Bolivia programme evaluation of a package to reach an underserved population: Community-based maternal and newborn care economic analysis

Abstract

To address inequitable access to health services of indigenous communities in the Bolivian highlands, the Bolivian Ministry of Health, with the support of Save the Children-Saving Newborn Lives, conducted operational research to identify, implement and test a package of maternal and newborn interventions using locally recruited, volunteer Community Health Workers (vCHW) between 2008 and 2010. The additional annual economic and financial costs of the intervention were estimated from the perspective of the Bolivian Ministry of Health in two municipalities. The cost of intervention-stimulated increases in facility attendance was estimated with national surveillance data using a pre-post comparison, adjusted for secular trends in facility attendance. Three scale-up scenarios were modelled by varying the levels of coverage and the number (per mother and child pair) and frequency of home visits. Average cost per mother and average cost per home visit are presented in constant 2015 US$. Eighteen per cent of expectant mothers in the catchment area were visited at least once. The annualized additional financial cost of the community-based intervention across both municipalities was $43 449 of which 3% ($1324) was intervention design, 20% ($8474) set-up and 77% ($33 651) implementation. Drivers of additional costs were additional paid staff (68%), 81% of which was for management and support by local implementing partner and 19% of which was for vCHW supervision. The annual financial cost per vCHW was $595. Modelled scale-up scenarios highlight potential efficiency gains. Recognizing local imperatives to reduce inequalities by targeting underserved populations, the observed low coverage by vCHWs resulted in a high cost per mother and child pair ($296). This evaluation raises important questions about this model's ability to achieve its ultimate goals of reducing neonatal mortality and inequalities through behaviour change and increased care seeking and has served to inform innovative alternative models, better equipped to tackle stagnant inequitable access to care.

Economic evaluation methods used in home‐visiting interventions: A systematic search and review

“…Two studies used parallel cluster RCT designs where randomisation was by geographical area (Barger et al, 2017;Mathewos et al, 2017). The study by Mathewos and colleagues used a cluster randomised trial, and randomly allocated 22 clusters in two regions to intervention and control (Mathewos et al, 2017).…”

“…In the paper by Barger and colleagues, interventions were conducted in two municipalities, and were compared to two control municipalities. The authors do not state how the regions were chosen, and there was no information supplied in the paper on the comparability of the different regions (Barger et al, 2017). The economic evaluation method used in this paper was a cost-comparison analysis.…”

“…Societal and governmental. In a third paper, the costs of three scale-up scenarios were assessed (Barger et al, 2017). Barger and colleagues found that high costs per home-visitor combined with low coverage of the mother/ child dyad impeded the study's aim to lowering infant mortality.…”

“…Whilst none of the included papers explicitly state that they were presenting a budget impact analysis, in the paper by Barger and colleagues (Barger et al, 2017) the additional annual costs to the municipality of an intervention to improve services to an underserved population of women during pregnancy in Bolivia were presented. Annualised average costs per mother for the intervention arms were compared with costs of the comparison group, and the feasibility of the study was measured at scale in terms the implications on finances and human resources.…”

“…Fourteen included studies compared the intervention to care-asusual or standard care (Browne et al, 2001;Au et al, 2006;Barlow et al, 2007Barlow et al, , 2019Aracena et al, 2009;McIntosh et al, 2009;Olds et al, 2011;Martin & Lotspeich, 2014;Thanh et al, 2015;Dalziel et al, 2015;Barger et al, 2017;Sonuga-Barke et al, 2018;Deidda et al, 2018;Bell et al, 2019;Kuklinski, Crowley, et al, 2020). Four studies compared two different home intervention strategies (Ammerman et al, 2017;Dijkstra et al, 2018;Mathewos et al, 2017;Sharac et al, 2011).…”