Effect of peak expiratory flow data quantity on diagnostic sensitivity and specificity in occupational asthma

Serial peak expiratory flow records are recommended in the first-line investigation of suspected occupational asthma. The effects of sequentially reducing the numbers of working weeks, consecutive days at work and readings taken per day on diagnostic sensitivity and specificity were investigated, using good quality peak expiratory flow records from 81 workers with independently confirmed occupational asthma and 60 asthmatics without occupational exposure. Sensitivity was 81.8% for records of 4 weeks' duration and 70% for those of 2 weeks' duration (specificity 93.8 and 82.4% respectively). The sensitivity fell to 56.7% if there were only 2 consecutive workdays in each work period. Although best at 8 readings x day(-1), sensitivity and specificity were acceptable with four daily readings (82.4 and 87%). The effect of defining a record as being of adequate quality if it was of > or = 2.5 weeks' duration, with > or = 4 readings x day(-1) and > or = 3 consecutive workdays in each work period, was tested in records not used in the initial data reduction process. The sensitivity and specificity respectively of adequate records were 78.1 and 91.8 versus 63.6 and 83.3% for inadequate records. Peak expiratory flow records for the diagnosis of occupational asthma should be interpreted with caution if they do not satisfy the suggested minimum data quantity criteria.