Unpredictable error in dynamometry measurements: a quantitative analysis of the literature

This research investigated factors influencing unpredictable variability in dynamometric measurements. The criterion for inclusion in this review was that the standard error of measurement (SEM) associated with test procedures could be estimated. SEMs were determined using reported retest correlations and standard deviations of measurements. Subject factors found to influence variability were impairment and gender. Test factors implicated were movement direction and test speed. A strong relationship was indicated between score magnitude and variability. SEMs derived from measurements of low scoring subjects do not appear appropriate for interpreting measurements of higher scoring subjects. Measurements of subjects who achieve higher dynamometry scores appear more variable than measurements of lower scoring subjects. Similarly, as test conditions facilitate higher scores eg lower test speeds, variability in measurements increases. SEMs employed by clinicians to interpret test results should be derived from a population with a mean measurement that is representative of the subject being tested.