The minimally important difference of the Asthma Control Test

Abstract

The Asthma Control Test (ACT) has been well validated, but a minimally important difference (MID) has not been established. We sought to identify an MID for the ACT. Data come from 4 independent samples of adult asthmatic patients. Distributional methods for determining the MID included 0.5 SD, 1 SEM, and 2 SEM. Anchor-based methods assessed the relationship of differences in ACT scores to (1) self-reported asthma severity, (2) asthma episode frequency in the past 4 weeks, (3) physician ratings of asthma control, (4) physician recommendation of a change in therapy, (5) FEV(1), (6) the risk over the next 12 months of excess short-acting beta-agonist use and exacerbations, and (7) patient-defined changes in asthma course over 3 months. Four thousand one hundred eighteen patients completed the ACT. The 0.5 SD criterion for MID ranged from 2.03 to 2.45 points (mean, 2.2 points). The 1 SEM criterion ranged from 1.77 to 2.05 points (mean, 1.88 points), and the 2 SEM criterion ranged from 3.55 to 4.10 points (mean, 3.75 points). Differences in mean ACT scores across patient groups differing on criterion measures ranged from 1.06 to 5.28 points (mean, 3.1 points). Predictive analyses showed that a difference of 3 points on the ACT was associated with a subsequent 76% increased risk (95% CI, 73% to 79%) of excess short-acting beta-agonist use and a 33% increased risk (95% CI, 31% to 35%) of exacerbations. The data support an MID for the ACT of 3 points.