IAFP 2009 Program Book and Abstract Book - AB166

P3-48 Restaurant Outbreaks 1990–2006 XUMAN A. TIAN and Caroline Smith DeWaal Center for Science in the Public Interest, Washington, D.C., USA Introduction: The Center for Science in the Public Interest (CSPI) maintains a database of foodborne illness outbreaks that have been linked to specific foods. CSPI identified a total of 5,778 outbreaks of illness linked to specific foods, involving 168,898 individual illnesses that occurred between 1990 and 2006. Purpose: The purpose of this study was to analyze the illnesses linked to restaurant outbreaks between 1990 and 2006. Methods: As outbreaks can range in size from 2 to over a thousand persons, the best tool for comparison is illnesses linked to outbreaks, with each illness equal to one person. Using this technique, CSPI researchers compared foodborne illness outbreaks linked to restaurants to those linked to private households and found that restaurant outbreaks were nearly twice as common as private household outbreaks, and restaurant outbreaks affected nearly four times as many people as those in private households. Results: CSPI found that forty-one percent (2,388) of outbreaks identified in CSPI’s database were attributed to restaurants, including 32% (53,399) of the total illnesses. Twenty-two percent (1,250) of outbreaks were attributed to private household, including 8.6% (14,551) of the total illnesses. The specific food categories with the greatest number of illnesses were produce (29%), multi-ingredient foods without meats such as salads, pizza, and sandwiches (17%), multi-ingredient foods with meats (11%), eggs (9%), seafood (8%), poultry (8%), and beef (7%). The most common pathogens implicated in restaurant outbreaks were Salmonella (38%), Norovirus (27%), Clostridum (9%), E. coli (5%) and Shigella (5%). Significance: This analysis showed that over two-thirds (69%) of illnesses linked to restaurant outbreaks were caused by non-meat foods such as seafood, multi-ingredient foods without meats, eggs, produce, dairy, breads, and beverages. Twenty percent were attributed to meats, such as beef, poultry, pork, and luncheon meats, and 11% were multi- ingredient foods or meals containing meat. P3-49 Relative Rates of Illnesses by Food Category, Adjusted for Consumption, 1999 to 2006 Xuman A. Tian and CAROLINE SMITH DEWAAL Center for Science in the Public Interest, Washington, D.C., USA Introduction: The Center for Science in the Public Interest (CSPI) maintains a database of foodborne illness outbreaks that have been linked to specific foods. CSPI identified a total of 5,778 outbreaks of illness linked to specific foods, involving 168,898 individual illnesses that occurred between 1990 and 2006. Purpose: The purpose of this study was to analyze the illnesses linked to outbreaks in CSPI’s database for selected food categories, and compared this data to consumption data for the years 1999 to 2006, to reflect current trends. Methods: As outbreaks can range in size from 2 to over a thousand persons, the best tool for comparison is illnesses linked to outbreak, with each illness equal to one person. CSPI researchers analyzed the illnesses linked to outbreaks for each food category, and compared it to consumption data for the years 1999 to 2006, to reflect current trends. To standardize its analysis, CSPI choose dairy products to reflect a nearly fully-pasteurized product. The data sources used were from USDA Economic Research Service for consumption data. The population estimates were from the US Census Bureau. CSPI provided the food attribution estimates, using data provided by the Centers for Disease Control and Prevention. Results: CSPI found that compared to dairy, reported illnesses were four times more frequent from produce, eight times more frequent from pork, eleven times more frequent from beef, thirteen times more frequent from eggs, fifteen times more frequent from poultry, and twenty-nine times more frequent from fish and shellfish. Significance: The study showed current trends of relative rates of illnesses linked to outbreaks among selected food categories when adjusted for consumption during the period of 1999 to 2006. Acccording to this analysis, fish and shellfish topped the list to cause more sicknesses per pound than any other food category. After dairy, produce is the second safest category of food, followed by pork. Instead of relying on recalls and warnings, the Food and Drug Administration should focus on preventing these problems from ever reaching consumers. Modernizing the food safety system should be an urgent priority to reduce outbreaks and illnesses from food and restore consumer confidence. P3-50 Beef Grinding and Record Keeping Practices, a Survey of Retail Establishments in Three States, 2008 HANNAH GOULD, Dawn Norton, Karen Everstine, Danny Ripley, David Reimann, Moshe Dreyfuss, Wu San Chen, Scott Seys and Carol A. Selman Centers for Disease Control and Prevention, Enteric Diseases Epidemiology Branch, Atlanta, GA, USA Introduction: Ground beef has been implicated as a transmission vehicle in outbreaks of disease caused by Escherichia coli, Salmonella, and other foodborne pathogens. Traceback of contaminated beef to the slaughter facility is often unsuccessful because of inadequate recordkeeping at retail establishments that grind beef products. Purpose: To describe beef grinding and record keeping practices at retail establishments. Methods: Environmental Health Specialists Network (EHS-Net) personnel in three states (California, Minnesota, Tennessee) surveyed a convenience sample of retail establishments that grind beef. In each establishment that kept grinding logs, three randomly selected records were reviewed to determine whether data elements critical for traceback investigations were completed. Results: We surveyed 124 stores, of which 60 (49%) kept grinding logs, including 54 (74%) of 73 chain and 6 (12%) of 51 independent stores. One hundred seventy-six grinding records were reviewed. These establishments ground beef a median of seven times per week (range 2 to 140), and ground a median of 40 pounds (range 2 to 800) per grind. Seventy-three percent of the records included the establishment code of the source beef, 72% included the grind date and time, and 59% 166 — Abstract Book Poster

IAFP 2009 Program Book and Abstract Book

Table of Contents for the Digital Edition of IAFP 2009 Program Book and Abstract Book