The change in a validated shoulder measure score will be used as a performance measure for surgeons performing shoulder arthroplasty which includes hemiarthroplasty, total anatomical and reverse arthroplasty procedures. Two measures will be created and reported by each surgeon. Surgeons will report the average shoulder measure change score for patients treated during the observation period. In addition, surgeons will produce a risk-adjusted shoulder measure change score ratio by dividing the average patient shoulder measure change score by the average predicted patient shoulder measure change score calculated using the formula provided. These measures will serve as sports medicine performance measures at the eligible surgeon level. Eligible validated shoulder patient reported outcome measures include: American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES), Oxford Shoulder Score (OSS), Single Assessment Numeric Evaluation (SANE) [1] Defining the population: CPT-4 codes will be used to identify patients who received shoulder arthroplasty surgery. Shoulder arthroplasty during the reporting period CPT-4 Codes: 23470, 23472 (excluding 23 473 and 23474). Surgeons who performed a minimum of 25 shoulder arthroplasty procedures a year will be eligible to submit this measure. Time period for data collection and reporting: This measure will be calculated using all patients who underwent shoulder arthroplasty during the observation period. Patients receiving shoulder arthroplasty during the period 18 to 6 months prior to the reporting data will be included in the analysis to ensure 6 months of patient-reported follow-up in the shoulder measure score are available. Measures: Two surgeon-level performance measures will be reported. The unadjusted measure will be the average 6-month shoulder measure change score across all shoulder arthroplasty patients treated by the surgeon. The adjusted measure will be the ratio of the average shoulder measure change score divided by the average predicted shoulder measure change score for all patients treated by the surgeon. Unadjusted measure: Average Shoulder Measure Change Score NUMERATOR: Sum of shoulder measure change scores from pre-surgery to 6-months post-surgery across all shoulder arthroplasty patients in the denominator. DENOMINATOR: The number of all patients who received shoulder arthroplasty during the observation window from the surgeon. Type of score: Validated shoulder measure change score average Risk adjusted measure: Shoulder Measure Change Score Ratio NUMERATOR: Average 6-month shoulder measure change score for all shoulder arthroplasty patients treated by surgeon. DENOMINATOR: Average predicted 6-month shoulder measure change score for all shoulder arthroplasty patients treated by surgeon. Risk adjustment methods and variables: The predicted 6-month shoulder measure change scores are estimated for each patient using the Center for Effectiveness Research in Orthopaedics’ (CERortho) risk adjustment regression model that includes the following patient variables: age, gender, BMI, smoking status, comorbidities, worker’s compensation status, prior shoulder surgery, fracture, glenoid type, arthroplasty type (reverse vs. traditional total, hemi), the baseline shoulder measure, and baseline measures of pain, shoulder function, and quality of life.[2-6] Variables included in the current CERortho risk model are based on information in available databases. CERortho plans to constantly update these specified variables based on input from evaluated surgeons and specialty organizations. Appendix A contains the list of baseline concepts from peer-reviewed literature that are thought to affect outcomes of treatment and will be candidates for future inclusion in the model. Type of score: Ratio score Interpretation of score: Surgeons with ratio scores equal to 1 have quality scores equal to the average surgeon given the characteristics of their patients. Surgeons with ratio scores greater than 1 had shoulder change scores higher than average. Surgeons with ratios scores less than 1 had shoulder change scores lower than average given the characteristics of their patients. Definitions: Patient’s Shoulder Measure Score: A shoulder measure score is produced when the patient answers a validated shoulder assessment administered in the clinic or through an online portal. Patient’s Shoulder Measure Change Score: A patient’s change score is calculated by subtracting the patient’s shoulder measure score at baseline from the patient’s shoulder measure score at 6 months. Patient’s Shoulder Measure Predicted Adjusted Change Score: Shoulder measure change scores for patients are risk adjusted using a regression model that includes the following independent variables: age, gender, BMI, smoking status, comorbidities, worker’s compensation status, prior shoulder surgery, fracture, glenoid type, arthroplasty type (reverse vs. traditional total, hemi), the baseline shoulder measure, and baseline measures of pain, shoulder function, and quality of life.[2-6] The patient’s predicted shoulder measure change score is the dependent variable. The regression model produces a risk-adjusted predicted shoulder measure change score for each patient. Risk-adjusted Shoulder Measure Change Score Ratio: The ratio between the actual change scores and the predicted change scores (after risk adjustment) is the ratio score. The risk-adjusted shoulder measure change score represents risk-adjusted change corrected for patient characteristics. A risk-adjusted ratio score of one or greater should be interpreted as change scores that were as good as or better than predicted given the risk-adjustment variables of the patient. Risk-adjusted ratio change scores less than one should be interpreted as shoulder measure change scores that were less than predicted given the risk-adjustment variables of the patient. The surgeon ratio scores can be used to make comparisons across surgeons. Steps: 1. Prior to shoulder arthroplasty surgery the patient completes a baseline validated shoulder measure assessment which generates the patient’s shoulder measure score at baseline. 2. Six months after surgery, the patient completes the same validated shoulder measure assessment again, which generates the patient’s shoulder measure score at 6 months. 3. The patient’s raw, unadjusted shoulder measure change score is generated by subtracting the baseline score from the 6-month score. Unadjusted measure: 4. Patients’ shoulder measure change scores are averaged by surgeon. Adjusted measure: 5. A risk-adjusted shoulder measure change score is predicted using the CERortho risk-adjustment regression model. 6. Patients’ unadjusted and adjusted shoulder measure change scores are averaged by surgeon. 7. A risk-adjusted Shoulder Measure Change Score Ratio is generated for each surgeon. 8. The ratio scores for all surgeons in the database are ranked. See attachment for references.