Step width variability as a discriminator of age-related gait changes.

Only healthy community-dwelling adults without overt neurological disorders or significant disabilities were included, so the results may not apply to patients with neurological conditions, assistive devices, or walking limitations. The threshold analysis used group averages from studies rather than individual participant data, so it does not establish a validated cutoff for a single patient. The authors state the study cannot claim that the optimal reference range discriminates older adults who fall from older adults who do not fall. Visual inspection of the funnel plot indicated publication bias, although Begg's test p-value was 0.07 and Egger's test p-value was 0.08. The study of Decker et al. was omitted as a potential outlier, leaving 10 studies for the meta-analysis. Of the seven included studies that used treadmill walking, only three had a 10 minute walking protocol, which may affect measurement accuracy. Of the six overground walking studies, five used repeated single walking protocols and only one used a continuous walking protocol, which may affect reliability. The included studies did not report power analyses. Some included studies did not describe how preferred walking speed was determined, and step width was calculated using different foot markers across studies.

Funding was provided by the National Institute of General Medical Sciences, the National Center for Medical Rehabilitation Research, and the University of Nebraska at Omaha. The supplied text does not state author conflicts of interest.

Do not use these thresholds to label an individual patient as high fall risk or to set a training target. The thresholds were derived from group averages of studies in healthy community-dwelling adults, and the authors state the study cannot claim they discriminate older adults who fall from those who do not fall.