Table 2.

Linear regression model and generalized linear model for incidence-based medical costs.

Model	Variable	Cervical Cancer (C53)				CIS (D06)
		Univariate		Multivariate		Univariate		Multivariate
		β	p Value	β	p Value	β	p Value	β	p Value
Linear Regression Model	Age (reference = 20–39 years)
	40–59	0.509	<0.0001	0.501	<0.0001	0.306	<0.0001	0.263	<0.0001
	60+	0.617	<0.0001	0.591	<0.0001	0.95	<0.0001	0.782	<0.0001
	Year (reference = 2013)
	2014	0.036	0.219	0.035	0.223	0.049	0.007	0.045	0.008
	2015	0.124	<0.0001	0.129	<0.0001	0.077	<0.0001	0.084	<0.0001
	2016	0.13	<0.0001	0.131	<0.0001	0.184	<0.0001	0.186	<0.0001
	CCI (reference = 0)
	1	0.164	<0.0001	0.118	0.000	0.42	<0.0001	0.27	<0.0001
	2	0.131	<0.0001	0.049	0.134	0.744	<0.0001	0.571	<0.0001
	3	0.278	<0.0001	0.17	0.000	1.149	<0.0001	0.827	<0.0001
	AIC			500,560				788,947
Generalized Linear Model	Age (reference = 20–39 years)
	40–59	0.348	<0.0001	0.335	<0.0001	0.469	<0.0001	0.398	<0.0001
	60+	0.35	<0.0001	0.321	<0.0001	0.998	<0.0001	0.821	<0.0001
	Year (reference = 2013)
	2014	0.036	0.12	0.038	0.093	0.09	<0.0001	0.076	<0.0001
	2015	0.177	<0.0001	0.177	<0.0001	0.068	<0.0001	0.076	<0.0001
	2016	0.205	<0.0001	0.198	<0.0001	0.19	<0.0001	0.181	<0.0001
	CCI (reference = 0)
	1	0.093	0.000	0.059	0.001	0.39	<0.0001	0.22	<0.0001
	2	0.089	0.001	0.046	0.014	0.924	<0.0001	0.785	<0.0001
	3	0.198	<0.0001	0.134	<0.0001	1.223	<0.0001	0.968	<0.0001
	AIC			542,775				872,825

CIS, carcinoma in situ of cervix uteri; CCI, Charlson Comorbidity Index; AIC, Akaike Information Criterion.