Validity of pedometers in people with physical disabilities: a systematic review.

Abstract

To review the literature for the criterion validity of pedometers for use in child and adult populations with physical disabilities. Academic Search Premier, ERIC, SPORTDiscus, MEDLINE, AMED, Scopus, CINAHL, Web of Science, and EMBASE databases, searched from inception to September 7, 2011. Studies were included if they were peer-reviewed articles, included populations with physical disabilities, and reported primary data for pedometer validity in comparison with direct observation. A consensus approach was used to apply the inclusion and exclusion criteria. Of the 163 articles identified in the database searches (excluding duplicates), 7 studies met the inclusion criteria. The quality of the studies was assessed independently by 2 reviewers, using a purpose-designed appraisal tool, with a consensus approach used to settle disagreement. A single reviewer extracted data relating to sample size, participant characteristics, pedometer model, main variables tested, duration of tests, and method of direct observation. The methodologic quality of the studies was generally high; however, there was a wide variation of population and methodology between studies. The correlation between pedometer step counts and directly observed step counts was moderate to excellent (intraclass correlation coefficient, .52-.87), and percent errors ranged from 0.5% to 24.7%. Secondary variables reported included the effect of speed of movement, pedometer placement, comparison of pedometer makes/models, and test-retest reliability. Available evidence suggests that pedometers are valid for use in clinical and research settings in people with physical disabilities. Further research examining the validity of pedometers in less heterogeneous populations of people with disabilities is warranted to determine validity for specific disability populations and to determine optimal pedometer placement.