Table 1.
List of non-imaging variables fed into logistic regression models to make predictions of whether a patient would undergo TKR within 5 years. Abbreviations used: Body Mass Index (BMI), Nonsteroidal Anti-inflammatory drugs (NSAIDS), Blood Pressure (BP), Physical Activity Scale for the Elderly (PASE), Knee Injury and Osteoarthritis Outcome Score (KOOS), Quality of Life (QOL), Western Ontario and McMaster Universities Arthritis Index (WOMAC), Short Form 12 (SF-12).
| Non-imaging variables used to augment image-based predictions | |
|---|---|
| Age | Comorbidity score |
| BMI | Injections to treat arthritis in previous 6 months |
| Education | Seen physician for arthritis in previous year |
| Ethnicity | Knee valgus negative alignment (degrees) |
| Income | Isometric leg strength |
| NSAID usage | Back pain in previous 30 days |
| Analgesics usage | Difficulty squatting in previous 7 days |
| Systolic BP | Difficulty kneeling in previous 7 days |
| Considering TKR | Baseline frequent knee pain status |
| PASE | Previous knee injury that limited walking |
| KOOS QOL | 0-10 global rating assessing effect of knee pain |
| KOOS pain | SF-12 physical component score |
| WOMAC pain | SF-12 mental component score |
| WOMAC disability | |