Minimal clinically important difference (MCID) is a vital tool in the analysis of clinical results. It allows the determination of clinical relevance of statistical data. Our hypothesis was that specific differences between preoperative and postoperative scores would be able to accurately predict patient perception of improvement and satisfaction as reflected by anchor and distribution-based questions.

Retrospective cohort with patients that underwent rotator cuff repair. We evaluated the University of California at Los Angeles Shoulder Rating Scale (UCLA) and the American Shoulder and Elbow Surgeons Assessment Form (ASES) before and 12-months after surgery. Anchor-based, distribution-based and minimum detectable change (MDC) approaches were utilized.

We evaluated 289 shoulders. The MCID for the UCLA scale was 4.5 points using the anchor method, 2.5 by the distribution method and 3.6 by MDC. Patients with a baseline score>20 presented a lower MCID (1.5, 1.1 and 1.7, respectively). For the ASES score, the MCID was 6.1 by the anchor method, 10.5 based on the distribution method and 26.3 by MDC. In the group of patients above the 60 point cutoff, the obtained values were 2.4, 4.9 and 13.6, respectively.

The mean MCID value for the UCLA shoulder score is 3.5 points, ranging from 2.5 points (distribution method) to 4.5 points (anchor method). The mean MCID value for the ASES score was 15.2 points, ranging from 6.1 (anchor method) to 26.3 (MDC). Patients groups presenting with higher preoperative scores showed lower MCID values. This fact needs to be considered in postoperative comparisons between treatment groups.

Basic Science Study, Validation of Outcomes Instruments/Classification Systems.