Table 4.

Two-part model for hospitalization costs of individuals using the private sector. ISA-Capital 2003–2015, São Paulo city, Brazil.

Variables	Categories	Logit			GLM			Marginal Effects
		β	SE	Sig.	β	SE	Sig.	dy/dx	SE	Sig.
Sex	(1 = female)	0.455	0.201	*	−0.098	0.108		18.67	11.63
Age	(years)	0.009	0.005		0.009	0.002	**	1.01	0.32	**
Skin color/ethnicity	(1 = white)	0.108	0.194		0.149	0.136		13.75	12.48
Marital status	(1 = accompanied)	0.678	0.193	***	0.089	0.117		40.49	11.80	**
Education	(years)	0.031	0.019		0.004	0.014		1.88	1.27
Occupational status	(1 = employed)	−0.313	0.189		0.026	0.101		−15.08	11.55
Income per capita	($PPP)	0.258	0.091	**	−0.033	0.061		11.79	5.87	*
Leisure-time PA	(1 = yes)	−0.494	0.225	*	−0.221	0.121		−37.89	13.72	**
Time spent in sitting position (ref. = lower tertile)	(1 = middle tertile)	−0.001	0.199		−0.041	0.129		−1.80	10.11
	(1 = upper tertile)	0.442	0.198	*	0.225	0.121		40.92	15.25	**
Obesity	(1 = yes)	−0.008	0.306		0.385	0.159	*	20.29	18.22
Tobacco use	(1 = yes)	−0.130	0.187		−0.132	0.111		−14.01	11.67
Health insurance	(1 = yes)	1.355	0.191	***	−0.114	0.093		65.10	11.48	**
Hypertension	(1 = yes)	0.130	0.207		0.318	0.132	*	17.87	13.14
Diabetes	(1 = yes)	0.177	0.512		−0.416	0.305		−13.08	31.63
CVD	(1 = yes)	0.681	0.545		0.058	0.306		38.96	33.28
Year (ref. = 2003)	(1 = 2008)	−0.068	0.207		0.089	0.129		1.20	12.83
	(1 = 2015)	0.075	0.218		−0.039	0.118		1.79	13.03
Constant		−7.045	0.602		7.415	0.400
N		7068			262			7068

GLM = generalized linear model (gamma family with log-link); β = regression coefficient; SE = robust standard error; dy/dx = average marginal effects. Statistical significance: *** = p < 0.001; ** = p < 0.01; * = p < 0.05.

PA = physical activity; CVD = cardiovascular diseases.