Developing a Reliable Testing Protocol for the Hydra-Fitness Upper Body OmniTron

Evaluation of the reliability of musculoskeletal testing equipment is an important step in establishing the usefulness of an assessment device's data. The purposes of this study were to determine the reliability of a specific upper body OmniTron testing protocol and to estimate reliabilities for several other protocols in order to determine the optimal one. After upper body warm-up, 32 subjects (22 men, 10 women; mean age = 20.7 +/- 2.6 years) were tested on the Hydra-Fitness OmniTron chest press (CP) and upper back pull (BP) exercise at a slow speed and high resistance. Peak force, work, and power values were obtained for each subject during a protocol consisting of two test sessions of four repetitions each, with an intervening 5-minute rest period. The first repetition of each test session was considered a warm-up, and data from these were discarded. A repeated measures ANOVA was conducted on the main effects (tests and repetitions), and a generalizability analysis was performed. Mean peak force output values were significantly higher (p < .0001) during the second test session (Test 1 CP = 654.9 +/- 167.5 N, Test 2 CP = 690.0 +/- 179.1 N; Test 1 BP = 547.5 +/- 139.0 N, Test 2 BP = 568.2 +/- 153.9 N), possibly indicating the presence of a learning effect. Work and power data showed similar trends, as peak force, work, and power were highly correlated with one another (r >/= .90). Reliability of the protocol was estimated for the peak force data of CP and BP. The coefficients were .993 and .985, respectively. Generalizability forecasting for alternate protocols demonstrated that for both CP and BP, a minimum of two tests of at least three repetitions each were necessary for optimum reliability, although all evaluated combinations exhibited high reliability (r >/= .949). This study suggests that the two test session/three repetition protocol (four repetitions, including the initial discarded one) with a 5-minute rest is highly reliable and may be desirable for use with subjects being tested for the first time in order to counteract a possible learning effect. The clinician may also choose another protocol shown to be reliable by this investigation. J Orthop Sports Phys Ther 1992;16(2):87-91.