Measurement of the prevalence of respiratory allergies by interview questionnaire

A recent publication of the National Center for Health Statistics estimated the prevalence of asthma in 1970 to be 30.2 per 1,000 noninstitutionalized population, and of hay fever (without asthma) as 54.2 per 1,000. These and other data on chronic respiratory conditions were obtained by means of a household survey employing lay interviewers and a standardized questionnaire. There may be a gap between the intended meaning of the key question-did the individual have a chronic condition during the past 12 months-and the respondent's understanding of it. As a consequence, some of those who have been asymptomatic in the recent past may have excluded themselves. Even among people with the chronic condition who have seen a physician about it during the prior year, significant numbers fail to report it to the interviewer, according to validation studies. A method is suggested for making rough adjustments for these undercounts. It results in raising the estimated prevalence to 43.0 for asthma and 71.1 for hay fever. These revisions are dependent on necessarily subjective assumptions about the magnitude of the biases involved.