Married women don´t use condoms: an analysis of population-based data to understand PMTCT knowledge and practice
R. Viswanathan1, R.E. Ekpini1, A. Chatterjee1, D. Burke1, K. Johnson2, C. Luo1
1UNICEF Headquarters, HIV and AIDS, New York, United States, 2ICF Macro, Calverton, United States
Background: ´Knowing your epidemic´ and response will shape future investments in HIV, especially the prevention of mother-to-child transmission (PMTCT). Since 2005, this knowledge is derived primarily from health facility-based data gathered by UNICEF, WHO and UNAIDS, bypassing rich population-based data sources. We used population-based data to measure knowledge, access and utilization of PMTCT services.
Methods: We analysed Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) collected between 2002 and 2007 in 60 low and high-prevalence countries to generate evidence on PMTCT knowledge and practice according to epidemic typology. Bivariate and logistic regression methods were used for the analysis.
Results: In high-prevalence countries in Africa, condom-use among women ranges between 3-24% in cohabiting/married women and is highest among unmarried/single women, younger women, HIV+ women and women with >2 sexual partners. Women´s knowledge of MTCT is associated with wealth and urban residence, and low-prevalence countries demonstrate a range (19% in Iraq to >73% in Uzbekistan) that is comparable to high-prevalence countries (25% in Ethiopia to >73% in Haiti); levels among women and men are similar. Women who know about ARVs for PMTCT are between 80% (Haiti) and 270% (Benin) more likely to have been offered an HIV test. Of countries with sufficient data, when testing is offered, acceptance rates are >60%, with Zimbabwe and Cambodia having the highest levels, and Niger, the lowest.
Conclusions: The association of low condom-use with being married poses a challenge, particularly in high-prevalence countries. PMTCT and MTCT knowledge is not linearly associated with epidemic type or sex (associated with exposure to antenatal care), suggesting that various factors mediate knowledge. The skew of knowledge/ being offered an HIV test toward urban residence suggests that a decentralized approach to PMTCT is needed. If made more accessible, evidence indicates that the majority of women will accept HIV tests.
Back to the Programme-at-a-Glance