Reproducibility of walking test results in chronic obstructive airways disease.

Thirty six patients with chronic airflow obstruction were studied to examine (1) the reproducibility and order effect of repeated walking tests when performed over consecutive days or consecutive weeks; (2) the correlation between walking distance and spirometric measurements; and (3) the effect of static visual clues on performance. In study 1, where 12 patients performed 12 walks over three consecutive days, five minute walking distance increased by 33% between walks 1 and 12, half of the increase occurring after the first three walks. In study 2, where 24 patients performed 12 walks over four consecutive weeks, five minute walking distance increased by 8.5% between walks 1 and 12. A learning effect was seen over the first nine walks. Static visual clues to performance did not affect the distance walked. Spirometric measurements showed no order effect in either study. Although walking distance correlated significantly with FEV1, forced vital capacity, and peak expiratory flow, these measurements were poor predictors of exercise performance. The learning effects seen on repeated performance of walking tests over short intervals should be considered when an individual's response to treatment is being interpreted. When walking tests are used in clinical trials a placebo group or randomised crossover design is essential.