Ordinal Logistic Regression Modeling for Perceived Decrease in Longevity with Observation (PDLO) among Men with Localized Prostate Cancer
| Unadjusted Effects | Adjusted Effects OR (95% CI) | ||||
|---|---|---|---|---|---|
| PDLO ≤0* | PDLO = 1† | PDLO ≥2‡ | OR (95% CI) | ||
| Age | 25 | 62 | 83 | 1.08 (1.04–1.13) | 1.01 (0.92–1.10) |
| PSA level | 25 | 62 | 82 | 0.94 (0.88–1.00) | 0.93 (0.86–1.00) |
| CALE | 25 | 62 | 83 | 0.92 (0.88–0.96) | 0.93 (0.84–1.02) |
| Anxiety score | 25 | 62 | 83 | 0.88 (0.81–0.96) | 0.96 (0.85–1.08) |
| Depression score | 25 | 58 | 82 | 0.79 (0.67–0.93) | 0.85 (0.69–1.04) |
* A (PDLO) ≤0 (reference group) indicates that CALE and self-reported survival expectation without treatment are within the same range or CALE is less.
† A (PDLO) = 1 suggests that CALE exceeds self-reported survival expectation without treatment by one response category.
‡ A (PDLO) ≥2 suggests that CALE exceeds self-reported survival expectation without treatment by at least 2 response categories (ie., about 10 years). All covariates in the ordinal logistic regression model are defined as continuous variables.
PDLO, perceived decrease in longevity with observation (categorized); PSA, prostate-specific antigen; CALE, comorbidity adjusted life expectancy; OR, odds ratio.