Table 5.

Catastrophic health expenditure and unmet need

VARIABLES	Unmet need
CHE10[t-1]	-0.01	-0.01
	(0.00)	(0.01)
Health expenditure [t-1]			-0.06*	-0.07#
			(0.02)	(0.04)
Age	0.18**	0.13**	0.18**	0.13**
	(0.01)	(0.01)	(0.01)	(0.01)
Age2	-0.00**	-0.00*	-0.00**	-0.00*
	(0.00)	(0.00)	(0.00)	(0.00)
Age3	0.00**	0.00**	0.00**	0.00**
	(0.00)	(0.00)	(0.00)	(0.00)
CHE10[t-1] * Age		0.00
		(0.00)
CHE10[t-1] * Age2		-0.00
		(0.00)
CHE10[t-1] * Age3		0.00
		(0.00)
Health expenditure [t-1] * Age				0.00
				(0.00)
Health expenditure [t-1] * Age2				-0.00
				(0.00)
Health expenditure [t-1] * Age3				0.00
				(0.00)
Observations	23,458
Individuals	4,190

Note: CHE10 denotes catastrophic health expenditure at a 10% threshold; Health expenditure (% of total consumption) is mean-centralised and transformed by the inverse hyperbolic sine transformation; Estimates by fixed-effects linear probability models; ** p < 0.01, * p < 0.05, ^# p < 0.1; Values are coefficients with cluster-robust standard errors in parentheses; Controlled for employment status of respondents (working or not), marital status of respondents (single or not), educational attainment of respondents (bachelor or higher), house ownership, household size, current cigarette smoking, current alcohol consumption, exercise days, individual fixed-effects, and city-by-year fixed-effects; Age of respondents is centralised when interactions with CHE were analysed; Weighted by longitudinal weights to address for attrition bias; singleton observations are not used for estimations.