Feasibility and acceptability of home-based HIV testing among refugees: a pilot study in Nakivale refugee settlement in southwestern Uganda.

BACKGROUND: Refugees in sub-Saharan Africa face both the risk of HIV infection and barriers to HIV testing. We conducted a pilot study to determine the feasibility and acceptability of home-based HIV testing in Nakivale Refugee Settlement in Uganda and to compare home-based and clinic-based testing participants in Nakivale. METHODS: From February-March 2014, we visited homes in 3 villages in Nakivale up to 3 times and offered HIV testing. We enrolled adults who spoke English, Kiswahili, Kinyarwanda, or Runyankore; some were refugees and some Ugandan nationals. We surveyed them about their socio-demographic characteristics. We evaluated the proportion of individuals encountered (feasibility) and assessed participation in HIV testing among those encountered (acceptability). We compared characteristics of home-based and clinic-based testers (from a prior study in Nakivale) using Wilcoxon rank sum and Pearson's chi-square tests. We examined the relationship between a limited number of factors (time of visit, sex, and number of individuals at home) on willingness to test, using logistic regression models with the generalized estimating equations approach to account for clustering. RESULTS: Of 566 adults living in 319 homes, we encountered 507 (feasibility = 90%): 353 (62%) were present at visit one, 127 (22%) additional people at visit two, and 27 (5%) additional people at visit three. Home-based HIV testing participants totaled 378 (acceptability = 75%). Compared to clinic-based testers, home-based testers were older (median age 30 [IQR 24-40] vs 28 [IQR 22-37], p < 0.001), more likely refugee than Ugandan national (93% vs 79%, < 0.001), and more likely to live ≥1 h from clinic (74% vs 52%, < 0.001). The HIV prevalence was lower, but not significantly, in home-based compared to clinic-based testing participants (1.9 vs 3.4% respectively, p = 0.27). Testing was not associated with time of visit (p = 0.50) or sex (p = 0.66), but for each additional person at home, the odds of accepting HIV testing increased by over 50% (OR 1.52, 95%CI 1.12-2.06, p = 0.007). CONCLUSIONS: Home-based HIV testing in Nakivale Refugee Settlement was feasible, with 90% of eligible individuals encountered within 3 visits, and acceptable with 75% willing to test for HIV, with a yield of nearly 2% individuals tested identified as HIV-positive.

Changes in HIV-related behaviours, knowledge and testing among refugees and surrounding national populations: a multicountry study.

To our knowledge, there is currently no published data on the prevalence of risky sex over time as displaced populations settle into long-term post-emergency refugee camps. To measure trends in HIV-related behaviours, we conducted a series of cross-sectional HIV behavioural surveillance surveys among refugees and surrounding community residents living in Kenya, Tanzania and Uganda, at baseline in 2004/2005 and at follow-up in 2010/2011. We selected participants using two-stage cluster sampling, except in the Tanzanian refugee camp where systematic random sampling was employed. Participants had to reside in a selected household for more than weeks and aged between 15 and 49 years. We interviewed 11,582 participants (6448 at baseline and 5134 at follow-up) in three camps and their surrounding communities. The prevalence of multiple sexual partnerships ranged between 10.1 and 32.6% at baseline and 4.2 and 20.1% at follow-up, casual partnerships ranged between 8.0 and 33.2% at baseline and 3.5 and 17.4% at follow-up, and transactional partnerships between 1.1 and 14.0% at baseline and 0.8 and 12.0% at follow-up. The prevalence of multiple partnerships and casual sex in the Kenyan and Ugandan camps was not higher than among nationals. To our knowledge these data are the first to describe and compare trends in the prevalence of risky sex among conflict-affected populations and nationals living nearby. The large reductions in risky sexual partnerships are promising and possibly indicative of the success of HIV prevention programs. However, evaluation of specific prevention programmes remains necessary to assess which, and to what extent, specific activities contributed to behavioural change. Notably, refugees had lower levels of multiple and casual sexual partnerships than nationals in Kenya and Uganda and thus should not automatically be assumed to have higher levels of risky sexual behaviours than neighbouring nationals elsewhere.

Differences in HIV-related behaviors at Lugufu refugee camp and surrounding host villages, Tanzania.

BACKGROUND: An HIV behavioral surveillance survey was undertaken in November 2005 at Lugufu refugee camp and surrounding host villages, located near western Tanzania's border with the Democratic Republic of Congo (DRC). METHODS: The sample size was 1,743 persons based on cluster survey methodology. All members of selected households between 15-49 years old were eligible respondents. Questions included HIV-related behaviors, population displacement, mobility, networking and forced sex. Data was analyzed using Stata to measure differences in proportions (chi-square) and differences in means (t-test) between gender, age groups, and settlement location for variables of interest. RESULTS: Study results reflect the complexity of factors that may promote or inhibit HIV transmission in conflict-affected and displaced populations. Within this setting, factors that may increase the risk of HIV infections among refugees compared to the population in surrounding villages include young age of sexual initiation among males (15.9 years vs. 19.8 years, p = .000), high-risk sex partners in the 15-24 year age group (40% vs. 21%, chi2 33.83, p = .000), limited access to income (16% vs. 51% chi2 222.94, p = .000), and the vulnerability of refugee women, especially widowed, divorced and never-married women, to transactional sex (married vs. never married, divorced, widowed: for 15-24 age group, 4% and 18% respectively, chi2 8.07, p = .004; for 25-49 age group, 4% and 23% respectively, chi2 21.46, p = .000). A majority of both refugee and host village respondents who experienced forced sex in the past 12 months identified their partner as perpetrator (64% camp and 87% in villages). Although restrictions on movements in and out of the camp exist, there was regular interaction between communities. Condom use was found to be below 50%, and expanded population networks may also increase opportunities for HIV transmission. Availability of refugee health services may be a protective factor. Most respondents knew where to go for HIV testing (84% of refugee respondents and 78% of respondents in surrounding villages), while more refugees than respondents from villages had ever been tested (42% vs. 22%, chi2 63.69, p = .000). CONCLUSION: This research has important programmatic implications. Regardless of differences between camp and village populations, study results point to the need for targeted activities within each population. Services should include youth education and life skills programs emphasizing the benefits of delayed sexual initiation and the risks involved in transactional sex, especially in the camp where greater proportions of youth are affected by these issues relative to the surrounding host villages. As well, programs should stress the importance of correct and consistent condom use to increase usage in both populations. Further investigation into forced sex within regular partnerships, and programs that encourage male involvement in addressing this issue are needed. Program managers should verify that current commodity distribution systems ensure vulnerable women's access to resources, and consider additional program responses.