Heterosexual transmission has become the leading route of transmission of new HIV cases in China, surpassing intravenous drug use (IDU) for the first time since China's HIV epidemic began in the late 1980s.1 Due to high rates of unprotected sex and drug use behaviors, China's “floating population” of migrant workers “liudong renkou” is at risk for HIV/sexually transmitted infections (STIs) and may play a critical role in the spread of HIV in China2-5 as has been observed in other contexts.6 Largely because of China's gender imbalance and economic reforms over the last 20 years, an estimated 100-150 million migrants from the countryside who are mostly male, young, poor, and uneducated have migrated to urban centers in search of work.7 As mobility has increased and the economy has grown, the commercial sex industry has boomed and STI rates have risen.8,9 Separated from home and working long hours in difficult conditions, male migrants may patronize female sex workers (FSWs), increasing their risk of HIV and helping to fuel epidemics of HIV and STIs.4 Though China's HIV epidemic is still concentrated in high-risk groups, some regions of Yunnan province already have generalized epidemics with more than 1% of pregnant women in antenatal clinics testing HIV positive.10,11 Clients, especially in provinces such as Yunnan, may subsequently “bridge” these infections through unprotected intercourse with both sex workers and regular partners.
To our knowledge, this is the first study to examine HIV/STI prevalence and the associated risk behaviors among a large sample of miners in China. Only one other study has examined HIV risk among male migrant clients in China.12 Attention, however, has been given to HIV/STIs in migrant workers in China, with variable results. For instance, STIs were detected in 18% of market vendors of Eastern China13 and 20% of a community sample of migrant workers in Beijing, Shanghai, and Nanjing reported a history of STIs.12 Studies in Zhejiang14 and Shanghai,15 however, found minimal cases of STIs and no HIV. Still, behavioral studies in China have found that migrant populations have high-risk sexual behaviors, including multiple sex partners, engagement in commercial sex, and scarce condom use.3,16
Mining townships around Gejiu City in Yunnan, like other mining communities studied in Africa,17-19 have great potential for HIV/STI spread. Gejiu City, known as the “tin capital” of China, has a population of 453,300, not including an estimated 67,900 migrant workers from rural areas, most of whom are miners. Development of the local mining industry and the influx of single migrant miners have contributed to the expansion of the local commercial sex industry. The first HIV infection in Gejiu was reported in an injecting drug user in 1996, but by December 2004, a total of 1774 HIV cases were confirmed in Gejiu.20 Though IDU still accounts for the majority of HIV infections in Yunnan,20 the percentage of cases associated with sexual transmission has risen steadily from 5.3% in 1996 to 11.8% in 2004.11,21 There are indications that Gejiu City is moving toward a generalized HIV epidemic; in 2003, HIV prevalence among women receiving antenatal care was 0.8%.10 A pilot study conducted in 2005 among 232 miners in Laochang and Kafeng mining townships outside of Gejiu indicated that almost all (94.8%) were not local residents; 9.4% reported sex with FSWs; and the prevalence of HIV, gonorrhea, and chlamydia was 0.5%, 0.5%, and 9.3%, respectively.22 Our study assessed HIV and STI prevalence and risk factors among a large sample of miners in Gejiu City.
Laochang and Kafang are 2 mining townships located approximately 20 and 25 kilometers from Gejiu, respectively. They have a combined population of 13,000. Five mines were selected as study sites based on the following criteria: (1) short distance from the entertainment facilities in town with transportation readily available; (2) private ownership; and (3) employment of ethnic minority miners. There were approximately 2000 miners working in the 5 selected mines.
All miners enrolled in the study were: (1) employed in 1 of the 5 selected mines; (2) aged 16 or above; and (3) willing and able to provide informed consent.
A cross-sectional study was conducted in the 5 selected mines of Laochang and Kafang townships from March to May 2006. Every effort was made to include all potential subjects. Subjects who met the above selection criteria and signed the informed consent forms were assigned a personal identification number. Study staff explained to every eligible participant the purpose of the study, the procedures, and the risks and benefits of study participation. The informed consent process, interviews, and specimen collection were all conducted in private rooms at the hospital or clinic in each of the 5 mining districts. Each participant was interviewed face-to-face by an interviewer of the same gender to collect information on sociodemographics, sexual behaviors, drug use, HIV/AIDS-related knowledge, and STI symptoms.
Individual pretest counseling sessions were provided to each participant before specimen collection. A 5-ml void urine sample and 7-ml venous blood specimen was collected from each participant and labeled only with a personal identification number. All participants were requested to return in 4 weeks to receive testing results and posttest counseling by trained study staff.
All specimens were collected and processed according to the manual of procedures approved by the National Institutes of Allergy and Infections Diseases and the Chinese Center for Disease Control and Prevention (CDC). Blood specimens were sent to the Gejiu CDC laboratory, where the plasma was separated for HIV testing by enzyme-linked immunosorbent assay (ELISA). Specimens found positive for HIV antibody by ELISA were confirmed by 2 additional ELISA tests and 1 Western blot test at the Yunnan Provincial CDC reference laboratory.
The Gejiu CDC laboratory performed herpes simplex virus (HSV-2) antibody testing and Treponema pallidum antibody testing by rapid plasma reagin (RPR). Positive RPR specimens were confirmed by T. pallidum particle assay at the Yunnan Provincial CDC laboratory. Syphilis cases were identified by positive T. pallidum particle assay and RPR. Urine specimens were tested with morphine gold conjugate test strip (ACON MOP) for morphine in the Gejiu CDC laboratory.
Data Management and Statistical Analysis
All data collected were transmitted via the DataFax system to the data management team in Beijing. The final data received via DataFax were manually entered into the SAS database for analysis. Sociodemographics, HIV/STI prevalence, knowledge, and attitudes were analyzed for all miners and compared between miner clients and miner nonclients.
Statistical analysis of data was performed with SAS version 9.1 (SAS Institute Inc, Cary, NC). The proportions between groups were compared by using χ2 test or Fisher exact probability. The means between groups were compared by using Student t test. Univariate and multivariate logistic regression were used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for associations between HIV and STIs and variables of interest. Risk factors associated with HIV infection were adjusted for confounding variables and were fitted to a logistic regression model using a forward stepwise regression method. For multivariate analysis, variables were only retained if they were statistically significant at the level of P < 0.05 or if the coefficients of regression of the other variables in the same computation were substantially changed by their inclusion in the model.
The study protocol was approved by the institutional review boards of the National Center for AIDS/STD Control and Prevention, Yunnan Provincial CDC, and the Division of AIDS Prevention Science Review Committee at the US National Institutes of Health.
A total of 1804 miners were identified and screened and 1798 enrolled. Reasons for not participating included age less than 16 (n = 1); refused to participate (n = 3); and withdrawal from the study after giving informed consent (n = 2). Of all the enrolled miners, 1796 completed the questionnaire, 1760 provided a blood specimen, and 1773 provided a urine specimen.
Miners ranged in age from 16 to 58, with an average age of 28.5 (±7.4) years. There was no significant difference in the mean ages of the miners enrolled from the 2 townships (P = 0.17). Among the 1796 miners who completed questionnaires, 81.6% were <35 years; 65.1% attended school for ≤6 years; 14.3% never attended school; 60.9% were married; and of those married, 36.4% lived apart from their spouses. Forty-two percent (42.6%) were Han (the major ethnic group in China accounting for 90% of the total Chinese population), 44.8% were Hani, and 12.6% were from other ethnic groups. Most miners (92.8%) had a “hukou” (residence permit) from outside Gejiu, only 7.2% had a Gejiu hukou. Most (84.7%) with a hukou from outside Gejiu were from elsewhere in Yunnan, with only 8.1% from other provinces. A significant proportion (41.3%) had stayed at the study site for <1 year. Before mining in Gejiu, many miners were farmers (64.8%), and others were miners (11.0%), students (8.4%), factory workers (2.1%), construction workers (7.1%), or other (6.6%).
Three hundred thirty-nine miners (339 of 1796; 18.9%) reported a history of patronizing FSWs. Comparison of sociodemographic characteristics between client and nonclient miners are presented in Table 1.
HIV and STI Prevalence
Thirteen miners (0.7%) (95% CI: 0.3% to 1.1%) were HIV positive. The relative prevalence of syphilis, HSV-2, Neisseria gonorrhea, and Chlamydia trachomatis were 1.8% (95% CI: 1.2% to 2.4%), 9.6% (95% CI: 8.2% to 10.9%), 0.8% (95% CI: 0.4% to 1.2%), and 4.8% (95% CI: 3.8% to 5.8%). Prevalence of any STI was 14.9% (95% CI: 13.2% to 16.5%). There was a statistically significant difference in prevalence of syphilis (P = 0.02), C. trachomatis (P = 0.01), and HSV-2 (P = 0.01) between Kafang and Laochang townships.
Miner clients had a significantly higher prevalence of HIV (1.8% vs. 0.5%, P < 0.05), HSV-2 (14.9% vs. 8.4%, P < 0.001), N. gonorrhea (2.1% vs. 0.5%, P < 0.01), C. trachomatis (6.9% vs. 4.3%, P < 0.05), compared with nonclients (Table 1). There was no difference in syphilis (P = 0.05) based on client status. Prevalence of any STI was also significantly higher in clients (23.2% vs. 4.3%, P < 0.0001).
Awareness of HIV and Other STIs Knowledge
Misconceptions about HIV transmission were high. Only 66.9% had heard of HIV/AIDS; 84.5% believed that mosquitoes can transmit HIV; and 73.0% believed that sharing bathroom facilities with others could transmit HIV. Only 4.1% knew the correct routes of HIV transmission. In addition, only 43.9% of miners knew that people living with HIV/AIDS may seem healthy.
Miner clients in general had higher knowledge and awareness of HIV/STIs compared with nonclients (Table 2). Significantly, more clients had heard of HIV/AIDS (82.3% vs. 63.3%, P < 0.0001), had heard of condoms (75.8% vs. 54.6%, P < 0.0001), and knew HIV can be sexually transmitted (76.4% vs. 59.4%, P < 0.0001). In general, more clients than nonclients knew the correct routes of HIV transmission; more clients believed that there is a risk of HIV transmission from sexual intercourse (67.6% vs. 48.5%, P < 0.0001), a risk from sharing injection needles (65.5% vs. 47.4%, P < 0.0001), and a risk from blood transfusion (68.7% vs. 49.1%, P < 0.0001). However, only 9.1% of clients and 3.8% of nonclients believed that they were at risk for HIV.
Sexual Behavior and Drug Use
High-risk sexual behaviors were frequent among miners. Eighty-seven percent (87.3%) of all miners reported having sexual intercourse, and 34.0% reported having 2 or more sex partners in the past 12 months. One fifth of miners (18.9%) reported having purchased sex, and of these, 45.1% had visited FSWs twice or more in the past 12 months (range: X-70 visits). Reported condom use with FSWs was very low, 72.0% reported never using condoms with FSWs. Levels of drug use among miners were low, with 19 miners (1.1%) reporting previous illegal drug use, and 4 of these 19 (21.1%) reporting injecting drug use. Sixteen miners tested positive for morphine in their urine (0.09%).
In general, miner clients had more risk behaviors but higher knowledge and perceived risk of HIV/STIs (Table 2). Significantly more miner clients than nonclients reported illegal drug use (3.5% vs. 1.3%, P < 0.01). They also reported more STI symptoms in the last 12 months (1.2% vs. 0.3%, P < 0.001), multiple sexual partners (>3 partners in the last 12 months, 65.8% vs. 14.5%, P < 0.0001), early age of sexual debut (<16 years old, 21.8% vs. 15.9%, P < 0.05), and frequenting karaoke halls in leisure time (36.9% vs. 11.1%), a common work venue for FSWs. There was no significant difference in condom use with regular partners or FSWs and tattoo or surgical operation history.
HIV/STI Risk Factors
Risk factors for HIV and STIs by client status significant in univariate and multivariate analyses are presented in Table 3. HIV infection in miner clients was independently associated with illegal drug use (adjusted OR: 266.9, 95% CI: 13.3 to 999.9, P < 0.01), tattoo history (adjusted OR: 42.3, CI: 3.6 to 502.1, P < 0.01), and 3 or more FSW partners in the past 12 months (adjusted OR: 24.0, CI: 2.1 to 276.9, P < 0.01). HIV infection in nonclients was independently associated with illegal drug use (adjusted OR: 11.5, CI: 1.3 to 105.3, P < 0.05) and 3 or more sexual partners in the last 12 months (adjusted OR: 5.9, CI: 1.2 to 31.7, P < 0.05).
STI infection in miner clients was independently associated with illegal drug use (adjusted OR: 3.5, CI: 1.1 to 11.3, P < 0.05) and reported STI symptoms in the last 12 months (adjusted OR: 2.5, CI: 1.4 to 4.5, P < 0.01). STI infection in nonclients was independently associated with being an ethnic minority (adjusted OR: 2.6, CI: 1.8 to 3.6, P < 0.01) and living with a spouse (adjusted OR: 1.6, CI: 1.2 to 2.2, P < 0.01). Going to a karaoke hall frequently in leisure time, however, was independently protective of STIs in miner nonclients (adjusted OR: 0.5, CI: 0.3 to 0.9, P < 0.05).
Mining communities are high-risk areas for HIV and STI transmission, largely due to the temporary nature of the work and the associated disruption of social ties.17-19 Mining areas often do not provide accommodation for spouses or families, and most miners migrate alone (only 36% of married miners lived with spouses). Mining districts, perceived to be thriving economically, also attract female migrants who engage in sex work.18
The miners in our study were primarily young ethnic minority migrants with low educational attainment and except for their ethnic minority status match the demographic characteristics of China's so-called “surplus males.”7 These migrant surplus men may be helping to fuel the HIV epidemic in China through unprotected sex with multiple partners or through IDU behaviors.5 HIV-related risks among FSWs (often also migrants) have been more widely studied in China,23-30 than those of migrant surplus men.
HIV among the participating miners (0.7%) was 10 times higher than China's last national HIV estimate among the general population (0.05%, end of 2007).1 Comparisons with other migrant population studies should be made with caution given their diversity across China; however, HIV prevalence among miners in this study was much higher than recent studies of migrants in eastern China13-15 and slightly higher than a smaller convenience sample of miners in the same area (0.5%).22 N. gonorrhea and C. trachomatis rates in our miners were also greater than in a population-based study of 20-year-old to 64-year-old Chinese male adults (0.02% and 2.1%)31 and among migrant workers in Shanghai (0.5% and 3.5%).15
HIV/STI rates may be higher in our study because of increased heterosexual risks. More than 2 times more males (18.9%) reported ever patronizing FSWs compared with a study of male adults aged 20-64 years old in China (9.4%)32 and migrants in Shanghai (3.2%).15 Other sexual risks (multiple sexual partners, young age at sexual debut, STI symptoms) were higher among clients compared with nonclients. This is consistent with a recent study of male migrant clients in China, in which multiple sex partners and history of STIs were associated with client status.12
Clients also had higher HIV rates (1.7% vs. 0.5%) and overall STI rates (23.3% vs. 4.3%). Yunnan CDC data show a relatively high HIV prevalence among FSW clients (0.3%-1.8%, 2003),33 and a study in southwest China of FSW male clients reported an STI prevalence of 37.8%.34 Multiple encounters with FSWs (≥3) conferred an independent HIV risk for clients in our study; and even for nonclients, multiple sex partners was associated with HIV. Though never using a condom with FSWs was not associated with HIV/STIs, condom use among clients was very low (only 13.3% reported always using condoms with FSWs) placing them at high risk of acquiring HIV/STIs. One interesting finding was that many clients were of an ethnic minority. It is possible that certain ethnic minority cultures facilitate liberal sexual norms, such as 2 Hani holidays when it is customary for men to have extramarital sexual contacts. Ethnic minorities in our study were twice more likely to be STI infected (OR: 2.1).
Drug use was an important factor in HIV/STI risk among miners and was strongly associated with HIV and STIs among clients (OR: 266.9 and 3.5, respectively) with half (3 of 6) of the HIV-positive clients being drug users (2 of 3). Clients were also more likely to be drug users than nonclients (3.5% vs. 1.3%, P < 0.05), a finding consistent with those of another study of male migrant clients in China.12 Although multiple sex contacts with FSWs may increase HIV/STI risk among miner clients, those clients infected with HIV by IDU and with inconsistent condom use may simultaneously put FSWs at risk. Some international studies have proposed that HIV/STI transmission occurs from the “core” group of sex workers to “bridging” groups of clients who may transmit to the general population.35,36 Similar to Africa,17,37 our findings suggest in mining communities both FSWs and miner clients may be both “core” and “bridging” groups for HIV/STI transmission to the general population. Mining districts as a whole may be considered high-risk places, and interventions targeting the entire mining community should be considered, such as those proposed in a study of a Tanzanian mining community.17
Several limitations to this study should be noted. First, as this was a cross-sectional study, causality cannot be confirmed. For instance, as both drug use and multiple encounters with FSWs were independent risk factors for HIV in miner clients, it is not known how infection occurred. It is also unclear whether risky behaviors preceded migration or were encouraged by it. Second, as sexual behavior is a sensitive topic and data were collected by a face-to-face interview, participants likely under-reported commercial sex and other risk behaviors. Even with under-reporting, however, rates of patronizing FSWs, HIV, and STIs were higher in this population compared with the previously mentioned studies. Third, although this was a large sample of miners, results should not be extrapolated to all miners in China due to important regional differences in terms of demographics, economic conditions, and illicit drug availability and use.
Despite these limitations, this study demonstrates that miners who patronize FSWs are at substantial risk of contracting HIV/STIs and potentially escalating these epidemics in China. Although targeting mobile populations can be a major challenge,38 some mining communities have reported declining STIs after prevention, education, and systematic treatment of miners and FSWs.39 Workplace-based peer education, improvement of STI services, condom promotion, and community-based voluntary testing and counseling has also proven effective in other migrant communities. HIV/STI treatment options should not be reliant on a local residence permit, therefore, other methods of health care delivery should be made available.
In conclusion, miners in Gejiu constitute a high-risk population that may bridge HIV/STIs to both FSWs and regular partners and subsequently to the general population. Future studies should explore the availability and effectiveness of HIV/STI prevention and treatment programs in this high-risk community.
Authors thank Gejiu CDC, Yunnan Provincial CDC, and the National Center for STD Control in Nanjing for their aid in laboratory testing and their enthusiastic help and support throughout the project. Authors also thank all the study participants and Naomi Juniper for editing drafts of this article.
1. State Council AIDS Working Committee Office, UN Theme Group on AIDS in China. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China (2007)
. State Council AIDS Working Committee Office, Beijing, China; December, 2007.
2. Lu F, Wang N, Wu Z, et al. Estimating the number of people at risk for and living with HIV in China in 2005: methods and results. Sex Transm Infect
. 2006;82(Suppl 3):87-91.
3. Li X, Fang X, Lin D, et al. HIV/STD risk behaviors and perceptions among rural-to-urban migrants in China. AIDS Educ Prev
4. Anderson AF, Qingsi Z, Hua X, et al. China's floating population and the potential for HIV transmission: a social-behavioural perspective. AIDS Care
5. Hertog S, Merli GM. Demographic Shifts and the Spread of HIV in China. Paper presented at: Population Association of America annual meetings; March, 2007; New York City, NY.
6. Quinn TC. Global burden of the HIV pandemic. Lancet
7. Tucker JD, Henderson GE, Wang TF, et al. Surplus men, sex work, and the spread of HIV in China. AIDS
8. Gil VE, Wang MS, Anderson AF, et al. Prostitutes, prostitution and STD/HIV transmission in mainland China. Soc Sci Med
9. Cohen MS, Henderson GE, Aiello P, et al. Successful eradication of sexually transmitted diseases in the People's Republic of China: implications for the 21st century. J Infect Dis
. 1996;174(Suppl 2):S223-S229.
10. China Ministry of Health, UN Theme Group on HIV/AIDS in China. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China
. China Ministry of Health, China CDC National Center for AIDS/STD Control and Prevention, UNAIDS China Office. Beijing, China; 2003.
11. Xiao Y, Kristensen S, Sun J, et al. Expansion of HIV/AIDS in China: lessons from Yunnan Province. Soc Sci Med
12. Wang B, Li X, Stanton B, et al. HIV-related risk behaviors and history of sexually transmitted diseases among male migrants who patronize commercial sex in China. Sex Transm Dis
13. Detels R, Wu Z, Rotheram MJ, et al. Sexually transmitted disease prevalence and characteristics of market vendors in eastern China. Sex Transm Dis
14. Hesketh T, Li L, Ye X, et al. HIV and syphilis in migrant workers in eastern China. Sex Transm Infect
15. He N, Detels R, Zhu J, et al. Characteristics and sexually transmitted diseases of male rural migrants in a metropolitan area of Eastern China. Sex Transm Dis
16. He N, Detels R, Chen Z, et al. Sexual behavior among employed male rural migrants in Shanghai, China. AIDS Educ Prev
17. Desmond N, Allen CF, Clift S, et al. A typology of groups at risk of HIV/STI in a gold mining town in north-western Tanzania. Soc Sci Med
18. Campbell C. Migrancy, masculine identities and AIDS: the psychosocial context of HIV transmission on the South African gold mines. Soc Sci Med
19. Clift S, Anemona A, Watson-Jones D, et al. Variations of HIV and STI prevalences within communities neighbouring new goldmines in Tanzania: importance for intervention design. Sex Transm Infect
20. Zhang CA, Duan Y, Tong JY. Breakthroughs and Discoveries of AIDS: Best Practice of AIDS Integrated Control and Prevention of China-UK Programme in Yunnan Province
. Kunming, China: Yunnan Provincial Bureau of Health; 2006.
21. Cheng H, Zhang J, Jia M, et al. The study on HIV epidemic and trend in Yunnan Province. J Chin AIDS/STD Prev Control
22. Zhao R, Gao H, Shi X, et al. Sexually transmitted disease/HIV and heterosexual risk among miners in townships of Yunnan Province, China. AIDS Patient Care STDS
23. Chen XS, Yin YP, Liang GJ, et al. Sexually transmitted infections among female sex workers in Yunnan, China. AIDS Patient Care STDS
24. Liao SS, Schensul J, Wolffers I. Sex-related health risks and implications for interventions with hospitality women in Hainan, China. AIDS Educ Prev
25. Qian HZ, Vermund SH, Wang N. Risk of HIV/AIDS in China: subpopulations of special importance. Sex Transm Infect
26. Lau JT, Tsui HY, Siah PC, et al. A study on female sex workers in southern China (Shenzhen): HIV-related knowledge, condom use and STD history. AIDS Care
27. Ding Y, Detels R, Zhao Z, et al. HIV infection and sexually transmitted diseases in female commercial sex workers in China. J Acquir Immune Defic Syndr
28. Rogers SJ, Ying L, Xin YT, et al. Reaching and identifying the STD/HIV risk of sex workers in Beijing. AIDS Educ Prev
29. Wei SB, Chen ZD, Zhou W, et al. A study of commercial sex and HIV/STI-related risk factors among hospitality girls in entertainment establishments in Wuhan, China. Sex Health
30. Hesketh T, Zhang J, Qiang DJ. HIV knowledge and risk behaviour of female sex workers in Yunnan Province, China: potential as bridging groups to the general population. AIDS Care
31. Parish WL, Laumann EO, Cohen MS, et al. Population-based study of chlamydia infection in China: a hidden epidemic. JAMA
32. Lu L, Jia MH, Zhang XB. Analysis for epidemic trend of acquired immunodeficiency syndrome in Yunnan province of China. Chin J Prev Med
33. Zeng YL, Luan RS, Huang Y. Qualitative study of behavior associated HIV in male clients. Modern Prev Med
34. Cowan FM, Langhaug LF, Hargrove JW, et al. Is sexual contact with sex workers important in driving the HIV epidemic among men in rural Zimbabwe? J Acquir Immune Defic Syndr
35. Gomes do Espirito Santo ME, Etheredge GD. Male clients of brothel prostitutes as a bridge for HIV infection between high risk and low risk groups of women in Senegal. Sex Transm Infect
36. Lowndes CM, Alary M, Gnintoungbe CA, et al. Management of sexually transmitted diseases and HIV prevention in men at high risk: targeting clients and non-paying sexual partners of female sex workers in Benin. AIDS
37. Williams BG, Taljaard D, Campbell CM, et al. Changing patterns of knowledge, reported behaviour and sexually transmitted infections in a South African gold mining community. AIDS
38. Steen R, Vuylsteke B, DeCoito T, et al. Evidence of declining STD prevalence in a South African mining community following a core-group intervention. Sex Transm Dis
39. Bao V. Expanding Workplace HIV Prevention Programs for a Highly Mobile Population in Ho Chi Minh City
. Washington, DC: Population Council; 2003.