Assessing the magnitude, distribution and determinants of catastrophic health expenditure in urban Lucknow, North India

Abstract

Introduction To assess the magnitude, distribution, and determinants of catastrophic health expenditures (CHE) of households in urban Lucknow, North India. Methods A cohort of 400 households was selected by 2-step cluster sampling and baseline demographic survey was done followed by two six-monthly health surveys. CHE was defined as health expenditures ≥10% of household's capacity to pay, measured by non-subsistence spending. Results From December 2011 to June 2012, 157/400 (39.25%) households reported ≥1 episodes of illness, with households suffering sickness in the first survey at increased risk for it in the second (Crude Odd's Ratio = 3.33, 95% CI: 2.02–5.45; p value <0.0001). Mean sickness days without hospitalization were 13.13 ± 36 per household. In 24 (6%) households, there was ≥1 hospitalization. Health expenditure was entirely met through out of pocket payments (OOP). CHE occurred in 45 (11.25%) households, with statistically significant differences across per capita income quintiles (p = 0.036) and 60% falling in the lower two. On logistic regression model, adjusting for per capital income quintile, CHE was associated with hospitalization (Adjusted OR = 100, 95% CI: 25.00–333.33; p < 0.0001) and >13 sickness days without hospitalization (Adjusted OR = 4.21, 95% CI: 1.862–9.524; p = 0.001). Conclusions Since not only hospitalization but also prolonged sickness days without hospitalization was associated with increased risk of CHE, and since almost half the households have sickness, steps should be taken to protect all households from financial hardship through tax based health financing, social health insurance or other forms of prepayment, as currently all health expenses were met through OOP payments.