Several reports have documented characteristics of children who die of asthma; however, to the best of our knowledge, no studies have used case controls to clarify the clinical characteristics associated with death. We conducted a case-controlled study of 21 patients with severe asthma hospitalized between 1973 and 1982 who died of asthma sometime following discharge. Average age at death was 13 years (range, 8 to 18 years). Twenty-one asthmatic control cases were matched for age at the time of hospitalization, sex, and severity of illness. Hospital records were evaluated for 57 physiologic and psychological variables. A stepwise discriminant analysis determined that the following eight variables could discriminate the two groups effectively: history of seizures associated with an asthma attack; conflicts between the patient's parents and hospital staff regarding medical management of the patient; self-care of asthma while in the hospital that was not appropriate for age; prednisone dosage having been decreased by more than 50% during the course of hospitalization; inhaled beclomethasone dipropionate required for treatment; increased asthmatic symptoms during the week preceding discharge; depressive symptoms; and disregard of asthmatic symptoms. Most of the clinical characteristics previously thought to place patients at greater risk for a fatal asthmatic attack were found as often in the control cases as in the children who died. This study indicates that psychologic risk factors were prominent in severely asthmatic children who subsequently died of asthma. The variables defined in this study may be important in identifying patients who are at high risk for dying of asthma and in developing treatment plans to prevent deaths.