National and Provincial Estimates of Catastrophic Health Expenditure and its Determinants in Nepal
Arjun Kumar Thapa, Achyut Raj Pandey
Journal of Nepal Health Research Council · 2020 · DOI: 10.33314/jnhrc.v18i4.2392
Countries: Nepal
What Was Studied
This cross-sectional study conducted a secondary analysis of Nepal Living Standard Survey III (2012) data, collected by the Central Bureau of Statistics across the (then) 75 districts, later recategorised into Nepal's 7 provinces following the 2015 federal restructuring. From a nationally representative sample of 5,988 households (28,460 individuals), the authors extracted health expenditure and illness data for 7,911 individuals who reported an acute (past 30 days) or chronic (past 12 months) illness. Catastrophic health expenditure (CHE) was defined using the standard threshold of household out-of-pocket (OOP) health spending exceeding 40% of non-food expenditure. The study used descriptive statistics (national and 7-province CHE prevalence) and multivariate logistic regression (n=3,357 with complete data) to identify determinants of CHE, including household size, household head literacy, caste/ethnicity, consumption quintile, urban/rural residence, illness type, and type of health facility visited. Ethical approval was granted by the Ethical Review Board of the Nepal Health Research Council.
What They Found
Among households with a member reporting acute or chronic illness (7,911 cases, from a nationally representative sample of 5,988 households), 11.11% experienced catastrophic health expenditure (CHE), defined as out-of-pocket spending exceeding 40% of household non-food expenditure. CHE prevalence varied by province, from 9.4% in Province 2 (Madhesh) to 13.4% in Sudurpaschim Province and 13.3% in Karnali Province, with Lumbini (11.7%) and Province 1 (11.3%) also above the national average. Lowering the CHE threshold sharply raised prevalence: 14.56% at a 30% cut-off, 20.44% at 20%, and 33.22% at 10% of non-food expenditure. In adjusted logistic regression (n=3,357), the poorest consumption quintile had 7.2 times higher odds of CHE than the richest (OR=7.2, 95% CI 4.9-10.7), with a clear gradient down through the second (OR=5.0), third (OR=2.9), and fourth (OR=1.5) quintiles. Households with a chronic illness had 3.2 times higher odds of CHE than those with acute illness only (OR=3.2, 95% CI 2.6-4.1). An illiterate household head raised the odds of CHE by 1.3 times (OR=1.3, 95% CI 1.1-1.6), while each additional household member lowered the odds of CHE by about 30% (OR=0.7, 95% CI 0.7-0.8). Rural residence was associated with 1.8 times higher odds of CHE than urban residence in the regression results (OR=1.8, 95% CI 1.4-2.3) -- though the paper's own discussion section states the opposite direction, an internal inconsistency readers should note. Caste/ethnicity and province were not statistically significant predictors in the adjusted model, likely reflecting small per-province samples (e.g., Karnali n=384).
What This Means for Nepal
This study's data pre-date the Health Insurance Board's (HIB) National Health Insurance Programme (NHIP), whose pilot began only in 2016 in Kailali, Baglung and Ilam districts and which now faces a 2026 solvency crisis (arrears exceeding Rs 16bn). It therefore functions as a pre-insurance baseline against which NHIP's progress on financial protection can be judged: with OOP still at 54.2% of current health expenditure and 1.7% of Nepalis pushed into poverty by health costs every year (NHA 2019/20), the CHE risk profile identified here -- poorest quintile (OR=7.2), chronic illness (OR=3.2), illiterate household head (OR=1.3), and rural residence (OR=1.8) -- maps onto NHIP's own weakest-coverage geography, including Madhesh (~8% NHIP enrolment vs ~42% in Koshi). Concretely, HIB should prioritise premium-subsidy promotion and empanelment expansion in Karnali and Sudurpaschim provinces, which recorded the highest CHE here (13.3% and 13.4%) and which today have Nepal's weakest health workforce (0% of sanctioned physician and consultant posts filled in Karnali) -- using FCHVs and ward-level outreach to specifically reach illiterate-headed and rural households, since this study identifies them as facing the highest CHE risk.
Contextualisation
This is among the first nationally representative estimates of catastrophic health expenditure (CHE) in Nepal at both national and provincial level, and its national CHE estimate of 11.11% closely matches the ~10-11% figure commonly cited for Nepal's catastrophic health spending. Because the data (Nepal Living Standard Survey III, 2012) pre-date the Health Insurance Board's NHIP pilot (2016) and the 2015 federal restructuring, the study is best read as a pre-insurance baseline: with out-of-pocket spending still at 54.2% of current health expenditure and 1.7% of the population pushed into poverty by health costs every year (NHA 2019/20), the highest CHE recorded here -- in Sudurpaschim (13.4%) and Karnali (13.3%) provinces -- maps directly onto the same two provinces that today have Nepal's weakest health workforce (0% of sanctioned physician and consultant posts filled in Karnali) and among the lowest NHIP enrolment.