Table 4. Candidate models incorporating various.
Variables | Number of variables included | |||||||
---|---|---|---|---|---|---|---|---|
2 | 3 | 3 | 4 | 4 | 5 | 5 | 6 | |
Mode l# | #1 | #2 | #3 | #4 | #5 | #6 | #7 | #8 |
IADL | √ | √ | √ | √ | √ | √ | √ | √ |
Falls | √ | √ | √ | √ | √ | √ | √ | √ |
Physical function | √ | √ | √ | √ | ||||
Social activities | √ | √ | √ | |||||
Loss of weight | √ | √ | √ | √ | √ | √ | ||
Hearing | √ | √ | √ | |||||
AIC | 170 | 170 | 171 | 170 | 170 | 171 | 171 | 172 |
C-statistic | 0.64 (0.53–0.75) | 0.67 (0.56–0.78) | 0.68 (0.57–0.78) | 0.68 (0.57–0.78) | 0.66 (0.55–0.77) | 0.69 (0.59–0.80) | 0.67 (0.56–0.78) | 0.70 (0.60–0.80) |
As a point of reference, in our cohort, the AIC and C-statistic for CTP was 168 and 0.66, for MELD was 172 and 0.67, for Frailty Index was 170 and 0.68. Higher AIC indicates relative lower model quality. IADL, instrumental activities of daily living; AIC, Akaike information criteria.