Affiliations: [a] Department of Health Sciences, Division of Physical
Therapy, Lund University, Lund, Sweden | [b] Department of Clinical Sciences, Division of
Rehabilitation Medicine, Lund University, Lund, Sweden | [c] Department of Health Sciences, Luleå University
of Technology, Luleå, Sweden
Note: [] Address for correspondence: Anna Maria Holmbäck, PhD, Lund
University, University Hospital, Division of Physical Therapy, SE-220 05 Lund,
Sweden. Tel.: +46 46 222 89 56; Fax: +46 46 22 42 02; E-mail:
[email protected]
Abstract: The aim of this study was to determine the test-retest (two
occasions seven days part) reproducibility of isokinetic (Biodex) concentric
(CON) and eccentric (ECC) ankle dorsiflexor strength and fatigue measurements
in older subjects using several statistical methods. Thirty healthy men (n=15)
and women (n=15) aged between 70 and 85 years participated in the study.
To determine the reproducibility of strength measurements, the 30 individuals
performed three maximal CON and ECC contractions at 30°/s and
90°/s. The intraclass correlation coefficient (ICC_{2.1}) for CON
and ECC strength measurements was high and ranged from 0.94 to 0.98. Bland
& Altman graphs and analyses indicted no systematic bias. The standard
error of measurement (SEM), representing the smallest change that indicates a
real improvement (or deterioration) for a group of individuals, was small
(< 2.3 Nm). The smallest real difference (SRD), representing the smallest
detectable change that indicates a real improvement (or deterioration) for a
single subject, was also small ('error bands' from −7.2 Nm to 5.6 Nm). A high
correlation (Pearson's r > 0.94) between CON peak torque at 30°/s
and 90°/s, as well as ECC peak torque at 30°/s and
90°/s, suggested that any of these two velocities could be used as a
reference. To determine the reproducibility of fatigue measurements, 28 of the
30 individuals performed 50 maximal CON contractions at 60°/s.
Reproducibility of the loss in work (work fatigue) and the relative loss in
peak torque, comparing the first three to the last three contractions (3-3),
was determined and the ICC was 0.71 and 0.60, respectively. In addition, the
SEM was small (< 8.4%) and the 'error bands' to define the SRD for a
single subject were also small (from −23.5% to 24.5%). In conclusion,
these values indicate small measurement errors and thus provide a clinically
acceptable basis for testing ankle dorsiflexion strength and fatigue of the
dominant side in healthy older people.
Keywords: Aging, ankle, muscle strength, muscle fatigue, reproducibility of results
