Assessing the properties of an encounter-based "process of recovery" measure of patient benefit following rotator cuff repair.

Background: Rotator cuff repair (RCR) is a common orthopedic procedure, with patient benefit often assessed using the American Shoulder and Elbow Surgeons (ASES) score. However, because ASES scores are not consistently collected across patient populations or care settings, they may be impractical for broad policy evaluation. The objective of this study was to assess whether the number of shoulder-related encounters a patient has during a Recovery Period after standard follow-up care (here 181-365 days post-RCR) can serve as a proxy benefit measure for 12-month ASES scores following RCR. Methods: Adult patients (18 years or older) who underwent a primary RCR between 2017 and 2022 and had complete post-operative ASES surveys at 12 months were included in this study. Health-care encounters with shoulder-related diagnoses were counted during the Recovery Period. Linear regression was used to contrast outcome models as a function of baseline covariates for Recovery Period shoulder-related encounter and 12-month ASES score benefit measures and to estimate the association between these measures. Results: Of an initial sample of 683 patients who underwent primary RCR and provided a baseline ASES score, 513 (75%) had complete 12-month ASES scores. Outcome models showed consistent relationships between patient baseline covariates and the 12-month ASES scores and Recovery Period shoulder-related encounters. Furthermore, regression showed a strong association between the number of Recovery Period shoulder-related encounters and ASES scores at 12 months ( Conclusion: Additional shoulder-related encounters during the Recovery Period were associated with clinically important lower 12-month ASES scores. Encounter-based measures should not replace patient-reported outcome measures for clinically focused research and individual clinical decision-making but may provide useful proxy measures of patient benefit for population-level orthopedic outcomes research when patient-reported outcome measure data are incomplete or unavailable.