Housing instability is associated with failure to achieve virologic control among pregnant individuals with HIV

Abstract

Homelessness and housing instability, which are on the rise nationally, are considered important social determinants of health. Among nonpregnant adults living with HIV, both have been associated with decreased linkage to medical care and virologic nonsuppression. This association may be particularly concerning in pregnancy, because virologic control is the primary determinant of HIV perinatal transmission. In addition, housing instability in pregnancy may be an independent risk factor for adverse perinatal outcomes, further amplifying perinatal risks in pregnant individuals living with HIV. However, the role of housing as a social determinant of health among such individuals is largely unstudied. The objective was to examine the association between housing instability and virologic control among pregnant individuals living with HIV. This was a retrospective cohort study of pregnant individuals seeking perinatal care in a specialty HIV clinic from 2007 to 2018. Markers of virologic control, including time from antiretroviral therapy initiation to virologic suppression, antiretroviral therapy adherence, and viral load at 36 weeks and at delivery, were assessed. All patients underwent assessment of housing status with a licensed clinical social worker and were classified as experiencing housing instability (ie, staying with family or friends, transitional housing, treatment program, shelter, outdoors or vehicle, hotel, or incarcerated) vs not experiencing instability (renting or owning). Multivariable regression models assessed the associations of housing instability with virologic control. Of 232 pregnant patients living with HIV with documented housing status, 41.4% (n=96) experienced housing instability. Patients with housing instability were younger and more likely to self-identify as non-Hispanic Black, have public or no insurance, and have a mental health or substance use disorder. They were less likely to be married, be employed, or have greater than a high school education. There were no differences in parity, number of prenatal visits, or timing of HIV diagnosis between groups. On adjusted analyses, patients with housing instability required an adjusted 2.45 weeks (95% confidence interval, 0.16-4.74) longer to achieve initial viral suppression and had greater odds of missing 5 or more doses of antiretroviral medications (adjusted odds ratio, 2.09; 95% confidence interval, 1.07-4.09) and having a detectable viral load at delivery (adjusted odds ratio, 2.13; 95% confidence interval, 1.02-4.47). Housing instability among pregnant individuals living with HIV is common and is associated with decreased virologic control during pregnancy. Given the association between virologic control and perinatal transmission, housing instability may be an important social determinant of HIV-related perinatal outcomes. Addressing housing instability during pregnancy may be a critical avenue to improve maternal and neonatal health and reduce the risk of perinatal transmission.