Changes in symptom score and peak expiratory flow (PEF) and the use of medications during the first acute exacerbation of asthma were studied in 41 patients and matched controls who took part in a panel study. An acute exacerbation was defined as the presence of at least one of the following: any unscheduled physician visit, visit to an emergency room, or hospitalization for treatment of asthma; a decrease in PEF by 30% from the patient's best reading; an increase in asthma symptoms during the day and night for over 48 h and not responding to usual medications; and the commencement or doubling of the dose of oral or inhaled steroids for any of the foregoing reasons. Data from -9 to -7 d before the onset of an acute exacerbation were taken as the baseline data, with which the values of the subsequent 14 d were compared. Significant increases in symptoms occurred before a significant reduction in PEF. None of the patients had a decrease of more than 30% in PEF before the onset of symptoms. Daily variation in PEF was not significantly increased from the baseline. Our results suggest that PEF monitoring is not as sensitive as a symptom diary for revealing acute exacerbations of asthma, and that a 30% decrease in PEF is too stringent a criterion for defining an acute exacerbation.