49.2) as seen in Table 2. More years of commercial sex experience was correlated with reduced fees (χ2, 95.3; P = 0.007).

Factors Associated With Sexually Transmitted Infection


Only 19 HIV infections have been documented in Mongolia through surveillance efforts and routine screening. However, high-risk populations such as FSWs, truck drivers, traders, and drug users made up only 4.8% of the population tested for HIV in 2003.11 The rate of detection of HIV has increased in recent years, and studies such as this one are needed to more completely understand STI and HIV prevalence and possible routes of spread in Mongolia.

In this population of low-income female commercial sex workers in Mongolia, we diagnosed no new cases of HIV despite a high prevalence (67%) of STIs. Our study results indicate that syphilis is the most prevalent STI in this population. Seventeen percent of women had mixed infections with syphilis/trichomoniasis representing the most common combination. Having more than 2 years of commercial sex experience was independently predictive of diagnosis with a STI. The high prevalence of syphilis (43%) is more than double the recent government estimates in this population6–9,11 (National Center for communicable diseases, unpublished data). Low-income female commercial sex workers may be underrepresented in disease prevalence estimates in Mongolian FSWs. The data suggest that they may be at substantially higher risk for acquiring, harboring, and spreading sexually transmitted disease and HIV.

The results of this study suggest that sexual risk behaviors in this group were closely linked to socioeconomic characteristics. Although most respondents claimed to use condoms regularly, the high prevalence of STIs, especially syphilis, suggests underreporting of risk factors and inconsistent or improper use of condoms. Women who reported no condom use were more likely to report lower education, longer commercial sex experience, fewer clients per night, lower fee per client, and less frequent alcohol consumption. However, no variable showed a clear trend on multivariate analysis.

Drug and alcohol use are typically associated with increased risky sexual behavior.21,22 However, more frequent alcohol consumption was associated with more frequent condom use in this group of women. The unexpected correlation between more frequent alcohol use and condom use could be explained by the prominence of vodka consumption in Mongolian culture; less consumption of alcohol may simply reflect having less expendable income. Indeed, we found that alcohol use was significantly correlated to higher fee per sexual encounter.

The most common barrier to condom use according to survey responses was client refusal followed by the high cost of purchasing condoms (100 tugrug, or 10 U.S. cents each). This is consistent with studies that indicate low acceptance of condom use in the general Mongolian population.10 Study participants were aware that condoms can be used to prevent infection, but indicated that they lack the negotiating power to ensure consistent use with their clients.

In this study, we attempted to optimize our sample of low-income FSWs by recruiting them in their working environments. This may have reduced the selection bias introduced by only sampling FSWs who seek care from STI clinics; however, the study sample depended on the locations we chose, on peer referral, and on self-identification as a FSW, potentially introducing other sources of bias. We attempted to capture a broad cross-section of low-income commercial sex workers in 3 different communities. However, the small sample size of some of the subgroups captured by the survey resulted in low statistical power, and this study is therefore inadequate to provide a complete description of the low-income FSW population. Specific questions not addressed by this study include their healthcare access and disease prevention strategies as well as detailed characteristics of their clients. Other important limitations of this study include bias introduced by underreporting of risk behaviors and the limitations of our laboratory methods. The only laboratory tests that were serology-based were syphilis and HIV, whereas the other tests relied on less sensitive microscopic and culture techniques. Finally, the study was unable to test for other important sexually transmitted pathogens such as chlamydia, herpes simplex virus type 2, and human papillomavirus.

The specific challenges of reducing the risk behaviors and improving the health of marginalized populations such as low-income female commercial sex workers require studies designed to understand their health status, needs, and their barriers to access. This study's focus on disease prevalence, behaviors, and risk factors of the lowest-income FSW is a strength over other FSW studies that may miss this group. Our data suggest that previous sampling efforts may be significantly underestimating the rates of infection among FSWs in Mongolia. Low-income FSWs in Mongolia are clearly at significant risk for HIV acquisition and are potentially facing a crisis of STI prevalence, notably syphilis, that remains unaddressed. The knowledge gained from this and further studies might lend itself to the design of effective intervention strategies that may both reduce a previously underaddressed burden of disease and help prevent the introduction of HIV into the apparently naïve Mongolian population.


1. Amindavaa O, Kristensen S, Pak CY, et al. Sexually transmitted infections among pregnant women attending antenatal clinics in Mongolia: Potential impact on the Mongolian HIV epidemic. Int J STD AIDS 2005; 16:153–157.
2. Ebright JR, Altantsetseg T, Oyungerel R. Emerging infectious diseases in Mongolia. Emerg Infect Dis 2003; 9:1509–1515.
3. Schwebke JR, Aira T, Jordan N, et al. Sexually transmitted diseases in Ulaanbaatar, Mongolia. Int J STD AIDS 1998; 9:354–358.
4. Purevdawa E, Moon TD, Baigalmaa C, et al. Rise in sexually transmitted diseases during democratization and economic crisis in Mongolia. Int J STD AIDS 1997; 8:398–401.
5. Manaseki S. Mongolia. A health system in transition. BMJ 1993; 307:1609–1611.
6. Garland SM, Tabrizi SN, Chen S, et al. Prevalence of sexually transmitted infections (Neisseria gonorrhoeae, Chlamydia trachomatis, Trichomonas vaginalis and human papillomavirus) in female attendees of a sexually transmitted diseases clinic in Ulaanbaatar, Mongolia. Infect Dis Obstet Gynecol 2001; 9:143–146.
7. National Statistical Office of Mongolia. Mongolian Statistical Yearbook, 1999. Ulaanbaatar, Mongolia: National Statistical Office of Mongolia, 2000:187.
8. Tellez I, Altankhuu M, Vermund S, et al. Hepatitis, syphilis, and HIV sentinel surveillance in Mongolia 1999–2000. Sex Transm Infect 2002; 78:223–224.
9. National Center for Health Development. Health Indicators 2001. Ulaanbaatar, Mongolia: Ministry of Health, 2001.
10. Status and Trends of STI, HIV and AIDS at the End of the Millennium. Manila, Philippines: World Health Organization, WPRO, 1999.
11. UNAIDS/WHO. Epidemiological Fact Sheet on HIV/AIDS and Sexually Transmitted Infections [UNAIDS/WHO web site]. 2004. Available at: Accessed September 23, 2004.
12. Schwebke JR, Vermund SH. Antimicrobial resistance among Neisseria gonorrhoeae isolates from Ulaanbaatar, Mongolia. Sex Transm Infect 2001; 77:463.
13. Lkhamsuren E, Shultz TR, Limnios EA, et al. The antibiotic susceptibility of Neisseria gonorrhoeae isolated in Ulaanbaatar, Mongolia. Sex Transm Infect 2001; 77:218–219.
14. Haley N, Roy E, Leclerc P, et al. HIV risk profile of male street youth involved in survival sex. Sex Transm Infect 2004; 80:526–530.
15. Gysels M, Pool R, Nnalusiba B. Women who sell sex in a Ugandan trading town: Life histories, survival strategies and risk. Soc Sci Med 2002; 54:179–192.
16. Piot P, Laga M. Prostitutes: A high risk group for HIV infection? Soz Praventivmed 1988; 33:336–339.
17. McKeganey NP, Prostitution and HIV: What do we know and where might research be targeted in the future? AIDS 1994; 8:1215–1226.
18. Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: The contribution of other sexually transmitted diseases to sexual transmission of HIV infection and STDs among sex workers. Sex Transm Infect 1999; 75:3–17.
19. Ford K, Wirawan DN, Fajans P. Factors related to condom use among four groups of female sex workers in Bali, Indonesia. AIDS Educ Prev 1998; 10:34–45.
20. Miyazaki M, Takagi S, Kato M, et al. Prevalences of and risk factors for sexually transmitted diseases among Japanese female commercial sex workers in middle- and high-class soaplands in Japan. Int J STD AIDS 2002; 13:833–838.
21. Kim J, Celentano DD, Crum RM. Alcohol consumption and sexually transmitted disease risk behavior: Partner mix among male Korean university students. Alcohol Clin Exp Res 1998; 22:126–131.
22. Vanable PA, McKirnan DJ, Buchbinder SP, et al. Alcohol use and high-risk sexual behavior among men who have sex with men: The effects of consumption level and partner type. Health Psychol 2004; 23:525–532.
© Copyright 2007 American Sexually Transmitted Diseases Association