Table 3.
Characteristics | IRR of Return Visits (95% CI) | Adjusted IRR of Return Visits(95% CI)** |
---|---|---|
Predisposing | ||
Female | 0.93 (0.87–1.00)* | |
< Age 30 | 0.51 (0.46–0.55)* | 0.19 (0.15–0.24)* |
Homeless | 0.99 (0.84–1.17) | |
High school degree/GED | 0.99 (0.93–1.06) | |
Race | ||
White | Ref | |
Black/African-American | ||
Other | ||
Marital Status | ||
Single | Ref | Ref |
Widowed/Separated/Divorced | 1.13 (1.05–1.21)* | Ref |
Married | 0.87 (0.78–0.97)* | 0.47 (0.33–0.68)* |
Enabling | ||
Same Zip Code as Enrollment Site | 1.45 (1.36–1.55)* | 4.16 (2.64–6.56)* |
Employed | 0.74 (0.68–0.81)* | 0.46 (0.26–0.81)* |
Need | ||
Drug treatment (ever) | 0.96 (0.89–1.04) | |
>20 years of injection drug use | 1.42 (1.34–1.51)* | 3.38 (2.48–4.61)* |
>4 injections per day | 1.01 (0.94–1.09) |
significant at p < .05
GEE was used to account for possible clustering effects by site; the exchangeable correlation (rho) for this model was .03. The overdispersion parameter for this model was 2.68. The offset was one year for all participants.