Responsiveness of measurements of lower-limb muscle strength obtained with a hand-held dynamometer from patients with stroke

BACKGROUND: The responsiveness of measurements obtained by hand-held dynamometry (HHD) is largely unexplored. OBJECTIVE: A secondary analysis of data from a clinical cohort of patients (N = 55) admitted for inpatient rehabilitation following acute stroke was completed to determine the responsiveness of measures of lower extremity muscle strength. METHOD: The isometric strength of hip flexion, knee extension, and ankle dorsiflexion was measured bilaterally in 55 patients using HHD. Independence in bed-to-chair transfers, level ground gait, and stair negotiation was determined using an ordinal mobility scale. All measures were obtained at admission and discharge. RESULTS: Over the course of rehabilitation muscle strength increased significantly (p <= 0.002) in all 3 lower limb muscle actions of both the weaker and stronger sides. Effect sizes (0.19-0.43) and standardized response means (0.45-0.79) were mostly small to moderate but tended to be greater on the weaker side than the stronger side. The minimum detectable change (95%) values ranged from 33.2 to 87.5 N and were higher on the stronger than on the weaker side. Receiver operating characteristic curve analysis for identifying minimal clinically important differences revealed cut-points between -12.9 and 52.9 Newtons for strength increases differentiating patients who did and did not demonstrate improved mobility. CONCLUSION: HHD is capable of detecting changes in lower limb strength after stroke but the responsiveness, as indicated by effect size, standardized response mean, minimum detectable change and minimal clinically importance is limited.