RTI uses cookies to offer you the best experience online. By clicking “accept” on this website, you opt in and you agree to the use of cookies. If you would like to know more about how RTI uses cookies and how to manage them please view our Privacy Policy here. You can “opt out” or change your mind by visiting: http://optout.aboutads.info/. Click “accept” to agree.
Short and longer-term impacts of health insurance on catastrophic health expenditures in Kwara State, Nigeria
Okunogbe, A., Hähnle, J., Rotimi, B. F., Akande, T. M., & Janssens, W. (2022). Short and longer-term impacts of health insurance on catastrophic health expenditures in Kwara State, Nigeria. BMC Health Services Research, 22(1), Article 1557. https://doi.org/10.1186/s12913-022-08917-z
BACKGROUND: Out- of-pocket health expenditures (OOPs) constitute a significant proportion of total health expenditures in many low- and middle-income countries (LMICs), leading to an increased likelihood of exposure to financial catastrophe in the event of illness. Health insurance has the potential to reduce catastrophic health expenditures (CHE), but rigorous evidence of its sustained impact is limited, especially in LMICs. This study examined the short- and longer-term effects of a health insurance program in Kwara State, Nigeria on CHE.
METHODS: The analysis is based on a panel dataset consisting of 3 waves of household surveys in program and comparison areas. The balanced data consists of 1,039 households and 3,450 individuals. We employed a difference-in-differences (DiD) regression approach to estimate intention-to-treat effects, and then computed average treatment effects on the treated by combining DiD with propensity score weighting and an instrumental variables analysis. CHE was measured as OOPs exceeding 10% of household consumption and 40% of capacity-to-pay (CTP).
RESULTS: Using 10% of consumption as a CHE measure, we found that living in the program area was associated with a 4.3 percentage point (pp) decrease in CHE occurrence (p < 0.05), while the effect on insured households was 5.7 pp (p < 0.05). The longer-term impact four years after program introduction was not significant. Heterogeneity analyses show a reduction in CHE of 7.2 pp (p < 0.01) in the short-term for the poorest tercile. No significant effects were found for the middle and richest terciles, nor in the longer-term. Households with a chronically ill member experienced a reduction in CHE of 9.4 pp (p < 0.01) in the short-term, but not in the longer-term. Most estimates based on the 40% of CTP measure were not statistically significant.
CONCLUSION: These findings highlight the critical role of health insurance in reducing the likelihood of catastrophic health expenditures, especially for vulnerable populations such as the poor and the chronically ill, and by extension in achieving universal health coverage. They also show that the beneficial impacts of health insurance may attenuate over time, as households potentially adjust their health-seeking behavior to the new scheme.