Trajectories of HIV Visit Engagement During the Perinatal Period among Women in Lilongwe, Malawi

Longitudinal patterns of engagement in care among women living with HIV (WHIV) during the perinatal period are poorly understood. We employed group-based trajectory modeling to (1) describe trajectories of HIV visit engagement; and (2) identify predictors of membership in suboptimal care trajectories. Data came from a prospective cohort study across five urban clinics in Lilongwe, Malawi conducted between February 2020 and August 2022. Participants were followed from entry into antenatal care (ANC) through 9 months postpartum. Trajectories were defined for 375 participants based on whether (1) ≥1 HIV care visit was attended; or (2) ≥1 HIV care visit was missed for each three-month interval during follow-up. Over a maximum of 15 months of follow-up, we identified two distinct trajectory groups for each HIV visit engagement outcome: "lower" (7%) vs. "consistently high" (93%) probability of attending a visit groups and "higher" (86%) vs. "consistently low" (14%) probability of missing a visit groups. Age <25 years (odds ratio (OR) = 3.82, 95% CI 1.25, 11.73) or >1 hour of travel time to the clinic (OR = 3.54, 95% CI 1.44, 8.69) were associated with membership in the "lower" probability of attending a visit trajectory group. Enrollment after the start of the COVID-19 pandemic (OR = 0.23, 95% CI 0.07, 0.74) and higher gestational age at enrollment (OR = 0.28, 95% CI 0.09, 0.86) were protective against membership in the "lower" probability of attending a visit and the "higher" probability of missing a visit trajectory groups, respectively. In settings where missed visits are common, visit attendance may be an important indicator of care engagement.