Lack of knowledge of HIV status a major barrier to HIV prevention, care and treatment efforts in Kenya: Results from a nationally representative study

Abstract

Background: We analyzed HIV testing rates, prevalence of undiagnosed HIV, and predictors of testing in the Kenya AIDS Indicator Survey (KAIS) 2007. Methods: KAIS was a nationally representative sero-survey that included demographic and behavioral indicators and testing for HIV, HSV-2, syphilis, and CD4 cell counts in the population aged 15-64 years. We used gender-specific multivariable regression models to identify factors independently associated with HIV testing in sexually active persons. Results: Of 19,840 eligible persons, 80% consented to interviews and blood specimen collection. National HIV prevalence was 7.1% (95% CI 6.5-7.7). Among ever sexually active persons, 27.4% (95% CI 25.6-29.2) of men and 44.2% (95% CI 42.5-46.0) of women reported previous HIV testing. Among HIV-infected persons, 83.6% (95% CI 76.2-91.0) were unaware of their HIV infection. Among sexually active women aged 15-49 years, 48.7% (95% CI 46.8-50.6) had their last HIV test during antenatal care (ANC). In multivariable analyses, the adjusted odds ratio (AOR) for ever HIV testing in women ≥35 versus 15-19 years was 0.2 (95% CI: 0.1-0.3; p<0.0001). Other independent associations with ever HIV testing included urban residence (AOR 1.6, 95% CI: 1.2-2.0; p = 0.0005, women only), highest wealth index versus the four lower quintiles combined (AOR 1.8, 95% CI: 1.3-2.5; p = 0.0006, men only), and an increasing testing trend with higher levels of education. Missed opportunities for testing were identified during general or pregnancy-specific contacts with health facilities; 89% of adults said they would participate in home-based HIV testing. Conclusions: The vast majority of HIV-infected persons in Kenya are unaware of their HIV status, posing a major barrier to HIV prevention, care and treatment efforts. New approaches to HIV testing provision and education, including home-based testing, may increase coverage. Targeted interventions should involve sexually active men, sexually active women without access to ANC, and rural and disadvantaged populations.