Table 3. Random Intercepts-Only Binary Logistic Mixed Effects Model Examining Time-to-Surgery.
| Fixed effects | Received surgery within 24 h, OR (95% CI) | P value |
|---|---|---|
| Intercept | 0.16 (0.04-0.44) | <.001 |
| Patient characteristics | ||
| Agea | 1.01 (1.00-1.02) | .003 |
| Sex | ||
| Women | 1 [Reference] | [Reference] |
| Male | 1.04 (0.83-1.29) | .75 |
| Raceb | ||
| Racial and ethnic minority | 1 [Reference] | [Reference] |
| White | 0.96 (0.72-1.29) | .79 |
| Fracture | ||
| Hip | 1 [Reference] | [Reference] |
| Femur | 1.09 (0.86-1.37) | .49 |
| ASA class | ||
| I | 1 [Reference] | [Reference] |
| II | 2.59 (1.18-5.67) | .02 |
| III | 3.11 (1.41-6.84) | .005 |
| IV | 5.80 (2.53-13.32) | <.001 |
| Employment | ||
| Not employed | 1 [Reference] | [Reference] |
| Employed | 0.74 (0.56-0.98) | .04 |
| Insurance coverage | ||
| Not covered | 1 [Reference] | [Reference] |
| Coverage | 0.69 (0.45-1.05) | .08 |
| BMIa | 1.02 (1.01-1.04) | .006 |
| Hospital characteristics | ||
| Population insurance coverage | 0.94 (0.89-0.98) | .005 |
| Racial composition | 0.97 (0.75-1.24) | .78 |
| Population insurance coverage × diversity | 1.03 (1.01-1.05) | .03 |
| Random effects | ||
| σ2 | 3.29 | NA |
| τ00 site | 0.31 | NA |
| ICC | 0.09 | NA |
| N site | 23 | NA |
| Total observations, No. | 2558 | NA |
| Marginal R2/Conditional R2 | 0.117/0.192 | NA |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); ICC, intraclass correlation coefficient; σ2, within-hospital variance; NA, not applicable; τ00, between-hospital variance.
a
Continuous variable centered on hospital site mean.
b
Race and ethnicity dichomotomized into White versus racial and ethnic minority categories from 12 initial categories; many of the 12 racial and ethnic categories had quite small patient numbers, which would make comparisons impossible.