BACKGROUND: There is a need for nonclinical measures in studies where objective clinical data used to determine asthma severity, such as spirometry and peak expiratory flow rate meter readings, are not available. OBJECTIVE: To determine whether different methods of defining asthma severity provide different distributions of patients across the categories of mild, moderate, and severe asthma. METHODS: Data were obtained from a mail survey and the database of a managed care organization located in Michigan. Seven methods of asthma severity that used either patient self-report or claims data were evaluated and divided into 3 categories: patient-perceived severity (1 method), symptom-derived severity (4 methods), and medication-derived severity (2 methods). To further evaluate the extent of the degree of agreement among the 7 severity methods, percent agreement and Cohen K scores were calculated. RESULTS: Surveys were sent to 1,139 persons, and 603 responses were usable. Substantial differences exist in the number of patients in each severity level for different methods of determining asthma severity. The proportion of persons identified as having mild, moderate, and severe asthma ranged from 48% to 82%, 16% to 38%, and 2% to 33%, respectively. The percent agreement and Cohen kappa scores were generally low among the 7 methods. CONCLUSIONS: Due to variations caused by the method of measuring asthma severity, comparability among studies using different methods of classifying asthma severity is limited. Although a uniform evaluation of asthma severity is needed, the results of this study do not identify one preferable method. Rather, investigators should know the limitations of the methods used.
BACKGROUND: There is a need for nonclinical measures in studies where objective clinical data used to determine asthma severity, such as spirometry and peak expiratory flow rate meter readings, are not available. OBJECTIVE: To determine whether different methods of defining asthma severity provide different distributions of patients across the categories of mild, moderate, and severe asthma. METHODS: Data were obtained from a mail survey and the database of a managed care organization located in Michigan. Seven methods of asthma severity that used either patient self-report or claims data were evaluated and divided into 3 categories: patient-perceived severity (1 method), symptom-derived severity (4 methods), and medication-derived severity (2 methods). To further evaluate the extent of the degree of agreement among the 7 severity methods, percent agreement and Cohen K scores were calculated. RESULTS: Surveys were sent to 1,139 persons, and 603 responses were usable. Substantial differences exist in the number of patients in each severity level for different methods of determining asthma severity. The proportion of persons identified as having mild, moderate, and severe asthma ranged from 48% to 82%, 16% to 38%, and 2% to 33%, respectively. The percent agreement and Cohen kappa scores were generally low among the 7 methods. CONCLUSIONS: Due to variations caused by the method of measuring asthma severity, comparability among studies using different methods of classifying asthma severity is limited. Although a uniform evaluation of asthma severity is needed, the results of this study do not identify one preferable method. Rather, investigators should know the limitations of the methods used.
Authors: Cynthia A Esteban; Robert B Klein; Elizabeth L McQuaid; Gregory K Fritz; Ronald Seifer; Sheryl J Kopel; Jose Rodriguez Santana; Angel Colon; Maria Alvarez; Daphne Koinis-Mitchell; Alexander N Ortega; Brenda Martinez-Nieves; Glorisa Canino Journal: J Allergy Clin Immunol Date: 2009-07-16 Impact factor: 10.793