Reliability and Agreement of Hand-Held Dynamometry Using Three Standard Rater Test Positions.

Only 10 raters took part, and the study analysed 9 tests, fewer than the 11 the authors calculated as needed, so it may have been underpowered. The forces were produced by a mannequin and pulley system, not by patients, so the study controlled patient variability and cannot show how measurements behave with actual patients. Only hip abduction, hip flexion, and knee extension were tested, and the mechanical setup could not add another test position. Raters received only about 30 minutes of training, much less than the 3 days and 20 hours reported by another study, which may limit how well they could stabilize. Two raters could not complete the highest knee extension force because they could not maintain a stable base, so the findings may not apply to forces beyond a rater's strength. The small sample prevented testing how rater, test position, and force magnitude interact. Outliers were identified but not removed from the analysis.

The authors reported no conflicts of interest. The work was supported in part by a Manchester University internal research grant, and the funding source had no involvement in study design, data collection, analysis, interpretation, manuscript writing, or the decision to submit for publication.

Do not conclude that hand-held dynamometry is accurate for patient strength testing because reliability was high. The study used a mechanical mannequin, not patients, and therapist-stabilized readings were consistently higher than fixed-support readings, with larger average error at low forces.