Clinically important differences in the intensity of chronic refractory breathlessness.

Abstract

Clinically important differences in chronic refractory breathlessness are ill defined but important in clinical practice and trial design. To estimate the clinical relevance of differences in breathlessness intensity using distribution and patient anchor methods. This was a retrospective data analysis from 213 datasets from four clinical trials for refractory breathlessness. Linear regression was used to explore the relationship between study effect size and change in breathlessness score (0-100mm visual analogue scale) and to estimate the change in score equivalent to small, moderate, and large effect sizes. Pooled individual blinded patient preference data from three randomized controlled trials were analyzed. The difference between the mean change in Day 4 minus baseline scores between preferred and non-preferred arms was calculated. There was a strong relationship between change in score and effect size (P = 0.001; R(2) = 0.98). Values for small, moderate, and large effects were -5.5, -11.3, and -18.2mm. The participant preference change in score was -9mm (95% CI, -15.8, -2.1) (P = 0.008). This larger dataset supports a clinically important difference of 10mm. Studies should be powered to detect this difference.