Table 5.
Models predicting discharge location post-TBI hospitalization.
| Model predictor(s) | Home with support vs home OR (95% CI) |
LTC vs home OR (95% CI) |
CCC vs home OR (95% CI) |
Other vs home OR (95% CI) |
Rehab vs home OR (95% CI) |
Likelihood ratio (df, LRT p-value) |
|---|---|---|---|---|---|---|
| Model 1 | ||||||
| Sex (female = 1) | 1.18 (0.98, 1.43) | 2.14 (1.09, 4.22) | 1.10 (0.72, 1.68) | 0.89 (0.76, 1.03) | 1.11 (0.93, 1.34) | 13.10 (5, 0.0224) |
| Model 2 | ||||||
| Gender score | 1.22 (0.74, 2.02) | 3.68 (0.57, 23.84) | 1.72 (0.58, 5.14) | 0.34 (0.22, 0.51) | 0.54 (0.32, 0.88) | 38.92 (5, < 0.0001) |
| Model 3 | ||||||
| Sex (female = 1) | 1.17 (0.96, 1.43) | 1.98 (0.98, 4.04) | 1.04 (0.67, 1.63) | 1.00 (0.85, 1.18) | 1.22 (1.01, 1.48) | 8.59 (5, 0.126) |
| Gender score | 1.07 (0.63, 1.81) | 2.14 (0.30, 15.46) | 1.63 (0.52, 5.14) | 0.34 (0.22, 0.52) | 0.45 (0.27, 0.77) | 34.41 (5, < 0.0001) |
OR: odds ratio, CI: confidence interval; LRT: Likelihood-Ratio Test; df: degrees of freedom; LTC: Long Term Care; CCC: Inpatient Complex Continuing Care. Baseline category logit model using binary sex and continuous gender score (1 is more woman-like vs. 0 is more man-like) for test set with N = 7343.