Measurement Properties of Handheld Dynamometry for Assessment of Shoulder Muscle Strength: A Systematic Review

Like any assessment tool, handheld dynamometry (HHD) must be valid and reliable in order to be meaningful in clinical practice and research. To summarize the evidence of measurement properties of HHD for the assessment of shoulder muscle strength. Cochrane Central Register of Controlled Trials (CENTRAL), PubMed, EMBASE, and PEDro were searched up to February 2020. Inclusion criteria were studies (a) evaluating HHD used on the glenohumeral joint, (b) evaluating measurement properties, and (c) included individuals ≥ 18 years old with or without shoulder symptoms. Exclusion criteria were studies (a) including patients with neurologic, neuromuscular, systemic diseases, or critical illness or bed-side patients and (b) that did not report the results separately for each movement. In total, 28 studies with 963 participants were included. The reliability results showed that 98% of the intraclass correlation coefficient (ICC) values were ≥0.70. The measurement error showed that the minimal detectable change in percent varied from 0% to 51.0%. The quality of evidence was high or moderate for the majority of movements and type of reliability examined. Based on the evidence of low or very low quality of evidence, the convergent validity and discriminative validity of HHD were either sufficient, indeterminate, or insufficient. The reliability of HHD was overall sufficient, and HHD can be used to distinguish between individuals on the group level. The measurement error was not sufficient, and evaluation of treatment effect on the individual level should be interpreted with caution.