Intra-rater reliability of hip abductor isometric strength assessment in older fallers and non-fallers

Abstract

Impaired muscle function at the hip has been implicated to be a major factor related to falls in older patients. Falls mostly occur in an upright position, therefore, the measurement of hip abductor strength in the often used side-lying or supine position is not functional. The aim was to investigate, for older persons at risk of falls, the feasibility and intra-rater reliability of a hip abductor test in standing position. To be included older persons had to be over 65 years. Fifteen fallers and 15 non-fallers participated in this study. Hip abductor strength was quantified with a hand-held dynamometer in standing position. Three trials were realized for both sessions. The raw force signals were exported and a low-pass filter was applied to attenuate high-frequency noise. The voluntary isometric strength (MVIS) was defined as the peak value reached within 0 to 4 s. The rate of force generation (RFG) was evaluated over 50 ms after 10% of the MVIS was reached. All variables were normalized for participants’ weight. Reliability was determined using intra-class correlation coefficient (ICCagreement), standard error measurement (SEM) and Bland and Altman plot analysis (mean differences between sessions and limit of agreement-LOA95%). The ICC above 0.75 indicated good reliability, those between 0.5 and 0.75 moderate reliability and those under 0.5, poor reliability. All recruited persons completed the strength test in standing position. The intra-rater reliability for MVIS parameter was good in non-fallers group (ICC = 0.94, SEM = 7.86 N) and fallers group (ICC = 0.91, SEM = 6.36 N). The RFG parameter was characterized with a moderate reliability in non-fallers group (ICC = 0.63, SEM = 114.24 N/s) and good reliability in fallers group (ICC = 0.83, SEM = 71.12 N/s). Bland and Altman analysis revealed small mean differences between sessions for MVIS (fallers: 1.41 N [LOA95%: −15.56;18.38] vs. non-fallers: 1.97 N [LOA95%: −21.26;25.20]) and larger mean differences for RFG (fallers: 44.28 N/s [LOA95%: −137.56;226.12] vs. non-fallers: −72.1 N/s [LOA95%: −428.60;284.10]). This standing approach is feasible to measure the hip abductors strength in older persons at risk of falls. Hip abductor strength assessment using a hand-held dynamometer provided consistent results when measured several times. For clinicians, this reliable test is easy to perform in daily use and interesting for testing the hip abductors strengths in a functional position.