The impact of free antiretroviral therapy for pregnant non‐citizens and their infants in Botswana

INTRODUCTION: In December 2019, the Botswana government expanded free antiretroviral therapy (ART) to include non‐citizens. We evaluated the impact of this policy change on antenatal care (ANC), antiretroviral therapy coverage and adverse birth outcomes. METHODS: The Tsepamo Surveillance study collects data at up to 18 delivery sites in Botswana. We compared outcomes in citizens and non‐citizens living with HIV before and after antiretroviral therapy expansion to non‐citizens. Adverse birth outcomes included preterm delivery (PTD) <37 weeks, very preterm delivery (VPTD) <32 weeks, small for gestational age (SGA) <10th percentile, very small for gestational age (VSGA) <3rd percentile, stillbirth and neonatal death. Log‐binomial regression models were constructed to generate risk ratios. RESULTS: From August 2014 to September 2021, 45,576 (96.5%) citizens and 1513 (3.2%) non‐citizens living with HIV delivered; 954 (62.9%) non‐citizen deliveries were before the antiretroviral therapy expansion, and 562 (37.1%) were after. Non‐citizen ANC attendance among pregnant people living with HIV increased from 79.2% pre‐expansion to 87.2% post‐expansion (p<0.001), and became more similar to citizens (96.0% post‐expansion). Non‐citizens receiving any antenatal antiretroviral therapy increased from 65.5% pre‐expansion to 89.9% post‐expansion (p < 0.001), also more similar to citizens (97.2% post‐expansion). Infants born to non‐citizens with singleton gestations in the pre‐expansion period had significantly greater risk of PTD (aRR = 1.28, 95% CI, 1.11, 1.46), VPTD (aRR = 1.89, 95% CI, 1.43, 2.44) and neonatal death (aRR = 1.69, 95% CI, 1.03, 2.60), but reduced SGA risk (aRR = 0.75; 95% CI, 0.62, 0.89) compared with citizens. Post‐expansion, greater declines in most adverse outcomes were observed in non‐citizens, with largely similar outcomes between non‐citizens and citizens. Non‐significant differences were observed for non‐citizenship in PTD (aRR = 0.84, 95% CI, 0.66, 1.06), VPTD (aRR = 0.57, 95% CI, 0.28, 1.01), SGA (aRR = 0.91, 95% CI, 0.72, 1.13), VSGA (aRR = 0.87, 95% CI, 0.58, 1.25), stillbirth (aRR = 0.71, 95% CI, 0.35, 1.27) and neonatal death (aRR = 1.35, 95% CI, 0.60, 2.62). CONCLUSIONS: Following the expansion of free antiretroviral therapy to non‐citizens, gaps narrowed in ANC and antiretroviral therapy use in pregnancy between citizens and non‐citizens living with HIV. Disparities in adverse birth outcomes were no longer observed.