Accuracy and reliability of the Mayo Elbow Performance Score

Abstract

Purpose: To test the reliability of the Mayo Elbow Performance Score (MEPS) and compare it with a validated outcomes instrument, the American Shoulder and Elbow Surgeons (ASES) score.

Methods: A total of 42 patients with the chief problem of elbow dysfunction formed the study cohort. Patients with an immediate surgical indication or treatment at the index visit were excluded. The others completed an MEPS questionnaire; at a second visit 2 to 3 weeks later, they completed another MEPS questionnaire and were evaluated with the ASES elbow assessment. Reliability and accuracy were calculated using 2-tailed Pearson correlation coefficients with 95% confidence intervals. Pearson coefficients greater than 0.8 indicated strong agreement.

Results: The average MEPS score at the initial visit was 58. At the second visit, the average MEPS score was 69 and the average ASES score was 78. The Pearson coefficient for MEPS scores at the 2 time points averaged 0.82, and between the MEPS and ASES scores averaged 0.83. Both coefficients indicated strong agreement.

Conclusions: The MEPS has strong reliability when assessed at different times and when compared with a validated elbow outcomes instrument. Differences in compared scores of approximately 10 points indicate some patient improvement between time points; however, 95% confidence intervals, standard deviations, and ranges were essentially equivalent between and among tests, indicating similar accuracy.

Clinical relevance: The MEPS is a reliable outcomes instrument for clinical studies of elbow function that is used to assess nonsurgical treatment.

Keywords: MEPS; Outcomes instrument; elbow pathology; reliability and accuracy; validation.