Understanding the Multidimensional Factors That Influence HIV Testing Among Women in the Democratic Republic of Congo

Approximately 37 million people worldwide are infected with the Human Immunodeficiency Virus (HIV), with the majority located in sub-Saharan Africa where women are more likely than men to become infected. Even though universal models such as the ‘HIV Continuum of Care’ and the ‘Bar Before the Bars’ exist that provide a context to identify barriers and tools to build the capacity for HIV testing and linkage to care, sub-Saharan Africa remains an HIV epicenter accounting for 72% of all worldwide HIV-related deaths in 2016. HIV testing rates vary significantly among many Central African countries; the DRC in particular has one of the lowest HIV testing rates in sub-Saharan Africa (21%) compared to neighboring countries Tanzania (67%) and Zambia (82%). However, why these testing discrepancies exist among countries that share tangible geographic borders is not well understood. The focus of this project was to identify which of the factors including HIV knowledge, cultural beliefs, stigma, and access to testing may directly contribute to the significantly lower HIV testing rates observed in women from the DRC. To evaluate this, a two-pronged approach was used. First, a retrospective cross-sectional explanatory study was designed using a dataset generated by the USAID Demographic and Health Survey Program containing data for 14,871 women between the ages of 18–49 who had been interviewed from 2013–2014. A series of variables addressing one of four primary areas including HIV knowledge, cultural beliefs, stigma, and access to testing were evaluated for their influence on the rate of HIV testing. Second, subjects were classified based on whether or not they had been HIV tested, then logistic regression was used to determine the capacity of the chosen variables to correctly predict a subject’s testing status. The prediction model generated using regression analysis based on the chosen variables was capable of identifying HIV-tested subjects with an accuracy of 81.8%. Women who were younger, of the poorest wealth index, those who lacked HIV knowledge, and those expressing behaviors associated with HIV-related stigma were substantially less likely to be HIV tested (p\u3c0.001). Location of residence, urban compared to rural, was the only statistically significant variable pertaining to access to care that influenced HIV testing (p\u3c0.001). Cultural beliefs did not significantly impact HIV testing (p=0.751-0.965). These gaps produced numerous limitations that prevented us from extrapolating the true impact of these factors on HIV testing in Congolese women. Future mixed method studies addressing considerations such as stigma and gender roles and conflict will be necessary to improve access to HIV testing and linkage to care for at-risk women in DRC