Cut points for Asthma Control Tests in Mexican children in Orange County, California

Abstract

Background: The Childhood Asthma Control Test (C-ACT) and the Asthma Control Test (ACT) are validated measures of asthma control in which a score of 19 is defined as uncontrolled according to published reports. However, different cut points may exist in different ethnic populations.

Objective: To determine the cut point for uncontrolled asthma in a Mexican descent population from Orange Country, California, compared with an age- and asthma severity-matched non-Hispanic cohort.

Methods: The C-ACT (in children 6-11 years old) and ACT (in children 12-17 years old) scores were collected from 151 children of Mexican descent and 48 non-Hispanic controls with mild-to-moderate asthma who lived in Orange County. Physicians were masked to C-ACT and ACT scores while assessing control based on National Asthma Education and Prevention program guidelines. The receiver operating characteristic method was used to examine the screening accuracy of the tests to detect uncontrolled asthma. The optimal cut points were selected by maximizing the total sensitivity and specificity.

Results: Cronbach α values for the C-ACT (0.76) and the ACT (0.80) confirmed that both tests were reliable in our study population. The C-ACT and ACT scores were statistically higher in children of Mexican descent than non-Hispanic children (P = .008). A cut point of 22 was optimal to detect uncontrolled asthma in children of Mexican descent 6 to 11 years old (group 1: sensitivity, 0.74; specificity, 0.86; area under the curve [AUC], 0.83) and children 12 to 17 years old (group 3: sensitivity, 0.78; specificity, 0.68; AUC, 0.79). For non-Hispanic controls, a cut point of 20 were optimal to detect uncontrolled asthma in children 6 to 11 years old (group 2: sensitivity, 0.70; specificity, 0.91; AUC, 0.86) and children 12 to 17 years old (group 4: sensitivity, 0.83; specificity, 0.87; AUC, 0.91).

Conclusion: In this cross-ethnic validation study, children of Mexican descent in Orange County seem to underreport asthma symptoms compared with a non-Hispanic population and may require higher C-ACT and ACT cut points to detect uncontrolled asthma.

Copyright © 2012 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

ROC curves of C-ACT/ACT in predicting physicians’ assessed uncontrolled asthma. Optimal C-ACT cut-points (A) for Mexican descent and non-Hispanic children were 22 (Group 1, AUC (Area under the curve) = 0.83) and 20 (Group 2, AUC = 0.86) respectively. Optimal ACT cut-points (B) for Mexican descent and non-Hispanic children were 22 (Group 3, AUC = 0.79) and 20 (Group 4, AUC = 0.91) respectively.

Figure 2

A) Percentage of patients (frequency) with uncontrolled asthma (C-ACT/ACT ≤ 22 for Mexican descent and C-ACT/ACT ≤ 20 for non-Hispanics) stratified by FEV1/FVC ratio. B) Percentage of patients with controlled asthma (C-ACT/ACT > 22 for Mexican descent and C-ACT/ACT > 20 for non-Hispanics stratified by FEV1/FVC ratio.