Background
The HIV epidemic in India remains predominantly concentrated in groups where individuals display high
risk behaviours, including men who have sex with men (MSM). Widespread behavioural changes are
crucial to the control of HIV, but need to be informed by an understanding of the risk factors for
infection. However, reliability and validity of self-reported behaviour are difficult to determine. This
thesis aims to contribute to the literature comparing innovative data collection modes for self-reported
HIV risk behaviour in developing countries.
Methods
The Avahan programme is a large-scale HIV-prevention project that focuses on the six states in India
with the highest HIV prevalence. The programme focuses on core and bridging groups, including MSM.
This thesis presents the findings of one aspect of the monitoring and evaluation: behavioural data
collected using face-to-face interviews (FTFI) and informal confidential voting interviews (ICVI) among
MSM sampled in public place and Hammam cruising sites in Bangalore.
Results
A review of empirical data collected in developing countries comparing FTFI with new interviewing
tools, found private data collection methods to have mixed success in reducing underreporting of risky
behaviour. A comparison of ICVI and FTFI in India found that ICVI significantly increased reporting of
stigmatised behaviours, but results did not adhere consistently to expectation. A number of self-identified
categories of MSM are commonly applied in the intervention context in India, each of which was
generally associated with different HIV-risk behaviours. Although there was evidence of role segregation
and identity-specific behaviour, the categories were found to be more fluid than has previously been
documented. Bisexual behaviour was common, and condom use with female partners was low, which
suggests a potential bridge of HIV transmission into the general population.
Conclusions
The dataset provided a solid description of HIV risk behaviours among MSM cruising in public places in
Bangalore, which has immediate implications for designing appropriate targeted HIV prevention
programmes that address fluidity in risk behaviour between MSM identities and reach out to
behaviourally bisexual men, rather than treating MSM as a homogenous group. Both the systematic
review and the comparison of ICVI and FTFI highlighted difficulties in gathering ‘truthful’ self-reported
behaviour, as determining the precise reasoning where individual responses departed from the presumed
norm was impossible. Qualitative research might contribute to a better understanding of the motivations
behind reporting biases amongst MSM.