Impact of community-based health insurance in low- and middle-income countries: A systematic review and meta-analysis
Paul Eze,Stanley Ilechukwu,Lucky Osaheni Lawani
DOI: https://doi.org/10.1371/journal.pone.0287600
IF: 3.7
2023-06-27
PLoS ONE
Abstract:Background: To systematically evaluate the empirical evidence on the impact of community-based health insurance (CBHI) on healthcare utilization and financial risk protection in low- and middle-income countries (LMIC). Methods: We searched PubMed, CINAHL, Cochrane CENTRAL, CNKI, PsycINFO, Scopus, WHO Global Index Medicus, and Web of Science including grey literature, Google Scholar®, and citation tracking for randomized controlled trials (RCTs), non-RCTs, and quasi-experimental studies that evaluated the impact of CBHI schemes on healthcare utilization and financial risk protection in LMICs. We assessed the risk of bias using Cochrane's Risk of Bias 2.0 and Risk of Bias in Non-randomized Studies of Interventions tools for RCTs and quasi/non-RCTs, respectively. We also performed a narrative synthesis of all included studies and meta-analyses of comparable studies using random-effects models. We pre-registered our study protocol on PROSPERO: CRD42022362796. Results: We identified 61 articles: 49 peer-reviewed publications, 10 working papers, 1 preprint, and 1 graduate dissertation covering a total of 221,568 households (1,012,542 persons) across 20 LMICs. Overall, CBHI schemes in LMICs substantially improved healthcare utilization, especially outpatient services, and improved financial risk protection in 24 out of 43 studies. Pooled estimates showed that insured households had higher odds of healthcare utilization (AOR = 1.60, 95% CI: 1.04-2.47), use of outpatient health services (AOR = 1.58, 95% CI: 1.22-2.05), and health facility delivery (AOR = 2.21, 95% CI: 1.61-3.02), but insignificant increase in inpatient hospitalization (AOR = 1.53, 95% CI: 0.74-3.14). The insured households had lower out-of-pocket health expenditure (AOR = 0.94, 95% CI: 0.92-0.97), lower incidence of catastrophic health expenditure at 10% total household expenditure (AOR = 0.69, 95% CI: 0.54-0.88), and 40% non-food expenditure (AOR = 0.72, 95% CI: 0.54-0.96). The main limitations of our study are the limited data available for meta-analyses and high heterogeneity persisted in subgroup and sensitivity analyses. Conclusions: Our study shows that CBHI generally improves healthcare utilization but inconsistently delivers financial protection from health expenditure shocks. With pragmatic context-specific policies and operational modifications, CBHI could be a promising mechanism for achieving universal health coverage (UHC) in LMICs.