Taslima Rahman

Background: All United Nations member countries, including Bangladesh, are committed to ensuring full financial risk protection to achieve Universal Health Coverage (UHC) by 2030. This study examines how Bangladesh is progressing to achieve financial risk protection in health care over time.

Method: We analyzed data from the latest three rounds of nationally representative Household Income and Expenditure Surveys (HIES) in Bangladesh (2005, 2010, and 2016) with sample sizes of 10,080, 12,240, and 46,076 households, respectively. We refined the normative food, housing (rent), and utilities approach developed by the WHO Barcelona Office for Health Systems Strengthening to measure the levels and distributions of catastrophic health expenditure (CHE) and impoverishment through a detailed exploration of households with zero out-of-pocket (OOP) health expenditure. We also examined the level and distribution of CHE through the traditional budget-share and capacity-to-pay approaches. Due to substantial differences between OOP health expenditures calculated from health (OOPh) and consumption modules (OOPc) in HIES, we employed three models for estimating incidences of financial risk protection indicators consisting of OOP in both their numerator and the denominator; Model 1: OOPh in the numerator but OOPc in the denominator; Model 2: OOPc in both numerator and denominator; Model 3: OOPh in both numerator and denominator.

Findings: OOP expenditure soared during the study period, particularly between 2010 and 2016 (from US$106.3 to US$243.8 in models 1 and 3; from US$68.5 to US$106.3 in model 2). CHE incidence showed increasing trend over three periods: 13.6%, 13.8%, 19.2% (model 1); and 11.5%, 11.9%, 16.6% (model 3) with a minor variation (model 2). Between 2005 and 2016, the incidence of impoverishment due to OOP also increased from 4%-5.8% (5.6-9.2 million individuals), 1.2%-1.3% (1.7-2.1 million), and 1.3%-2.6% (2.1-3.6 million) in model 1, 2, and 3, respectively. Further impoverishment of the poor households was a more severe problem than impoverishment of the non-poor. Additionally, at least 5% of households were at-risk of impoverishment in all three models during the study period. Households who did not spend on health care declined from about 50% in 2005 and 2010 separately to 25% (models 1 and 3). However, the proportion of households who forgo health care due to financial reasons always remained small (less than 1%). The poorest households were invariably the least financially protected group throughout the study period while in 2016, the burden on households with chronically ill individuals was also notable. CHE measured through 10% of budget share threshold (the Sustainable Development Goals indicator 3.8.2) shows the highest incidence among all methods in 2016 (25.6%, 10.4%, 24.6% in models 1, 2, and 3 respectively) but has a pro-rich distribution.

Conclusion: Financial protection in health care in Bangladesh exhibit deteriorated trajectory over 2005-2016. The poorest and chronically ill households bear a disproportionate burden of OOP expenditure. Reversing the worsening trends of CHE and impoverishment and addressing the inequities in their distributions in Bangladesh will require a significant shift from the country's excessive dependence on private OOP sources to public sources in financing health care.