BACKGROUND AND PURPOSE

Knowledge of the measurement error and reliability of measurement tools is required to judge whether true changes in performance have occurred. How a patient's performance relates to that of a healthy individual, and which factors would influence performance, also need to be considered to assess whether a patient's performance is 'normal'. The aim of the present study was to assess within-session and test-retest reliability, and measurement error, of a hierarchical series of functional tests of balance and walking in healthy participants. Obtaining indicative data in a group of healthy participants was an additional aim.

METHOD

Forty healthy participants aged 20-60 years were recruited from staff and students of Cardiff University. The participants completed eight functional balance and mobility tests on two occasions on the same day. Intra-class correlation co-efficients (ICCs), assessed within-session, and test-retest reliability and measurement error were calculated from the mean squares error term of a repeated measures analysis of variance (ANOVA). The relationship to the overall mean score was calculated and linear regression investigated the factors influencing performance.

RESULTS

Within-session and test-retest reliability for each of the tests was moderate to high (ICCs = 0.88-0.98 and 0.77-0.94, respectively). The overall measurement error was 3% to 11% of the mean scores. Age was the most frequent factor influencing performance; level of activity and body mass index (BMI) did not influence performance on any of the tests.

CONCLUSIONS

Functional balance and mobility tests are reliable but are subject to random error up to 11% of the mean. The present study provides initial reference data for physiotherapists in clinical practice. It is important to begin to develop a database of a standard range of scores to give a context with which to judge more accurately the importance and relevance of clinical measurements from patients.