Background Female sex workers (FSWs) are considered a high-risk group for HIV and other STIs. Sex work practices of FSWs in Maharashtra are diverse and vary according to the sex work milieu (eg, street-, brothel- and bar-based). We examined changes in HIV and STI prevalence, associated risk behaviours and socio-economic variables among street-based (SB) and brothel-based (BB) FSWs in Maharashtra.

Methods Two rounds of cross-sectional surveys were conducted in 2005 (R1) and 2009 (R2) in selected districts of Maharashtra. Probability sampling methods, such as conventional and time-location cluster sampling, were used to sample 2257 (1212 BB and 1045 SB) FSWs in R1 and 2228 (1182 BB and 1046 SB) FSWs in R2. Sexual behaviours were assessed, blood and urine samples were collected to measure the prevalence of HIV, Neisseria gonorrhoeae (NG), Chlamydia trachomatis (CT) and syphilis. Changes were analysed using bivariate and multivariate methods.

Results HIV prevalence increased significantly among SB FSWs from 17.2% in R1 to 24.3% in R2 (OR 1.54, p=0.04), whereas among BB FSWs, there was no significant change (28.4 in R1 to 28.9 in R2, OR 1.03, p=0.88). Prevalence of active syphilis (TPHA and RPR positive at any titre) declined among BB FSWs from 15.4% (R1) to 11.0% (R2) (OR 0.67, p=0.05), while the decline among SB FSW was not significant (R1—13.6% to R2—10.6%; OR 0.75, p=0.22). NG prevalence fell from 7.7% (R1) to 3.9% (R2) (OR 0.48, p=0.006) for BB FSWs, whereas for SB FSWs, it declined from 7.2% in R1 to 4.2% in R2 (OR 0.56, p=0.07). No significant change was observed in the prevalence of CT in either group. Self-perceived risk of HIV among BB FSWs increased from 52.5% (R1) to 62.5% (R2), (OR 1.5, p=0.02); whereas among SB FSWs, it was unchanged (49.8% in R1 to 50.3% in R2; OR 1.02, p=0.23). The proportion of illiterate BB FSWs decreased from 80.1% (R1) to 74.3% (R2), (OR 0.72, p=0.02) while for SB FSWs, illiteracy increased significantly from 64.4% (R1) to 77.1% (R2), (OR 1.86, p<0.001). The proportion of FSWs having more than 10 clients per week was same over both rounds for BB FSWs but for SB FSWs, it increased significantly to 49.9% (R2) from 32.8% (R1) (OR 2.04, p<0.001).

Conclusion SB FSWs are at higher risk of contracting HIV and other STIs and are more vulnerable compared to BB FSWs. Prevention interventions must account for differences in profiles, risk behaviours and sex work typology to customise strategies for these groups for maximum impact.