Background: Among HIV-infected women, perinatal depression compromises clinical, maternal, and child health outcomes. Antiretroviral therapy (ART) is associated with lower depression symptom severity but the uniformity of effect through pregnancy and postpartum periods is unknown.

Methods: We analyzed prospective data from 447 HIV-infected women (18–49 years) initiating ART in rural Uganda (2005–2012). Participants completed blood work and comprehensive questionnaires quarterly. Pregnancy status was assessed by self-report. Analysis time periods were defined as currently pregnant, postpartum (0–12 months post-pregnancy outcome), or non–pregnancy-related. Depression symptom severity was measured using a modified Hopkins Symptom Checklist 15, with scores ranging from 1 to 4. Probable depression was defined as >1.75. Linear regression with generalized estimating equations was used to compare mean depression scores over the 3 periods.

Results: At enrollment, median age was 32 years (interquartile range: 27–37), median CD4 count was 160 cells per cubic millimeter (interquartile range: 95–245), and mean depression score was 1.75 (s = 0.58) (39% with probable depression). Over 4.1 median years of follow-up, 104 women experienced 151 pregnancies. Mean depression scores did not differ across the time periods (P = 0.75). Multivariable models yielded similar findings. Increasing time on ART, viral suppression, better physical health, and “never married” were independently associated with lower mean depression scores. Findings were consistent when assessing probable depression.

Conclusions: Although the lack of association between depression and perinatal periods is reassuring, high depression prevalence at treatment initiation and continued incidence across pregnancy and non–pregnancy-related periods of follow-up highlight the critical need for mental health services for HIV-infected women to optimize both maternal and perinatal health.