Reducing Alcohol-Related HIV Risk in Katutura, Namibia: Results from a Multi-Level Intervention

This 141-page paper shares information about a 3-year AIDSTAR-One demonstration project designed to reduce heavy drinking and risky sexual behaviour among bar patrons in a low-income neighbourhood on the outskirts of Namibia's capital, Windhoek. According to the report, a strong body of evidence shows that alcohol consumption is associated with the sexual behaviors that put people at risk for HIV and other sexually transmitted infections. This report describes how the intervention, funded by United States President's Emergency Plan for AIDS Relief, was implemented, monitored, and evaluated, and reports the final assessment results and key recommendations.

The project included two phases: Phase 1 focused on gathering and analysing formative research data about the prevalence and socioeconomic context of alcohol consumption and HIV risk. Phase 2 included intervention design, implementation, monitoring, and evaluation. The intervention activities were designed by AIDSTAR-One and Society for Family Health (SFH), based on the formative research, and in consultation with community leaders, community members, and bar owners. The intervention programme activities were coordinated by SFH. The final design of the pilot intervention included two main interlocking components: 1) mobilising the community to address hazardous alcohol consumption and 2) creating risk-averse bar environments. Methods for monitoring and evaluating the programme included: event-level monitoring and attendance data for the community mobilisation activities; 22 in-depth interviews and three focus group discussions with bar patrons, owners, staff, and other community members and leaders who had participated in the programme; and two rounds of quantitative survey data—prior to the intervention 500 bar patrons were recruited (from a sample of bars representative of the study area) and post-intervention 507 bar patrons were recruited from the 24 bars that had participated in the programme.
Key findings from the evaluation include:

• Binge drinking decreased from 54% at baseline to 25% at endline. The decrease was similar among women and men.
• Patrons of bars with highest exposure to programme activities consumed less alcohol per sitting when compared to patrons of less-exposed bars: 4.1 standard drinks among the unexposed group, and 3.3 among the exposed.
• Women with highest exposure to programme activities had lower rates of regular binge drinking (2% compared to 18%); among men, however, there was no statistically significant difference between the exposed and unexposed groups (36% and 34%, respectively).
• Heavy-drinking bar patrons were more likely to be exposed to intervention activities and showed significantly more favourable outcomes with respect to sexual risk. They were significantly more likely to have discussed condoms with a partner (87% compared to 72%); have obtained condoms (93% compared to 77%); and refused to have sex without a condom (62% compared to 47%).
• Results from partner-by-partner sexual behaviour data are mixed and suggest that positive intentions toward safer sex may not yet have translated into safer sexual behaviours, particularly with regard to having multiple sex partners and low rates of condom usage with regular partners. Reported condom use with casual sex partners is high across all subsamples.
• Bar owners and staff found it feasible to implement changes to their bar environments and these shifts were noted by their customers. Sixty-four percent of patrons report noticing educational materials about alcohol and HIV on display and 33% of the patrons noted shorter bar hours. Moreover, patrons from highly adherent bars were more likely to report favourable perceptions of safety and violence at the bar, suggesting that the programme may have created more risk-averse environments.
• Community mobilisation activities were feasible and popular. Community mobilisers conducted 77 events over the course of the project period, reaching over 750 community members, split almost equally between women and men.
• Penetration and reach of the community mobilisation events was effective. Thirty-two percent of bar patrons indicated they had participated in one of the two most frequent community mobilisation events (home visits and public meetings) and 56% reported they had heard of others participating in the mobilisation activities.

The report concludes that results from this small-scale and short-term programme are promising, suggesting that community mobilisation and delivery of prevention advice within the bar setting may motivate patrons to alter their attitudes and behaviours —both around the quantity of alcohol they consume as well as their intentions toward using and obtaining condoms. Although there is clearly much more work to be done and the evaluation notes that researchers did not observe consistent improvements in some key measures or average number of sex partners — such as condom use with regular sex partners — overall the project achieved several initial steps toward reducing alcohol-related HIV risk, with the full support of community members and bar owners. The evaluation makes the following recommendations for future programming:

• Include formative research. Both to ensure programme materials are contextually relevant as well as to begin encouraging community involvement from the initial stages of the project.
• Create partnerships with unlicensed, informal bars and their owners. Given the ubiquitous sale of home-brewed alcohol, and how important it is for impoverished communities as a cornerstone of the economy, campaigns to simply shut down illegal drinking outlets will be unsustainable until other employment and small business opportunities become available. In the meantime, this programme's results suggest that unlicensed bar owners can be willing and productive partners in the community-based response to harmful alcohol use. Given that unlicensed bars are both high-risk environments and willing partners, targeting these venues appears to be an expedient approach to addressing alcohol related HIV-risk.
• Encourage community involvement, especially during the initial project stages. Nest discussions of HIV within other alcohol-related issues that community members are most concerned about (e.g., crime prevention, economic development, child safety, etc.). Ensure, however, that the HIV thread is not lost in these discussions, despite community members' discomfort in discussing sexual behaviour.
• Target binge drinking. Heavy episodic drinking, more popularly known as binge drinking, has been shown to uniquely contribute to alcohol-related harm. Because community members and bar owners alike have observed the social dysfunction that can erupt when individuals drink very heavily, binge drinking is a discrete behaviour that can be isolated and intervened upon.
• Encourage bar owners and/or staff to engage in interpersonal communication with patrons around HIV and alcohol risk reduction. Results from this programme show significant associations between positive sexual risk behaviours (e.g., discussing, obtaining, and refusing to have sex without condoms) and exposure to the interpersonal interventions by bar staff. The potential power of this approach is that bar staff are uniquely situated to provide individuals with risk-reduction information and advice at the very times and places where opportunity for risk is highest. Results from this project also show that bar owners/staff were very capable of targeting most-at-risk individuals for intervention.