Measurement properties of PROMIS short forms for pain and function in orthopedic foot and ankle surgery patients

Purpose To examine the measurement properties of the German PROMIS short forms for pain intensity (PAIN), pain interference (PI) and physical function (PF) in orthopedic foot and ankle surgery patients. Methods Patient-rated outcomes were collected from consecutive patients of our foot and ankle registry before and 6 months after surgery. Measurement properties were tested according to the COnsensus-based Standards for the selection of health status Measurement Instruments (COSMIN). The German Foot Function Index (FFI-D) served as a legacy measure. Results 748 patients were included in our cross-sectional sample. Longitudinal and test-retest data were available for 202 and 65 patients, respectively. Construct validity of all short forms was good. All Cronbach's alpha and intraclass correlation coefficients were > 0.7. The smallest detectable change (SDC) was highest for PF (8.9) and lowest for PI (6.5). Minimal important change was 4 to 5 points and thus smaller than SDC for all instruments. We observed a baseline ceiling effect for PF. PI showed insufficiently correlated change scores with FFI-D disability change scores, and therefore failed the responsiveness testing. Conclusion Our study showed some adequate psychometric properties, but also certain aspects regarding interpretability and responsiveness that researchers must be aware of when using PROMIS short forms of pain and function in foot and ankle surgery patients.