Table 4.

Coefficients of noninteraction terms estimated from the final logistic regression model.

		Estimate	P value
	Intercept*	−3.180	<2e − 16
Discharge	Home	Reference
	Other	0.302	0.119
Race	African American	Reference
	Caucasian	0.015	0.760
	Missing	−0.335	0.012
	Other*	−0.267	0.009
Admission	Emergency	Reference
	Other*	−0.155	<0.001
	referral	−0.020	0.517
Medical specialty	Cardiology	Reference
	General practice	0.388	0.035
	Internal medicine	0.377	0.022
	Missing*	0.463	0.002
	Other	0.306	0.059
	Surgery	0.443	0.032
Time in hospital*		0.130	0.000
Age	[30, 60)	reference
	[60, 100)	0.286	0.041
	<30	1.833	0.031
Diagnosis	Diabetes	reference
	Circulatory	0.143	0.171
	Digestive	−0.066	0.604
	Genitourinary	−0.288	0.056
	Injury	0.022	0.878
	Musculoskeletal*	−0.627	0.000
	Neoplasms	0.146	0.375
	Other	0.065	0.558
	Respiratory	−0.299	0.013
HbA1c	Not measured	reference
	High, changed*	−0.398	0.004
	High, not changed*	−0.579	0.009
	Normal	0.003	0.982

“Diagnosis” stands for a primary diagnosis with possible values: “circulatory” for icd9: 390–459, 785, “digestive” for icd9: 520–579, 787, “genitourinary” for icd9: 580–629, 788, “diabetes” for icd9: 250.xx, “injury” for icd9: 800–999, “musculoskeletal” for icd9: 710–739, “neoplasms” for icd9: 140–239, “respiratory” for icd9: 460–519, 786, and “other” for otherwise.

“HbA1c” variable has four values: “not measured” when the test was not measured, “normal” if the test was measured and the result was normal, “high, changed” when the result of HbA1c test was high and diabetic mediations were changed, and “high, not changed” when the result of HbA1c test was high but diabetic mediations were not changed.

*Coefficients significant at the 0.01 significance level.