Cost-effectiveness analysis of proactive home visits compared with site-based community health worker care on antenatal care outcomes in Mali: a cluster-randomised trial.

Abstract

Introduction: Despite recommendations from the WHO, antenatal care (ANC) coverage remains low in many low-income and middle-income countries (LMICs). Community health workers (CHWs) can play an important role in expanding ANC coverage through pregnancy identification, provision of health education, screening for complications, delivery of therapeutic care and referral to higher levels of care. However, despite the success of CHW programmes in various countries, WHO has called for additional research to develop evidence-based models that optimise CHW service delivery and that can be replicated across geographies.

Methods: The ProCCM Trial was a cluster-randomised controlled trial to compare proactive home visits by CHWs (intervention, 69 village clusters) to the provision of CHW care at community fixed sites only (control, 68 village clusters) in the Bankass health district in Central Mali. In this study, we conducted a cost-effectiveness analysis of proactive CHW home visits in improving ANC utilisation, a secondary outcome of the ProCCM trial. We analysed five ANC outcomes: (1) number of ANC contacts, (2) at least one ANC contact, (3) at least four ANC contacts, (4) at least eight ANC contacts and (5) ANC initiated in the first trimester. We assumed two perspectives, a CHW programme's and the Full ANC programme's perspective, which included facility-based as well as community-based ANC. We estimated programme costs, incremental cost-effectiveness ratios (ICERs) and probabilities of the intervention being more cost-effective than the control at different willingness-to-pay (WTP) thresholds.

Results: Proactive home visits were cost-saving from the CHW programme's perspective (ICERs: -$21.39 to -$79.20 per ANC utilisation outcome) and from the Full ANC programme perspective (ICERs: -$1.70 to -$6.30 per ANC utilisation outcome) compared with the fixed-site CHW care. The likelihood of the intervention being more cost-effective than the control was 100% at WTP thresholds $0 per ANC utilisation outcome and between $12.5 and $50.00 per ANC utilisation outcome in the CHW- and Full ANC programme perspectives, respectively.

Conclusion: Our results provide evidence that proactive home visits produce more value per dollar spent as a means of improving the uptake of ANC services compared with fixed-site CHW services.

Trial registration number: NCT02694055.