University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Declines in risk behaviour and sexually transmitted infection prevalence following a community-led HIV preventive intervention among female sex workers in Mysore, India
Global Health Sciences Literature Digest
Published June 30, 2009
Journal Article

Reza-Paul S, Beattie T, Syed HUR, et al. Declines in risk behaviour and sexually transmitted infection prevalence following a community-led HIV preventive intervention among female sex workers in Mysore, India. AIDS 2008;22 (suppl);S91-S100.


To measure the effect of a community-led intervention on sexual behavior and sexually transmitted infections (STIs) among female sex workers (FSWs) in Mysore, India


Mysore, Karnataka state, southern India

Study Design

Two unlinked cross-sectional studies conducted at 6 months following a community risk-reduction/STI control intervention and then again 30 months after the first survey


Female sex workers


Condom use with commercial and regular sex partners, number of unprotected sex acts in the past month, HIV, syphilis, herpes simplex virus-2 (HSV-2), Trichomonas, Chlamydia, and gonorrhea infections


Intervention: The intervention began with mapping and enumeration of sex workers followed by peer-mediated outreach, which addressed difficulties reported by the sex workers and promoted a community among the sex workers. Sexual health services were established to provide syndromic management of STIs, physical examinations, syphilis treatment, and presumptive treatment every 3-6 months for Chlamydia and gonorrhea. The health services included the promotion of routine STI check-ups in addition to helping people seek care for symptomatic disease. Condom promotion was done through social marketing and making condoms available in non-traditional outlets. A drop-in center that provided a safe place, social networking, community kitchen, and literacy classes was created for the sex workers.

Sampling: A sample size of 400 was estimated for each survey and recruited using time location cluster sampling.

Data collection: Behavioral data were collected anonymously using face-to-face interviews conducted by peers in Kannada (the local language). The measures included demographics, sex work, migration patterns, sexual history, sexual behaviors, condom use, drug injection practices, and use of the intervention drop-in center and sexual health clinic. Blood and urine were collected to test for HIV, HSV-2, syphilis, gonorrhea, Chlamydia, and Trichomonas infections. All participants were referred for free a health check-up and treatment following the interview and collection of biological specimens.

Statistical analysis: x-square tests were used to measure statistical associations, and odds ratios were used to measure the strength and direction of the association. Participants may have enrolled in more than one survey


There were 431 women invited to participate; of this number, 429 completed the baseline assessment. In the second round, all 425 invited women agree to participate. There were 175 who participated in both.

The median age was 30 years (range 18-45 years) and most women were illiterate. Median age at sexual debut was 16 (range 2-27 years) and most women were either currently or formerly married. The median age of initiating sex work was 25 years. The majority of women were street-based sex workers.

Of the 429 women sampled 6 months after the intervention, 64% had been visited by a peer educator, 56% visited the drop-in center, 52% visited the health center, 45% received treatment for STIs, 38% witnessed a condom demonstration, 13% were currently carrying a condom, and 33% reported a broken condom in the past month.

Among the 425 women surveyed 30 months after the first survey, there were increases in the proportion that had visited a peer educator (95%), visited the drop-in center (90%), visited the clinic (92%), received treatment for STIs (90%), witnessed a condom demonstration (85%), and were carrying a condom (30%). All of these increases were statistically significant (P=0.001). The proportion of women who reported condom breakage in the past month decreased to 28% (P=0.07).

Similarly, there were increases in the proportion of women reporting zero unprotected sex acts in the past month with customers and with their regular partner. There were also statistically significant decreases in the proportion of women diagnosed with syphilis, Trichomonas, Chlamydia, and gonorrhea. The proportion of women testing positive for HIV was stable and the proportion of women testing positive for HSV-2 increased from 64% to 79%.

Participation in components of the intervention appeared to be associated with decreases in risk. Among those who were visited by a peer educator, the adjusted odds ratio (OR) for condom use with the last occasional commercial partner was 3.87 (95% confidence interval [CI]: 2.43-6.15), and for the last repeat client was (OR 1.95, 95% CI: 1.23-3.09). Among those who visited the drop-in center, the adjusted OR for condom use with the last occasional commercial partner was 2.72 (95% CI: 1.79-4.15), and for the last repeat client was 1.94 (95% CI: 1.29-2.93).


Self-reported condom use increased, STIs decreased, and HIV prevalence remained stable in the 6-30 months following the community-level intervention, and this increase was associated with having participated in components of the intervention.

Quality Rating

This study uses repeat cross-sectional surveys to measure the effect of a community-level intervention. A randomized controlled trial would have been able to measure the efficacy of the intervention but would have presented ethical and logistical difficulties. The cross-sectional design offers some evidence of effect, particularly in self-reported condom use. Unfortunately, the authors did not collect data on the study measures prior to implementation of the intervention. The sample size was adequate for both surveys and the sampling methods are likely to have produced a relatively representative sample. Although behavioral data were collected through self-report, the use of biological specimens strengthened the study. The lack of independence of the samples is a serious weakness in the study. Overall this is a good study.

In Context

HIV in India occurs in a variety of groups, with FSWs accounting for a large proportion of heterosexually transmitted infections.(1) Interventions targeting this population have been limited.(2,3,4) An effective prevention intervention in FSWs could have a large effect on HIV incidence in India.

Programmatic Implications

The evidence to date suggests that combining sexual risk reduction, condom promotion, and improved access to STI treatment reduces STIs and HIV infections. This study demonstrates the effect of using a multipronged approach to reduction in sexual risk and STI and highlights the benefits of community-level interventions and the use of structural interventions (e.g., the use of the drop-in center and health clinic). Providing basic health services combined with condom promotion and use of peer-educators is a feasible and likely an effective method to address the HIV epidemic in resource-constrained areas.


  1. Halli SS, Blanchard J, Satihal DG, Moses S. Migration and HIV transmission in rural South India: an ethnographic study. Cult Health Sex 2007;9:85-94.
  2. Fung IC, Guinness L, Vickerman P, et al. Modelling the impact and cost-effectiveness of the HIV intervention programme amongst commercial sex workers in Ahmedabad, Gujarat, India. BMC Public Health 2007;7:195.
  3. Gangopadhyay DN, Chanda M, Sarkar K, et al. Evaluation of sexually transmitted diseases/human immunodeficiency virus intervention programs for sex workers in Calcutta, India. Sex Transm Dis 2005;32:680-4.
  4. Jana S, Basu I, Rotheram-Borus MJ, Newman PA. The Sonagachi Project: a sustainable community intervention program. AIDS Educ Prev 2004;16:405-14.