Table 2.
Results of Cox Proportional Hazards Regression Modeling
| Covariate | AIC w/Covariates | AIC Changed | HR (95% CI) | P-value |
|---|---|---|---|---|
| Univariable Models | ||||
| Distress Thermometera | 97.38 | −9.56 | 1.28 (1.11, 1.47) | <.001 |
| Age at Diagnosisa | 106.70 | −0.24 | 1.04 (0.99, 1.09) | 0.135 |
| Risk Stratification at Diagnosisb | 96.24 | −10.70 | 6.92 (2.30, 20.79) | <.001 |
| Charlson Scorea | 107.34 | 0.40 | 1.18 (0.93, 1.48) | 0.206 |
| Ever Use of Therapyc | 104.48 | −2.46 | 3.99 (0.89, 17.87) | 0.035 |
| Ever Reported Fatiguec | 108.27 | 1.33 | 1.58 (0.51, 4.91) | 0.414 |
| Frequency of Fatiguea | 107.46 | 0.52 | 0.89 (0.73, 1.09) | 0.224 |
| Final Model | ||||
| Distress Thermometera | 93.53 | -13.41 | 1.18 (1.01, 1.36) | 0.030 |
| Risk Stratification at Diagnosisb | 4.02 (1.26, 12.83) | 0.016 | ||
AIC=Akaike information criteria. HR=hazard ratio. CI=confidence interval
Single point increase.
High vs. Intermediate/Low.
Yes vs. No.
AIC change is calculated from the AIC of the model without covariates, 106.94.