Assessing excess deaths in St. Louis, Missouri: 1980 to 1994.

Excess deaths in the city of St. Louis from 1980 to 1994, their trends over time, and racial death patterns were assessed using the CDC's Wonder mortality database. Death rates in the city were compared with the three lowest statistically valid county death rates in the state of Missouri. The number and percent of preventable deaths in the city population also were estimated. Findings show that approximately 50 percent of city deaths from the nine leading causes may be preventable.

Evaluation of an outbreak of foodborne illness initiated in the emergency department.

STUDY OBJECTIVES: For many foodborne outbreaks, the pathogen and food vehicle never are identified. Delayed collection of epidemiologic and microbiologic information may contribute to this. We postulated that collection of this information from ill persons as they presented to the emergency department during a recent outbreak might contribute to earlier identification of the pathogen and vehicle. DESIGN: At least 690 of 1,900 conventioneers developed gastrointestinal symptoms after attending a banquet. A questionnaire was developed to collect information on specific food histories, incubation periods, symptoms, physical findings, and demographics. These results were compared with results of investigations by the city and state departments of public health. SETTING: The ED of Rush-Presbyterian-St Luke's Medical Center, a tertiary care university hospital in Chicago, Illinois. TYPE OF PARTICIPANTS: Adults (24 men and nine women) presenting to the ED with gastrointestinal symptoms after eating a common meal. MEASUREMENTS AND MAIN RESULTS: The clinical syndrome suggested an invasive pathogen. Based on this, clinical microbiology laboratory procedures were modified (isolation plates were reviewed during the evening shift). This led to early identification of the first isolates (Salmonella enteritidis) from the outbreak. The questionnaire also narrowed the vehicle to one of two foods served. Investigations by the departments of public health subsequently identified one of these, bread pudding with a raw egg based-sauce, as the vehicle. CONCLUSION: Outbreak evaluations can begin in the ED or any other patient care facility. This evaluation need not always add significantly to the expenditure of time, manpower, or laboratory studies. The evaluation of even a small percentage of ill persons from a large outbreak may provide useful epidemiologic information and be particularly important in settings with limited public health resources.