Preferences of pregnant women attending antenatal care regarding prevention of mother-to-child HIV transmission service delivery models in Ethiopia: Applying a discrete choice experiment technique

Abstract

Background: The prevention of mother-to-child transmission (PMTCT) of HIV/AIDS, often delivered as part of a comprehensive range of antenatal care services, is a vital part of the HIV response and well aligned with the universal health coverage agenda. Expanding service utilisation is increasingly a priority, especially in resource-limited settings. In Ethiopia, PMTCT service is a priority strategy to eliminate mother-to-child transmission (MTCT) (MOH, 2021). Understanding the demand-side factors that drive this low service uptake is important for policymakers and service providers to increase demand and avail client-centric health systems. Objectives: This paper assesses the preferences and drivers of choice regarding the attributes of PMTCT service delivery models, considering urban high-HIV prevalence and rural low-HIV prevalence settings in Ethiopia. Methods: A total of 275 pregnant women attending antenatal care (ANC) across twelve health facilities (six urban high-HIV prevalence and six rural low-HIV prevalence) were interviewed, applying a discrete choice experiment (DCE) technique. Six attributes: pretest counselling (individual/couple); service integration (family planning/antiretroviral treatment); disclosure counselling (one-to-one/partner); waiting time (30 minutes/45 minutes/1 hour/2 hours); cost (free/paying 25 ETB/paying 50 ETB/receiving 100 ETB) and PMTCT service location (health post/health centre/general hospital/specialised hospital) were included in the design, generated to maximise D-efficiency. A conditional random effect logit econometric model was employed. The econometric model analysis introduced an interaction term (urban versus rural setting). Results: Couple pretest counselling was preferred over individual pretest counselling (OR 1.23, p = 0.000). A pregnant woman waiting for 1 hour and 2 hours was less likely to prefer the PMTCT service than waiting for 30 minutes (OR 0.75, p=0.001; OR 0.76, p=0.000). The respondents preferred not to pay for the services (USD 1.27, USD 2.54). Pregnant women preferred PMTCT services at the health centre (OR 1.26, p=0.001). The odds of choosing couple pretest counselling by a pregnant woman from the rural areas were lower than the urban respondents (OR 0.77, p=0.003). Urban-dwelling pregnant women were less likely to prefer waiting for longer time periods (OR 0.72, p=0.72, p=0.04). Pregnant women from rural areas were less willing to pay USD 2.54 for the service (OR 0.52, p=0.000) and more willing to receive a payment of USD 5.08 (OR 2.09, p=0.000). On the other hand, women from urban areas were more willing to pay USD 2.54 (OR 1.49, p=0.013) and were less likely to receive a payment of USD 5.08 for the PMTCT service (OD 0.40, p=0.000). Conclusion/Policy Implication: Pretest counselling, waiting time, service fees, and location were the critical attributes impacting pregnant women’s preferences towards PMTCT service. Particularly, the preferences of urban and rural pregnant women were varied for the PMTCT service attributes of service integration, waiting time, cost/service fees and location.