The purpose of this study was to estimate relationships between asthma control and hospital contacts (visits to emergency rooms and hospitalizations) in a group of patients suffering from persistent asthma, after adjustment for prior use of inhaled corticosteroids. A computerized family practice database was used to identify patients (aged 6-50 years) with persistent asthma who received asthma therapy from January 1995. The database provided information on patient demographics and drug therapy. Asthma control was estimated by a survey of patients at the end of a 12-month study period. Frequency of hospital contacts during the study period was related to demographics, asthma control, and prescribed doses of inhaled corticosteroids during a prestudy period. Review of computerized medical files of 497 family practice physicians identified 1,966 patients with persistent asthma who met the study criteria. Of these patients, 1,251 completed the survey (63.6%). Asthma control was assessed in 1,130 patients; it was moderate or poor in 42% of the cases. During the 12-month study period, 14.8% of patients reported at least one hospital contact. The level of asthma control was significantly (p < 0.001) associated with hospital contacts. The odds ratio (OR) for hospital contact for good and poor asthma control was 0.5 (95% confidence interval [CI] 0.2-0.7) and 2.2 (95% CI 1.2-4.4), respectively. Asthma control was related to hospital contacts independently of use of inhaled corticosteroids before the study period. Overall, control of asthma was not optimal in this population. The occurrence of hospital contacts was closely related with the level of control. This association was independent of the dose of inhaled corticosteroids prescribed before the study, suggesting that in asthma, hospital contacts are primarily related to the level of control experienced by the patients.