The objective of this study was to determine stable estimates of the incidence, case fatality, and epidemiologic features of adult epiglottitis, and risk factors for intubation. The authors designed a retrospective cohort combined with a nested case-control study, followed by detailed analysis of cases from two tertiary care institutions. Among 813 cases, the incidence was 2.02 cases/10(5) population per year. Ten recorded deaths constituted a case fatality rate of 1.2% (95% confidence interval [CI]: 0.5% to 1.9%). The eight fully documented deaths indicated no sudden episodes of catastrophic upper airway obstructions without previous dyspnea. A detailed review of 51 cases revealed that 18% of patients underwent expeditious intubation. Patients managed without initially requiring intubation did not need emergency airway interventions. Only the presence of dyspnea (noted in 29% of patients) at the time of admission (P < 0.001) predicted the need for intubation. A low case fatality rate in a conservatively managed cohort and the absence of sudden upper airway catastrophes in patients without dyspnea suggest that prophylactic intubation and intensive care unit monitoring is not warranted in all patients. An early complaint of dyspnea may safely discriminate between patients requiring invasive airway management and close observation.