LAUSR

Objective The purpose of this study was to evaluate the Patient Reported Outcomes Measurement Information System (PROMIS) computerized adaptive testing Pain Interference (PROMIS PI) item bank in patients undergoing shoulder surgery. We hypothesized that PROMIS PI would exhibit a strong positive correlation with the numerical pain scale for the operative shoulder (shoulder NPS) with less floor and ceiling effects. Secondary study aims included assessing the relationships between patient characteristics and PROMIS PI. Design Analytical cross-sectional study. Setting Urban academic medical center. Patients One-hundred and ninety-five patients undergoing shoulder surgery between June 2015 to June 2017. Main outcome measures All patients completed a series of patient-reported outcomes measures, including PROMIS PI and NPS. Non-parametric tests were used for bivariate analysis. Multivariable regression models were used to determine independent associations. Results There was a moderate correlation between the PROMIS PI and shoulder NPS scores (rs = 0.53; p < 0.001). PROMIS PI had no ceiling or floor effects while shoulder NPS had 26 patients (13.3%) at either the floor or the ceiling. PROMIS PI demonstrated a strong correlation with PROMIS Physical Function (rs = -0.65; p < 0.001), ASES total score (rs = -0.67; p < 0.001), and PROMIS Fatigue (rs = 0.64; p < 0.001). Conclusions The strong association noted between PROMIS PI and psychosocial and behavioral factors, versus that of NPS, demonstrates that PROMIS Pain Interference may be a more appropriate choice as an outcome measure where the surgeon is concerned about a patient's improvement of functioning and decrease in the impact of pain in other aspects of wellbeing following shoulder surgery. Level of evidence IV.