Table 4.
Binary logistic regression model predicting BMI ≥ 19 kg/m2 at 6-month follow-up using continuous discharge BMI as a predictor variable (n = 91).
| Predictor | B | S.E. | Wald | df | Sig. | Exp(B) | 95% C.I. for Exp(B) | |
|---|---|---|---|---|---|---|---|---|
| Lower | Upper | |||||||
| Admission BMIa | 0.070 | 0.170 | 0.173 | 1 | 0.678 | 1.073 | 0.770 | 1.496 |
| Discharge BMIa | 0.470 | 0.187 | 6.332 | 1 | 0.012 | 1.600 | 1.110 | 2.307 |
| Illness duration | −0.003 | 0.023 | 0.017 | 1 | 0.897 | 0.997 | 0.954 | 1.042 |
| Lifetime nadir BMIa | 0.206 | 0.188 | 1.203 | 1 | 0.273 | 1.229 | 0.850 | 1.775 |
The model overall is significant (chi-square = 22.30; p < 0.001).
a
BMI = body mass index, kg/m2.