Early Development and Reliability of the Timed Functional Arm and Shoulder Test

Study Design Repeated-measures clinical measurement reliability study. Background While there are some shoulder functional tests for athletes, no widely used performance test of arm and shoulder function currently exists to assess lower-level upper extremity functional demands in, for example, a nonathlete population or elderly individuals. In these individuals, functional measures rely on patient self-report. Objectives Describe the development of the Timed Functional Arm and Shoulder Test (TFAST), age-related scores, and between-session reliability in a group of asymptomatic high school athletes, young adults, middle-aged adults, older adults, and a preliminary group of symptomatic patients. Methods One hundred forty asymptomatic individuals participated in the study: 36 high school athletes (14-18 years of age), 34 young adults (19-35 years of age), 37 middle-aged adults (36-65 years of age), 33 older adults (over 65 years of age), and 16 symptomatic patients (22-66 years of age). The TFAST is a functional test that includes 3 tasks: hand to head and back, wall wash, and gallon lift. Total repetitions were noted for each task, and the total TFAST score was calculated. Results Mean total TFAST scores were higher for young adults (107.9; 95% confidence interval [CI]: 102.5, 113.4) and middle-aged adults (105.2; 95% CI: 99.1, 111.3) as compared to the high school athletes (89.9; 95% CI: 81.2, 98.5) and older adults (74.5; 95% CI: 65.6, 83.5). All groups were significantly different (P<.05) from each other, except the young and middle-aged adults. For patients, the mean score for the symptomatic side was 100.1 (95% CI: 89.6, 110.5). The between-session reliability values for the total TFAST scores in the asymptomatic individuals were as follows: intraclass correlation coefficient (ICC) = 0.93; 95% CI: 0.60, 0.98; standard error of measurement, 6.7; and minimal detectable change based on a 95% CI, 18.5 repetitions. The ICC values for individual tasks ranged from 0.80 to 0.94 (95% CI range, 0.44-0.98). The reliability for the patient group was 0.83 (95% CI: 0.51, 0.94). Conclusion The TFAST was sensitive to detect differences in functional performance between age groups, demonstrated adequate between-session reliability, and demonstrated feasibility in a symptomatic patient group. Further assessment is needed to refine the TFAST. Development of a feasible and valid test of arm function would enhance clinical evaluation and outcome measurement. J Orthop Sports Phys Ther 2017;47(6):420-431. Epub 3 Mar 2017. doi:10.2519/jospt.2017.7136.