Differing HIV vulnerability among female sex workers in a high HIV burden Indian state

INTRODUCTION The HIV sentinel surveillance [HSS] conducted in 2010-11 among female sex workers [FSW] in the state of Maharashtra, India provided an opportunity to assess characteristics of different types of FSWs and their HIV risk. It is important for India's National AIDS Control Program, to understand the differences in vulnerability among these FSW, in order to define more specific and effective risk reduction intervention strategies. Therefore, we analyzed data from HSS with the objective of understanding the HIV vulnerability among different types of FSW in Maharashtra. MATERIAL AND METHODS Cross sectional data collected as a part of HSS among FSWs in year 2010-11 from 21 sentinel sites in the state of Maharashtra were analyzed to understand the vulnerability and characteristics of different types of female sex workers based on their place of solicitation using multinomial logistic regression. RESULTS While the HIV prevalence was 6.6% among all FSWs, it was 9.9% among brothel based [BB], 9% among street based [SB] and 3.1% and 3.7% among home based [HB], and bar based [Bar-B] sex workers respectively. SB FSWs were least likely to be located in HIV low burden districts [ANC] [ARRR: 0.61[95% CI: 0.49, 0.77]], but were 6 times more likely to be recently [<1 year] involved in sex work [ARRR: 6.15 [95% CI: 3.15, 12.0]]. The number of clients of SB FSWs in the preceding week were lower than 11% [ARRR: 0.89 [95%CI: 0.87, 0.90]] as compared to the BB FSWs denoting lesser client load. The duration since last paid sex was shorter [ARRR: 0.94[95%CI: 0.91, 0.96]] as compared to the BB FSWs. CONCLUSION Street based FSWs have emerged as one of the most vulnerable types of FSW with a high HIV prevalence similar to BB FSWs. Our study reveals that they have more frequent sex acts despite lower client loads, and are more likely to be located in districts highly affected by HIV (ANC prevalence >1%). We identify them as a group to be focused on for prevention interventions and it is likely that they would be easily amenable to novel interventions due to their higher literacy rate as compared to other typologies.