Table 4.

Determinants of income-related health inequalities in the COVID-19 era

	CIa ()	Elasticity ()	Contribution ()	Contribution (%)
Continuous covariates
Age in years	0.023***	0.212*	0.005	-4.08
	(0.003)	(0.115)	(0.003)
Years of schooling	0.057***	-0.033	-0.002	1.63
	(0.005)	(0.129)	(0.007)
Log of per capita household income	0.202***	-0.277**	-0.056**	45.67
	(0.007)	(0.119)	(0.024)
Qualitative covariates
Male (reference = female)	0.146***	0.015	0.002	-1.63
	(0.030)	(0.037)	(0.005)
Race (reference = White)
African	-0.286***	0.558***	-0.159***	129.67
	(0.034)	(0.107)	(0.032)
Coloured	0.028*	0.015	< 0.001	-0.08
	(0.017)	(0.016)	(0.001)
Asian	0.022*	0.007	< 0.001	-0.08
	(0.013)	(0.006)	(0.000)
Employed and earning income (reference = not employed/not earning income)	0.441***	-0.037	-0.016	13.05
	(0.026)	(0.037)	(0.016)
Dwelling type (reference = formal dwelling)
Traditional dwelling (e.g. hut)	-0.078***	-0.007	0.001	-0.82
	(0.014)	(0.011)	(0.001)
Informal dwelling (e.g. shack)	-0.075***	-0.009	0.001	-0.82
	(0.020)	(0.018)	(0.001)
Chronic illness (reference = no chronic illness)	-0.021	0.053**	-0.001	0.82
	(0.022)	(0.021)	(0.001)
Household experienced hunger (reference = no household experience of hunger)	-0.217***	0.051**	-0.011**	8.97
	(0.021)	(0.024)	(0.005)
Has breathing problem (reference = has no breathing problem)	0.005	0.031**	< 0.001	-0.08
	(0.014)	(0.013)	(0.000)
Has fever, sore throat or cough (reference = has none of these symptoms)	0.03	0.037**	0.001	-0.82
	(0.019)	(0.017)	(0.001)
Error			0.236***
			(0.035)

^aConcentration index; Estimates weighted by NIDS-CRAM design weights; Estimation sample = 4 124; Jackknife standard errors with 1 014 replications in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1