Table 2.
Results of the multivariate logistic regression analyses
| Covariates | Standard analysis | Interaction analysis | ||
|---|---|---|---|---|
| OR (95% CI) | P-value | OR (95% CI) | P-value | |
| Age > 65 years | 2.76 (2.12–3.60) | < 0.001 | – | * |
| Non-HIV immunosuppressive disease | – | NS | – | NS |
| Obesity | – | NS | – | NS |
| Infectious disease | 5.26 (1.66–16.71) | 0.005 | 4.31 (1.39–13.39) | 0.012 |
| Kidney disease | 2.60 (1.24–5.42) | 0.011 | 2.59 (1.27–5.27) | 0.009 |
| Heart disease | 1.67 (1.08–2.57) | 0.021 | 2.00 (1.31–3.04) | 0.001 |
| Hypertension | – | NS | – | * |
| Diabetes | – | NS | – | * |
| Age > 65 years × hypertension × diabetes | – | * | 1.50 (1.02–2.19) | 0.039 |
| Model summary | ||||
| BIC | 1,417.28 | – | 1,467.84 | – |
| BS | 0.191 | – | 0.201 | – |
| AUC | 0.654 | – | 0.588 | – |
AUC = area under the curve: measure of discriminatory performance where higher values indicate better performance; BIC = Bayesian information criterion: metric used for the comparison of regression models; BS = Brier score: metric used for overall model performance in which scores range from 0 for a perfect model to 0.25 for a non-informative model; NS = nonsignificant; OR = odds ratio. Data are given as OR and 95% CI.
*
Not included in the model.