OBJECTIVES: To develop and validate a patient-reported outcome measure for elbow surgery. METHODS: A questionnaire was created following semi-structured interviews with 18 patients being treated for elbow problems to determine candidate questions. These 18 questions were then prospectively tested, pre- and again at 6 months post-surgery, on 104 patients undergoing a variety of surgical treatments for elbow problems. Assessment of test/re-test reliability involved repeated OES completion, at home, 2 days later. Candidate questions were excluded if they showed high ceiling or floor effects, or if they cross-loaded to more than one domain. Underlying factor structure, dimensionality, internal and test retest reliability, construct validity and responsiveness of the questionnaire items were assessed in relation to the: 1) Mayo Elbow Performance Score (MEPS) clinical scale; 2) Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire; and 3) SF-36 general health survey. RESULTS: A total of 75% of patients were followed-up. Six candidate questions were excluded due to ceiling effect, cross loading or disordered response in Rasch analysis. A final 12-item questionnaire resulted, consisting of 3 unidimensional domains/scales: ‘Elbow Function’, ‘Pain’, ‘Social-Psychological’. Internal reliability (Cronbach’s alpha) for the 3 OES domains were optimal: Pain 0.89, Elbow function 0.90, Social-psychological 0.84. Test-retest reliability, (intra-class correlation coefficient), was good for all three domains (0.98; 0.90; 0,87). The OES showed high convergent correlation with the DASH and MEPS, with the exception of the social-psychological with the MEPS, which showed moderate correlation; correlation with related domains of the SF-36 was high. Divergent validity was confirmed by moderately low correlation with the general and mental health domains of the SF-36. Responsiveness was good, with effect sizes of >1.0 for the OES pain and social-psychological domains and for the MEPS; 0.80 for OES function and -0.74 for the DASH. CONCLUSIONS: The 12-question OES has good measurement properties in the context of elbow surgery.