Measuring gait – how the choice of measure can affect the statistical results and the clinical interpretation

Abstract

Aims: The aim of this study was to illustrate how the choice of gait measure could affect the statistical analysis of data and the resulting clinical conclusions.

Methodology: A descriptive design in which the results from different tests from 10 patients with normal pressure hydrocephalus illustrates the potential to generate different clinical conclusions.

Major findings and principal conclusion: The results illustrate how the choice of measure can affect the statistical results and the clinical interpretation of a study. It is possible to have the paradoxical situation in which one group has a better walking ability if the variable speed is used but the other group has a better walking ability if the variable time is used. An important message is that the choice of measurement and the transformation of data are not primarily statistical issues. If the statistical results are to be useful for clinical decisions, the variables used must be directly related to the utility for the subjects. An understanding of the clinical relevance of different outcomes is required. The distinction between when numbers are purely descriptive and when numbers represent a valuation is subtle and difficult to comprehend.

Keywords:

• Walking
• measurement
• statistics
• transformations
• utility

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.