Incremental Healthcare Expenditures Associated with Thyroid Disorders among Individuals with Diabetes

Table 4

Intercept and parameter estimates for the presence of thyroid disorders from ordinary least squares regression on logged healthcare expenditures individuals with diabetes from the Medical Expenditure Panel Survey, 2007 and 2009.

Thyroid disease	Beta	SE	% Change	Significance
			(exp(beta)−1)
Total
Model 1	0.363	0.005	43.8	***
Model 2	0.295	0.004	34.3	***
Model 3	0.213	0.005	23.7	***
Model 4	0.194	0.006	21.4	***
Inpatient
Model 1	0.389	0.02	47.6	***
Model 2	0.299	0.016	34.9	***
Model 3	0.2	0.017	22.1	***
Model 4	0.152	0.02	16.4	***
Emergency room
Model 1	0.124	0.006	13.2	***
Model 2	0.089	0.007	9.3	***
Model 3	0.01	0.006	1.0
Model 4	−0.02	0.005	−2.0	***
Outpatient
Model 1	0.393	0.011	48.1	***
Model 2	0.299	0.016	34.9	***
Model 3	0.208	0.008	23.1	***
Model 4	0.196	0.009	21.7	***
Prescription drug
Model 1	0.348	0.009	41.6	***
Model 2	0.282	0.009	32.6	***
Model 3	0.193	0.01	21.3	***
Model 4	0.179	0.01	19.6	***
Other
Model 1	0.752	0.03	112.1	***
Model 2	0.727	0.03	106.9	***
Model 3	0.646	0.03	90.8	***
Model 4	0.630	0.03	87.8	***

Based on 4,490 adults with diabetes, aged 21 years or older and were alive during the calendar year 2007 and 2009. Asterisks represent significant group differences by the presence of thyroid disorders using 𝑡-tests.
Model 1, included age, gender, and race/ethnicity as independent variables. In Model 2, in addition to the variables used in Model 1, other demographic (marital status and metro status), socioeconomic (education, poverty status, and health insurance) health status (perceived physical and mental health) and lifestyle factors (body mass index categories and smoking status) were included. We controlled for number of cooccurring chronic conditions as an additional variable in Model 3., Finally, in Model 4, we also additionally entered presence of any macrovascular comorbid conditions.
* 𝑃<.001; **.001≤𝑃<.01; *.01≤𝑃<.05.