A historical and epidemiologic perspective on respiratory symptoms questionnaires.

Since 1960 the British Medical Research Council (MRC) standardized questionnaire on respiratory symptoms has been used in epidemiologic investigations of chronic respiratory diseases. This review describes the validity, reliability, and susceptibility to bias of the MRC questionnaire. The validity of the phlegm questions as a sensitive index of sputum production has been established but adequate validation of other symptom questions has been hindered by the absence of appropriate, independent criteria for comparison. Consistency of response averages 80 per cent for symptom questions and may reach 99 per cent for reporting of smoking habits. With proper use, the results of the MRC questionnaire are not biased by the observer, by the season of administration, or by method of administration. Modification of the questionnaire by individual investigators may lead to bias when comparisons are made between studies. Because a large literature based on the MRC questionnaire is now available, any new respiratory symptoms questionnaires should be compatible with the MRC questionnaire, and should be calibrated against it.