Distance Traveled and Cross-State Commuting to Opioid Treatment Programs in the United States

Table 3

Multilevel logistic regression model predicting patient travel across state to OTP.

Predictor	95% Confidence interval of the unadjusted odds ratio			95% Confidence interval of the adjusted odds ratio
	Lower	Odds ratio	Upper	Lower	Odds ratio	Upper
Block one
U.S. Region
Southeast versus Northeast	0.314	2.516	20.140	0.126	1.051	8.737
Midwest versus Northeast	1.044	11.826	133.970	0.821	8.473	87.475
West versus Northeast	0.003	0.063	1.480	0.002	0.038	0.839
Urbanicity
250 K−1M versus Metro>1M	1.649	10.959	72.848	1.878	12.081	77.697
<250 K versus Metro >1 M	0.105	1.384	18.160	0.230	3.148	42.997
Block two
Patient ZIP code area	0.921	0.983	1.049	0.918	0.980	1.046
Program ZIP code area	1.024	2.007	3.933	1.317	2.928	6.510
Block three
Age
30–43 versus 18–29	0.943	1.115	1.320	0.950	1.125	1.332
43–81 versus 18–29	0.891	1.126	1.423	0.900	1.142	1.448
Female	0.872	1.020	1.192	0.837	0.987	1.164
Race/Ethnicity
Hispanic versus white	0.262	0.559	1.192	0.266	0.570	1.220
Black versus white	0.198	0.416	0.875	0.191	0.402	0.847
Other versus white	0.196	0.403	0.830	0.195	0.403	0.829
Employed	0.778	0.911	1.068	0.767	0.906	1.072
Block four
Pain a reason for treatment	0.882	1.032	1.209	0.880	1.032	1.211
First methadone treatment	0.722	0.849	0.997	0.734	0.868	1.026
Strong urge to use	0.667	0.861	1.112	0.676	0.895	1.185
Severe withdrawal	0.778	0.934	1.123	0.801	0.978	1.194
Prescription opioid use in past 30 days	0.654	0.988	1.493	0.602	0.935	1.450
Heroin use in past 30 days	0.854	1.064	1.327	0.858	1.084	1.369

Patient ZIP code area and program ZIP code area are continuous measures; all other covariates are coded 0, 1.