The value of questionnaires and spirometry in asthma surveillance programmes in the workplace

Shipyard workers and job applicants (n = 1126) completed an asthma questionnaire, and underwent measurements of ventilatory lung function (FEV1, FEV1/FVC and PEFR) and airway responsiveness. Airway responsiveness (to inhaled methacholine) was expressed as the dose calculated to provoke a 20 per cent fall in FEV1, ie the PD20. Results were categorized into four levels: high (PD20 < 200 micrograms), medium (200-1000 micrograms), low (1001-6400 micrograms) and unquantifiable (> 6400 micrograms), which correspond with definite, possible, doubtful and no asthmatic activity. These categories agreed closely with diagnoses of asthma by general practitioners and the use of bronchodilator medication. The sensitivity and specificity of the questionnaire and the measurements of ventilatory function for detecting asthmatic activity were then determined. The questionnaire symptoms (wheeze, chest tightness, undue coughing or abnormal breathlessness) had a low (28 per cent) sensitivity for detecting definite or possible asthmatic activity and a specificity of only 73 per cent. The sensitivity of the ventilatory function tests (any one abnormal) was also low at 21 per cent with a specificity of 92 per cent. When the FEV1 < 80 per cent predicted criterion was considered separately, its sensitivity was 11 per cent and its specificity was 98 per cent. These results illustrate that caution is needed when interpreting the results of questionnaires and measurements of ventilatory lung function in the diagnosis of asthma among working populations.