Factors associated with social group membership among HIV-positive and -negative women in Tabora, Tanzania

Background: The interaction of people in groups (social capital)?utilized as an indicator of empowerment?is associated with positive health outcomes and behaviors. There are few studies examining the factors associated with social capital among women attending prevention of mother-to-child HIV transmission (MTCT) and maternal and child health services in rural communities of developing countries.
Methods: The SAFI study in Tabora, Tanzania is a facility-based trial evaluating the effect of SMS appointment reminders and cash transport payments on attendance and facility delivery among pregnant women from 27 antenatal clinics. This 2015 baseline analysis of eligible HIV-positive and HIV-negative pregnant women (n=1,329; age≥18,) utilized a multivariable logistic regression model to describe factors associated with social capital, defined as membership of a community-based group/club who have met at least once in the past three months. Variables in the model included education (woman and partner), partner work, knowledge that ARVs reduce sex transmission of HIV, and MTCT, knowledge of partner status, joint decision-making on health care for self, being formerly employed in the past year and ever delivering in a facility.
Results: Overall, 12.6% (172/1,329) women were group members, with comparable HIV prevalence among group and non-group members (26% vs. 27%). Comparable proportions of group and non-group members reported joint decision-making on health care for self, partner work, were married, knew ARV decreases HIV sexual transmission, knew partner HIV status, ever delivered in a facility. However, greater proportions of group members (and their partners) completed secondary school (p=0.001, and p=0.003 respectively), were formerly employed in the past year, (p< 0.001), and knew antiretrovirals (ARVs) reduced MTCT (p=0.009). In the multivariate analysis being formally employed in the past year (adjusted odds ratio [AOR] 3.04, 95% confidence interval [CI]: 1.99, 4.65); knowing that ARVs reduced MTCT (AOR 2.65, 95% CI: 1.32, 5.33) and knowing partner HIV status (AOR 1.79; 95% CI: 1.18, 2.72) were significantly associated with social capital.
Conclusions: Irrespective of HIV status, women who earned an income, had enhanced MTCT knowledge, and knew their partners'' status were more likely to be in social groups. Group membership may enable gains in health knowledge and facilitate partner disclosure.

G. Woelk1, G. Antelman2, G. Mbita2, R. van de Ven2, P. Njao3, R. Machekano1
1Elizabeth Glaser Paediatric AIDS Foundation, Washington DC, United States, 2Elizabeth Glaser Paediatric AIDS Foundation, Dar es Salaam, Tanzania, United Republic of, 3Ministry of Health and Social Welfare, Dar es Salaam, Tanzania, United Republic of