Send to

Choose Destination
Ann Allergy Asthma Immunol. 2008 Nov;101(5):474-81. doi: 10.1016/S1081-1206(10)60285-2.

Comparison of guideline-based control definitions and associations with outcomes in severe or difficult-to-treat asthma.

Author information

Pharmaceutical Outcomes Research and Policy Program, University of Washington, Seattle, Washington 98195, USA.



The National Asthma Education and Prevention Program put forth guidance on how to measure asthma control. This guidance is flexible regarding how the impairment component of control is assessed.


To evaluate and compare 3 different National Asthma Education and Prevention Program-informed definitions of asthma control and to explore the strength of association between levels of asthma control and future control, asthma-related costs, and health-related quality-of-life utilities.


We used descriptive statistics, the weighted K statistic, and regression methods to compare 3 definitions in a large registry cohort and to explore associations with clinical and health economic outcomes.


We followed up 3,061 patients (> or = 12 years old) in a representative sample of severe or difficult-to-treat asthma. At baseline, 9.1%, 17.1%, or 33.5% of patients were considered well controlled depending on the definition. Lung function was not a component of the definition where 33.5% of patients were well controlled. Weighted K statistics comparing the baseline control definitions ranged from 0.36 to 0.67. The adjusted R2 statistic from the models predicting the future odds of being well controlled, mean asthma-related costs, and mean utilities did not yield consistent results for ranking the definitions of control.


Although all 3 control definitions were correlated with future control and health economic outcomes, there were striking differences in the descriptive results across the definitions, including the proportion of patients in each category of control. Differences among definitions of asthma control may lead to divergent research conclusions or treatment practices. We do not advocate for one particular definition, but we call for standardization of control definitions for specific uses.

Comment in

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center