Table 3.
Logistic regression analyses predicting having LGBT programs among facilities with residential treatment.
Factor | Adjusted odds ratio | P-value | Standard error | 95% Confidence interval |
---|---|---|---|---|
Region (ref: South) | ||||
Northeast | 1.354 | .008 | 0.115 | 1.082-1.696 |
Midwest | 0.952 | .671 | 0.117 | 0.757-1.197 |
West | 1.625 | <.001* | 0.094 | 1.351-1.956 |
Ownership (ref: Private non-profit) | ||||
Private for-profit | 1.581 | <.001* | 0.092 | 1.321-1.893 |
Government | 0.670 | .027 | 0.181 | 0.470-0.956 |
Pay assistance (ref: No) | ||||
Yes | 1.146 | .106 | 0.084 | 0.971-1.352 |
Sliding fee scale (ref: No) | ||||
Yes | 0.821 | .014 | 0.080 | 0.701-0.961 |
Medicaid (ref: No) | ||||
Yes | 1.051 | .550 | 0.084 | 0.892-1.239 |
Community outreach (ref: No) | ||||
Yes | 1.983 | <.001* | 0.080 | 1.694-2.322 |
Telemedicine/telehealth services (ref: No) | ||||
Yes | 1.239 | .004 | 0.076 | 1.069-1.437 |
N = 3554.
Due to the large sample size, P < .001 was deemed statistically significant.