A multilevel intervention to reduce stigma among alcohol consuming men living with HIV receiving antiretroviral therapy: findings from a randomized control trial in India.

OBJECTIVE: To examine the effectiveness of a multilevel intervention to reduce HIV stigma among alcohol consuming men living with HIV in India. DESIGN: A crossover randomized controlled trial in four sites. SETTING: Government ART centres (ARTCs) offering core services in the greater Mumbai area. PARTICIPANTS: Seven hundred and fifty two (188 per site) alcohol-consuming male PLHIV on ART were recruited. INTERVENTION: Multilevel intervention to reduce alcohol consumption and promote adherence by addressing stigma, implemented at the individual (individual counselling, IC), group (group intervention, GI) and community levels (collective advocacy, CA) in three distinct sequences over three cycles of 9 months each. MAIN OUTCOME MEASURE: HIV stigma, measured using the 16-item Berger Stigma scale. METHODS: The article examines the effectiveness of the interventions to reduce stigma using Linear Mixed Model regression. RESULTS: At baseline, 57% of participants had moderate-high levels of stigma (scores >40). All three counseling interventions were effective in reducing stigma when delivered individually, in the first cycle (collective advocacy: ßcoeff = -9.71; p < 0.001; group intervention: ßcoeff = -5.22; p < 0.001; individual counselling: ßcoeff = -4.43; p < 0.001). At then end of the second cycle, effects from the first cycle were sustained with no significant change in stigma scores. At the end of the third cycle, the site, which received CA+IC+GI sequence had maximum reduction in stigma scores (ßcoeff = -10.29; p < 0.001), followed by GI+CA+IC (ßcoeff = -8.23, p < 0.001). CONCLUSION: Baseline findings suggest that stigma remains a problem even with experienced patients, despite advances in treatment and adherence. Results of multilevel stigma reduction interventions argue for inclusion in HIV prevention and treatment program.

Viral load outcomes in a cohort of alcohol-consuming people living with HIV receiving antiretroviral therapy in Mumbai, India.

We examined the association between alcohol consumption, adherence and viral load (VL) in a cohort of 940 alcohol-consuming, human immunodeficiency virus (HIV)-positive men receiving antiretroviral therapy in Mumbai. Some of the participants (16.7%) had missed >1 doses in the last four days and 17.6% reported >1 treatment interruptions in the last 12 months; 60.8% of the participants consumed alcohol >2 times/month, 62.8% consumed >3 drinks/typical day and 11.5% reported binge drinking >1 times/month; 76.4% of the participants had VL<200 copies/mL. Higher alcohol consumption was associated with nonadherence (odds ratio [OR]: 1.21; 95% confidence interval [CI]: 1.11-1.31) and treatment interruptions (OR: 1.20; 95% CI: 1.11-1.31). We found no association between alcohol use and VL. There was, however, a significant interaction effect of alcohol use and nonadherence on virological failure (Adjusted Odds Ratio [AOR]: 1.23; 95% CI: 1.03-1.47) implying that alcohol negatively impacts VL outcomes but mediated through adherence. Alcohol use was associated with poor adherence, which in turn was a strong predictor of virological failure.

Circular labor migration and HIV in India: exploring heterogeneity in bridge populations connecting areas of high and low HIV infection prevalence.

BACKGROUND: The emerging human immunodeficiency virus (HIV) epidemics in rural areas of India are hypothesized to be linked to circular migrants who are introducing HIV from destination areas were the prevalence of HIV infection is higher. We explore the heterogeneity in potential roles of circular migrants in driving an HIV epidemic in a rural area in north India and examine the characteristics of the "sustaining bridge population", which comprises individuals at risk of HIV acquisition at destination and of HIV transmission into networks at origin capable of sustaining an epidemic. METHODS: Results of a behavioral survey of 639 male migrants from Azamgarh district, India, were analyzed using χ(2) tests and logistic regression. RESULTS: We estimated the size of various subgroups defined by specific sexual behaviors across different locations and over time. Only 20% fit our definition of a sustaining bridge population, with the majority making no apparent contribution to geographical connectedness between high- and low-prevalence areas. However, we found evidence of sexual contacts at origin that could potentially sustain an epidemic once HIV is introduced. Variables associated with sustaining bridge population membership were self-perceived HIV risk, current migrant status, and age. CONCLUSIONS: Circular migrants represent a heterogeneous population in terms of their role as a bridge group. Self-perception of heightened risk could be exploited in designing prevention programs.

Understanding out-migration among female sex workers in South India.

BACKGROUND: Migrant sex workers are known to be vulnerable to HIV. There is substantial female sex worker (FSW) mobility between the borders of Maharashtra and Karnataka, but little programming emphasis on migrant FSWs in India. We sought to understand the individual/cultural, structural, and contextual determinants of migration among FSWs from Karnataka. METHODS: A cross-sectional face-to-face interview of 1567 FSWs from 142 villages in 3 districts of northern Karnataka, India was conducted from January to June 2008. Villages having 10+ FSWs, a large number of whom were migrant, were selected following mapping of FSWs. Multinomial logistic regression was conducted to identify characteristics associated with migrant (travelled for ≥ 2 weeks outside the district past year) and mobile (travelled for <2 weeks outside the district past year) FSWs; adjusting for age and district. RESULTS: Compared with nonmigrants, migrant FSWs were more likely to be brothel than street based (Adjusted Odds Ratio (AOR): 5.7; 95% confidence interval: 1.6-20.0), have higher income from sex work (Adjusted Odds Ratio (AOR): 42.2; 12.6-142.1), speak >2 languages (AOR: 5.6; 2.6-12.0), have more clients (AOR per client: 2.9; 1.2-7.2), and have more sex acts per day (AOR per sex act: 3.5; 1.3-9.3). Mobile FSWs had higher income from sex work (AOR: 13.2; 3.9-44.6) relative to nonmigrants, but not as strongly as for migrant FSWs. CONCLUSION: Out-migration of FSWs in Karnataka was strongly tied to sex work characteristics; thus, the structure inherent in sex work should be capitalized on when developing HIV preventive interventions. The important role of FSWs in HIV epidemics, coupled with the potential for rapid spread of HIV with migration, requires the most effective interventions possible for mobile and migrant FSWs.