Financial Incentives and Coverage of Child Health Interventions: a Systematic Review and Meta-analysis

MAIN FINDINGS: Headline Findings: a summary statement Financial incentives can lead to increased uptake and coverage of a number of health interventions and practices for children under the age of five, but the evidence base is limited and generally of low quality. Evidence Base The authors include 25 studies in this review, of which 13 took place in Latin America and the Caribbean, eight in Sub-Saharan Africa and four in South Asia. Of the studies, 48 percent assess cash-transfer programmes (41 percent conditional cash-transfer programmes and 7 percent unconditional cash-transfer programmes), 22 percent assess the effects of removing user fees and 25 percent assess microcredit programmes. In terms of child health coverage indicators, 12 studies look at vaccination coverage, 19 studies look at health-care use, four studies look at the management of diarrhoeal diseases, four studies look at other preventive health and three studies look at breastfeeding practices. Implications for policy and practice Breastfeeding practices: Financial incentives can have a positive impact on some breastfeeding practices. Evidence from two studies in Bolivia and Ghana shows that microcredit programmes conditional on participation of mothers in health and nutrition education lead to a 22 percent increase in new-born babies receiving colostrum, on average (mean risk difference [MD] = 0.22; 95 percent confidence interval [CI]: 0.08, 0.35). However, two studies of conditional and unconditional microcredit interventions in Honduras and Ecuador found no significant effects on prevalence of breastfeeding. The overall size and quality of the evidence base is low.Vaccination coverage: Evidence from conditional cash-transfer and conditional microcredit programmes shows no impact on vaccination coverage, specifically BCG, DPT-1, DPT-3, measles or polio. Evidence from four conditional cash-transfer programmes suggests that these programmes may increase coverage of full age-appropriate vaccination, but the average effect is not statistically significant (MD = 0.05; CI: −0.01, 0.10). The available evidence base is only of moderate to low quality. Use of preventative health care: Evidence from five conditional cash-transfer programmes shows a 14 percent increase, on average, in the use of preventive health care by children under the age of five (MD = 0.14; CI: −0.00, 0.29), though the average effect is not statistically significant. The authors also note that the effects vary across studies and the evidence is of variable quality. Removal of user fees lead to a 33 percent increase in the prevalence of health-care use, on average (MD = 0.33; CI: 0.24, 0.43), and a 99 percent increase in the frequency of health-care use, on average (MD = 0.99; CI: 0.71, 1.27)—but, again, the quality of evidence is low.Management of child diarrhoea: The evidence for the impact of financial-incentive programmes on management of childhood diarrhoea is poor, with only single studies looking at the impact of either unconditional or conditional microcredit on use of oral rehydration or care-seeking during diarrhoea. Two studies of conditional microcredit interventions suggest no effect of the intervention on the practice of continuing child feeding during diarrhoea. Other preventative health care: Evidence from two conditional cash-transfer programmes finds a 16 percent increase in the average coverage of Vitamin A supplements, but the results are not statistically significant (MD = 0.16; CI: −0.01, 0.34). Implications for further research The review team calls for future systematic reviews to assess the impact of financial incentives on child morbidity and mortality. They also state that future primary research should try to isolate the effects of the financial and non-financial components of programmes. BACKGROUND: Financial incentives that provide direct or indirect monetary incentives to households, such as voucher schemes or cash-transfer programmes, are commonly used to help overcome poverty and improve health of populations in low- and middle-income countries. One channel by which they are expected to do this is through the removal of financial barriers to health care. This is expected to improve access, uptake and coverage of health services such as immunisation, treatment of diarrhoea and preventative health interventions. Previous reviews have looked at the impact of different types of financial incentives on health outcomes. However, this is the first review to comprehensively review the effectiveness of financial-incentive programmes in low- and middle-income countries targeted at the coverage and uptake of health services and behaviours of children under the age of five. RESEARCH OBJECTIVES: The authors aimed to investigate the effects of six different types of financial-incentive programmes—unconditional cash transfers, conditional cash transfers, unconditional microcredit, conditional microcredit, unconditional voucher, conditional voucher and user-fee removal—on uptake and coverage of health interventions for children under the age of five. METHODOLOGY: The authors included randomised controlled trials, cluster-randomised controlled trials and observational studies that assessed the effect of financial incentives on indicators of health-care coverage for children under the age of five. Eligible studies reported the effect of financial incentives on the following types of health-coverage outcomes: breastfeeding, vaccination, health-care use, diarrhoeal diseases and other preventive health interventions. The authors used the Child Health Evaluation Reference Group (CHERG) systematic review guidelines and searched for studies published in either peer-reviewed journals or institutional/commissioned reports covering the period up to 1 September 2012. They searched in the databases PubMed, EMBASE and AMED and used Google Scholar and the Microfinance Gateway library to complement this search. They assessed the quality of the included studies and used meta-analysis to synthesise effect sizes.