Background Tensions between religion/religious beliefs and personal behaviours may impact HIV-related health seeking for persons at risk for HIV. We evaluated the relationship between self-reported church attendance and sexual partner characteristics on CD4+ T lymphocyte count (CD4 count) and history of HIV screening among persons presenting for the first time to establish HIV care.

Methods This was a secondary analysis of prospectively collected data at a University HIV Clinic in the U.S between April 2008 and June 2012. Socio-demographic and behavioural information, including church attendance, is collected and linked to laboratory data. Using logistic regression, we assessed for a 2-way interaction between church attendance and sexual behaviour on CD4 count at entry into care and on history of prior HIV testing in univariate and multivariable models.

Results Of 508 participants, current church attendance was reported by 56%. There was a significant interaction between church attendance and sexual behaviour (p = 0.02) with CD4 count at the time of entry into care. MSM who reported church attendance were significantly more likely to present with a CD4 count < 200 cells/mm³ (p = 0.013) than MSM who did not. No difference in CD4 count was observed for MSW or WSM when evaluated by report of church attendance. There was a significant interaction between church attendance and sexual behaviour (p = 0.012) on history of previous HIV testing. WSM who attended church were more likely to report previous HIV testing (p = 0.01). MSM who attended church were less likely to report previous HIV testing (unadj p = 0.041) but this difference did not maintain significance in adjusted models.

Discussion These findings frame a potentially important interaction between church attendance and sexual behaviour on timing of HIV diagnosis and presentation into care with important implications for individual health outcomes and secondary HIV prevention.