provides homogenous categories of patients whose characteristics can be reported.The preponderance of literature that describes outcomes of surgical management for shoulder disorders relates to rotator cuff repair, 12,18 subacromial decomt he incidence of shoulder pain has been estimated at 11.2 per 1000 persons per year and ranks as the third most frequent musculoskeletal complaint of patients visiting a primary care provider. 35 The prevalence of shoulder pain in the general population ranges between 6.9% and 34%.
15The etiology and pathology of shoulder pain can be unclear, which often results in a lack of consensus regarding the appropriate classification of shoulder disorders. 15,35 However, once a patient undergoes a surgical procedure for shoulder pain, the categorization of the patient's shoulder problem is typically described on the basis of the surgical procedure itself. Grouping patients postoperatively on the basis of surgical intervention t studY dEsiGn: Retrospective cohort study. t MEthods: This study included 856 patients (43.7% female; mean SD age, 51.8 14.2 years) who received outpatient physical therapy following shoulder surgery. Standardized methods for classification of patients to type of shoulder surgery and collection of outcome variables were used. Data were gathered from 57 therapists working in 12 clinics. Patients included had been classified into 1 of 4 surgical categories: repair of a unidirectional instability, rotator cuff repair, rotator cuff repair with a subacromial decompression, or subacromial decompression alone. Descriptive statistics were calculated for baseline characteristics of patients in each surgical category. For all patients, scores on the Disability of the Arm Shoulder and Hand (DASH) questionnaire and a numeric pain rating scale (NPRS) were obtained at the initial and final physical therapy visits, and the change between visits was calculated. Data on number of physical therapy sessions and length of stay (LOS) were collected. For each surgical category, independent-samples t tests were used to determine differences between genders for each initial and final clinical outcome of pain and disability, change scores, utilization of visits, and LOS. The percentage of patients who achieved a minimal clinically important difference (MCID) on the DASH was also determined for each surgical group. For each gender in each surgical category, paired t tests were used to determine if patients achieved significant change in pain and disability.t rEsults: Means for each clinical outcome for the initial and final pain and disability scores, change scores, and the percentage of patients that achieved an MCID are provided. Significant differences were observed between genders for clinical outcomes. In the group treated with unilateral instability repair, women reported significantly greater initial disability than men, and their DASH change scores were significantly greater. In the group that had rotator cuff repairs, women reported significantly greater disability initially a...