This study describes a national survey estimating the number and characteristics of anesthesia-related deaths. Method—Death certificates from 1999 were selected using ICD codes and variable sampling fractions. A questionnaire was sent to certifying physicians (response rate 97%), and a peer review meeting was offered to the anesthesiologist (acceptance rate 97%). The records were analyzed to determine the mechanism of the incident and its attribution to anesthesia. Mortality rates were estimated based on data from a survey covering the year 1996 and compared to those from a survey conducted between 1978 and 1982. Results - The rates of deaths wholly or partially related to anesthesia were 6.9 (95% CI 2.2–12) and 47 (31–63) per million, respectively. They increased with age and underlying conditions, ranging from 5 to 554 per million across ASA classes I through IV. Among the causes, inhalation pneumonias were the leading cause, but intraoperative hypotension and anemia associated with myocardial ischemia were also significant. Deviations from professional standards and organizational deficiencies were frequently associated. Conclusion - Compared to the 1978–1982 survey, the rate of anesthesia-related deaths appears to have been reduced by a factor of ten. (R.A.)