Studies seeking to identify factors predictive of asthma mortality have relied on information obtained from relatives, other close acquaintances, and doctors who cared for the deceased. We wanted to determine whether asthmatics who have suffered a near-fatal asthma attack (NFA) are similar to asthmatics who have died of asthma with respect to important features, because studies of NFA asthmatics may provide a better insight into causes of asthma death. Such studies would avoid the difficulties associated with seeking information secondhand from proxy informants. Two groups were studied: asthmatics who had suffered a near-fatal asthma attack resulting in a visit to the accident and emergency departments of teaching hospitals (n=154), and asthmatics certified as dying of asthma who, following panel review, were confirmed to have died from this disease (n=80). For each case in the two groups, an interview questionnaire was administered to a close acquaintance (household or family member) and to the general practitioner. Both groups shared many important characteristics. Similarities related to: frequency of symptoms; frequency of hospital and intensive care unit admissions for asthma; use of asthma crisis plans; compliance with prescribed medications; quality of personal asthma management; and asthma severity. The two groups also showed similar psychiatric profiles, and similar use of asthma medications on a regular basis and with increased symptoms. However, NFA cases tended to be younger, were more likely to be male, and less likely to have concurrent medical conditions. Where the final episode was considered definitely or probably preventable, NFA cases were less likely to have experienced delay in actually receiving medical care, and were found to have had more ready access to acute medical care.