Background-Severe life threatening asthma (SLTA) is important in its own right and as a proxy for asthma death. In order to target hospital based intervention strategies to those most likely to benefit, risk factors for SLTA among those admitted to hospital need to be identified. A case-control study was undertaken to determine whether, in comparison with patients admitted to hospital with acute asthma, those with SLTA have diVerent sociodemographic and clinical characteristics, evidence of inadequate ongoing medical care, barriers to health care, or deficiencies in management of the acute attack. Methods-Seventy seven patients with SLTA were admitted to an intensive care unit (pH 7.17 (0.15), PaCO 2 10.7 (5.0) kPa) and 239 matched controls (by date of index attack) with acute asthma were admitted to general medical wards. A questionnaire was administered 24-48 hours after admission. Results-The risk of SLTA in comparison with other patients admitted with acute asthma increased with age (odds ratio (OR) 1.04/year, 95% CI 1.01 to 1.07) and was less for women (OR 0.36, 95% CI 0.20 to 0.68). These variables were controlled for in all subsequent analyses. There were no diVerences in other sociodemographic features. Cases were more likely to have experienced a previous SLTA (OR 2.04, 95% CI 1.20 to 3.45) and to have had a hospital admission in the last year (OR 1.86, 95% CI 1.09 to 3.18). There were no diVerences between cases and controls in terms of indicators of quality of ongoing asthma specific medical care, nor was there evidence of disproportionate barriers to health care. During the index attack cases had more severe asthma at the time of presentation, were less likely to have presented to general practitioners, and were more likely to have called an ambulance or presented to an emergency department. In terms of pharmacological management, those with SLTA were more likely to have been using oral theophylline (OR 2.14, 95% CI 1.35 to 3.68) and less likely to have been using inhaled corticosteroids in the two weeks before the index attack (OR 0.69, 95% CI 0.47 to 0.99). While there was no diVerence in selfmanagement knowledge or behaviour scores, those with SLTA were more likely to have inappropriately used oral corticosteroids during the acute attack (OR 2.09, 95% CI 1.02 to 4.47). Conclusions-In comparison with those admitted to hospital with acute severe asthma, patients with SLTA were indistinguishable on sociodemographic criteria (apart from male predominance), were more likely to have had a previous SLTA or hospital admission in the previous year, had similar quality ongoing asthma care, had no evidence of increased physical, economic or other barriers to health care, but had demonstrable deficiencies in the management of the acute index attack. Educational interventions, while not losing sight of the need for good quality ongoing care, should focus on providing individual patients with better advice on self-management of acute exacerbations.