The U.S. Department of Agriculture has been directed by Congress to provide an annual report on the incidence of foodborne illness in the United States. This report is to be submitted to the House and Senate Committees on Appropriations.

This first report includes a description of the Sentinel Site Study and foodborne illness data for calendar year 1996.

Questions about this report or about FSIS may be directed to the Food Safety and Inspection Service, U.S. Department of Agriculture, Washington, D.C. 20250.



Overview of the Sentinel Site Study

The Food Safety and Inspection Service (FSIS) of the U. S. Department of Agriculture (USDA) administers a comprehensive system of inspection laws to ensure that meat, poultry, and egg products moving in interstate and foreign commerce for use as human food are safe, wholesome, and accurately labeled.

Over the past decade, foodborne illnesses caused by bacterial contamination have heightened concern about the safety of food. The Centers for Disease Control and Prevention (CDC) has identified bacterial pathogens as the most common cause of foodborne illnesses because they can be easily transmitted and multiply rapidly in food, thus making them difficult to control. CDC has targeted seven bacterial foodborne pathogens (Campylobacter, E. coli O157:H7, Listeria, Salmonella, Shigella, Vibrio, and Yersinia) as those of greatest concern to public health. Government economists have estimated that foodborne illnesses cost billions of dollars each year in lost productivity and medical care. However, an accurate picture of both the number and causes of foodborne illnesses in the United States is unknown.

In July 1995, as part of the federal government's campaign to improve the safety of the nation's food supply, FSIS began a collaborative project with CDC and the Food and Drug Administration, known originally as the Sentinel Site Study and more recently as FoodNet, to collect more precise information on the incidence of foodborne disease in the United States. This project includes establishing direct working links with state and local health departments at five selected sites nationwide as a means of obtaining the most complete and current foodborne illness data available. This study, which is expected to continue for several years, will provide much-needed baseline data regarding the incidence of foodborne illness in the United States which is attributable to consumption of meat, poultry, and other food products.

Background

Bacterial foodborne diseases are currently reported to local and state health departments and CDC through passive surveillance systems. As with all passive systems, cases are frequently unreported. In fact, only 1% to 5% of foodborne disease cases are believed to be reported to CDC. Extrapolation of this data suggests there are between 6 to 81 million cases in the United States annually. There is clearly a need for more accurate data.

In July 1994, the USDA Pathogen Reduction Task Force recommended that FSIS work with CDC on research and surveillance activities that will better characterize risks for foodborne pathogens. This recommendation echoed a National Academy of Sciences recommendation for more community-based surveillance of foodborne disease. To improve data on the incidence and causes of foodborne illness, FSIS and CDC have established five foodborne illness sentinel sites to study the epidemiological links among outbreaks of foodborne illness and to explore what relationships may exist between outbreaks and the types of meat, poultry, and egg products consumed. The five selected sites are Northern California, Oregon, the Minneapolis/St. Paul metropolitan area, the metropolitan Atlanta area, and most of Connecticut.

Objectives

The major objectives of the Sentinel Site Study are to:

• determine the yearly incidence of diarrheal illness due to bacterial foodborne pathogens (i.e., Campylobacter, E. coli O157:H7, Listeria, Salmonella, Shigella, Vibrio, and Yersinia);

• develop a network to collaboratively respond to emerging foodborne diseases;



• determine the proportion of foodborne disease cases attributed to specific commodities; and



• determine whether federal interventions are having a measurable effect on the incidence of foodborne illness attributable to consumption of meat, poultry, and other foods.

Scope and Duration of Study

The Sentinel Site Study is a comprehensive effort to track major pathogens that cause foodborne illness, to examine the epidemiological links among outbreaks of foodborne illness, and to explore what relationships may exist between outbreaks and the types of food products consumed.

Active surveillance data will be collected on reported illnesses associated with the seven bacterial pathogens targeted by CDC. Collected data will be used to identify emerging

pathogens and monitor illness incidence. The results will be used to help monitor the effectiveness of new food safety programs aimed at reducing the levels of these pathogens in meat, poultry, and other foods.

For the project to be successful, data must be collected over a number of years to chart national trends and consider the effectiveness of control strategies. An inherent caution in such a study must be against using the data too soon to draw conclusions and develop future policies and plans. The data at this point are very preliminary and cannot be relied upon as an indicator of either the prevalence of foodborne illness or its causes. It is hoped that the success of this Study will prompt the future collection of such data in state health departments all over the country so that a truly accurate picture of the extent and causes of foodborne illness will be achieved.

Components of the Study

The Study was undertaken in stages following an initial 3-month pilot. Survey instruments (questionnaires) were developed jointly by CDC, FSIS, FDA, and the five sites. The components of the Study are:

• Laboratory Survey: A baseline survey of 230 clinical labs which perform stool cultures in the five sites was completed in late 1995. This survey will be repeated beginning in January, 1997. The primary purpose of this survey was to determine whether or not laboratories within the site boundaries were performing cultures for foodborne pathogens.



• Physician Survey: Each site mailed questionnaires to randomly selected physicians during 1996. Approximately 5000 questionnaires were mailed. Site response rates range from 50 to 60 percent. The primary purpose of this survey was to determine whether or not physicians who see patients for diarrheal complaints were referring them for laboratory analysis.



• Population Survey: An outside contractor, Macro International, piloted the questionnaire with 100 interviews in February, 1996. Approximately 750 actual interviews are being completed each month. Both adult and pediatric surveys are being used. The primary purpose of this survey is to determine population behavior in the five site areas, particularly regarding foods consumed.



• Case-Control Studies: The E. coli O157:H7 study began on March 1, 1996 for both adults and children. The Salmonella study began on May 1 for adults and children. Infant Salmonella case control began shortly afterward. The primary purpose of these studies is to determine patient behavior just prior to becoming ill. Case-Control Studies for other pathogens are not being conducted due to limitations in funding.

Future Plans

Plan for 1997:

As of January 1, 1997, the Sentinel Site Study will be known as "FoodNet."

Plan for 1998:

Food Safety Initiatives

USDA places a high priority on improving the safety of the food supply through increased research, education, and improved inspection systems. The implementation of Hazard Analysis and Critical Control Point (HACCP) systems will help assure that food safety risks are being addressed throughout the production continuum.

Due to the high incidence of Campylobacter revealed by the study in 1996, it is critical to expand FoodNet to include case-control and population-based studies for this pathogen to determine the risk factors for infection and determination of antibiotic resistance patterns among Campylobacter strains. Therefore, for 1998, FSIS proposes an increase of $500,000 for the collection of Campylobacter data and $565,000 for HACCP training for state and local public health officials.



Foodborne Illness Data

The following information represents findings from the first year of data collection. It is important to note, however, that the tables and charts that follow are based solely on the number of laboratory-confirmed actual cases in the sites. There has been no analysis at this point of the probable causes of the illnesses reported herein, only the pathogen found in each patient.

Population in sites:

Active Cases:

There were 7,259 laboratory-confirmed diarrheal cases in 1996 that were attributable to the seven bacterial pathogens targeted by CDC (i.e.,Campylobacter, E. coli O157:H7, Listeria, Salmonella, Shigella, Vibrio, and Yersinia).

Outbreaks:

Although the primary focus of the Study is to determine the incidence of foodborne illness, one particularly useful benefit of data collection is the identification of outbreaks. For example, five outbreaks were identified during 1996: Yersinia in infants in Atlanta linked to chitterlings; Salmonella in Oregon and Vancouver linked to alfalfa sprouts;

E. coli O157:H7 in Connecticut, New York, and Illinois attributable to lettuce, E. coli O157:H7 in Connecticut and Massachusetts attributable to apple cider, and E. coli O157:H7 in a daycare center in Minnesota.

Most frequently isolated pathogen:

Campylobacter is the most frequently isolated bacterium from persons with diarrhea (45%); Salmonella is second (30%); Shigella is third (17%); E. coli O157:H7 is fourth (5%).

Deaths from Salmonella:

Of the 7,259 cases recorded during 1996, it is known that 33 resulted in death. It was decided that a hospital-based follow-up investigation would be conducted in Atlanta to see if the identified pathogen contributed to the patient's death. This type of investigation will be extended to other sites during 1997.

Data obtained from the case-control studies at the Atlanta site showed that eight deaths were attributed to Salmonella infections. Six among those who died had an underlying illness, e.g., three had AIDS. The infecting Salmonella serotype was Typhimurium in six cases. According to CDC, extrapolation of these data suggests that Salmonella may cause over 800 deaths yearly nationwide, and risk factors associated with death include infection with serotype Typhimurium. Persons with underlying illness may be at greater risk for fatal disease.

| Office of Public Health and Science | FSIS | USDA |