Documentation of patient outcomes following injury is critical to ensure that patients are receiving the best care. Several patient-reported outcome measures (PROMs) have been developed to assess knee-related function following injury; however, there is limited data investigating the measurement properties of these instruments using Rasch model analyses.
To evaluate the measurement properties of several PROMs through application of the Rasch measurement model.
Cross-sectional study.
Clinical setting.
A convenience sample of 160 adults (mean age = 28.08 [10.95] y; male = 38.10%) were recruited for this study. Data collected were combined with existing pilot data from an earlier study containing de-identified Knee Injury and Osteoarthritis Outcome Score (KOOS) data from 79 adults with knee injuries resulting in a total sample size of 239 adults.
Psychometric evaluation of the KOOS, International Knee Documentation Committee subjective knee form, and the Marx Activity Rating Scale using the Rasch partial credit model.
Infit and outfit statistics, item step difficulties, person ability parameters, category function, and item and test information functions.
Large ceiling effects were present in the KOOS and International Knee Documentation Committee subjective knee form. Of the 65 items examined in this study, 35 items showed poor measurement properties. Item step difficulty for the remaining 30 items ranged from -5.45 (least difficult) to 0.57 logits (most difficult). The 5-category response options did not function well. Measurement precision decreased significantly as ability score increased beyond 0.30 logits.
Despite their use in clinical practice, several items showed poor measurement properties. Future studies are needed to develop and evaluate novel items that are suitable for measuring knee-related function in high-functioning populations to ensure continuity of PROMs as individuals recover from injuries.