Predictors of catastrophic out-of-pocket health expenditure in rural Egypt: application of the heteroskedastic probit model.

Abdel-Rahman, S., F. Shoaeb, M. N. A. Fattah, and M. R. Abonazel, "Predictors of catastrophic out-of-pocket health expenditure in rural Egypt: application of the heteroskedastic probit model.", The Journal of the Egyptian Public Health Association, vol. 96, issue 1, pp. 23, 2021.


BACKGROUND: Out-of-pocket (OOP) health expenditure is a pressing issue in Egypt and far exceeds half of Egypt's total health spending, threatening the economic viability, and long-term sustainability of Egyptian households. Targeting households at risk of catastrophic health payments based on their characteristics is an obvious pathway to mitigate the impoverishing impacts of OOP health payments on livelihoods. This study was conducted to identify the risk factors of incurring catastrophic health payments hoping to formulate appropriate policies to protect households against financial catastrophes.

METHODS: Using data derived from the Egyptian Household Income, Expenditure, and Consumption Survey (HIECS), a multiplicative heteroskedastic probit model is applied to account for heteroskedasticity and avoid biased and inconsistent estimates.

RESULTS: Accounting for heteroskedasticity induces notable differences in marginal effects and demonstrates that the impact of some core variables is underestimated and insignificant and in the opposite direction in the homoscedastic probit model. Moreover, our results demonstrate the principal factors besides health status and socioeconomic characteristics responsible for incurring catastrophic health expenditure, such as the use of health services provided by the private sector, which has a dramatic effect on encountering catastrophic health payments.

CONCLUSIONS: The marked differences between estimates of probit and heteroskedastic probit models emphasize the importance of investigating homoscedasticity assumption to avoid policies based on incorrect evidence. Many policies can be built upon our findings, such as enhancing the role of social health insurances in rural areas, expanding health coverage for poor households and chronically ill household heads, and providing adequate financial coverage for households with a high proportion of elderly, sick members, and females. Also, there is an urgent need to limit OOP health payments absorbed by private sector to achieve an acceptable level of fair financing.