Association Between Pulmonary Function and Asthma Symptoms.

ABSTRACT

BACKGROUND: FEV1 as a percentage of predicted (FEV1%pred) is commonly measured in asthma clinical studies; however, reports vary on its association with asthma control instruments evaluating symptoms. OBJECTIVE: Assess the association between FEV1%pred and Asthma Control Questionnaire (ACQ) scores in a managed-care population with persistent asthma. METHODS: Retrospective analysis of survey responses and spirometry results of patients (aged ≥12 years) with persistent asthma from the Observational Study of Asthma Control and Outcomes was done. Eligible patients received 4 identical surveys including the 5-item ACQ (ACQ-5)/6-item ACQ (ACQ-6) and completed spirometry in parallel. Longitudinal analyses, comparisons of change over time, and fixed- and random-effects regression analyses were conducted, with/without adjustment for covariates. RESULTS: There were 1748 survey responses with valid spirometry results. In unadjusted models, coefficients for ACQ-5/ACQ-6 scores were not statistically significant and coefficient of determination (R2) was low (0.03). When adjusted for covariates, ACQ-5 and ACQ-6 scores were significantly associated with FEV1%pred (P < .001) and R2 increased to 0.11 and 0.12, respectively. In adjusted models, every 1-point increase in ACQ-5 and ACQ-6 scores was associated with a 1.7% and 1.9% decrease, respectively, in FEV1%pred. Change in FEV1%pred and change in ACQ-5/ACQ-6 scores were not significantly associated in regressions with/without covariates. CONCLUSIONS: The weak and statistically insignificant association between FEV1%pred and ACQ-5/ACQ-6 scores in unadjusted models suggests a high degree of unexplained variation between these measures. Results support the use of both symptoms and pulmonary function, rather than relying on one measure alone, to assess asthma control in clinical care and outcomes studies.