JAMA. 2006;295:2241-2243.
MMWR. 2006;55:392-395
Foodborne illnesses are a substantial health burden in the United States.1 The Foodborne Diseases Active Surveillance Network (FoodNet) of CDC's Emerging Infections Program collects data from 10 U.S. states* regarding diseases caused by enteric pathogens transmitted commonly through food. FoodNet quantifies and monitors the incidence of these infections by conducting active, population-based surveillance for laboratory-confirmed illness.2 This report describes preliminary surveillance data for 2005 and compares them with baseline data from the period 1996-1998. Incidence of infections caused by Campylobacter, Listeria, Salmonella, Shiga toxin--producing Escherichia coli O157 (STEC O157), Shigella, and Yersinia has declined, and Campylobacter and Listeria incidence are approaching levels targeted by national health objectives3. However, most of those declines occurred before 2005, and Vibrio infections have increased, indicating that further measures are needed to prevent foodborne illness.
In 1996, FoodNet began active, population-based surveillance for laboratory-confirmed cases of infection from Campylobacter, Listeria, Salmonella, STEC O157, Shigella, Vibrio, and Yersinia. In 1997, FoodNet added surveillance for cases of Cryptosporidium and Cyclospora infection. In 2000, FoodNet began collecting data on STEC non-O157 and comprehensive information on hemolytic uremic syndrome (HUS). FoodNet personnel ascertain cases through contact with all clinical laboratories in their surveillance areas. HUS surveillance is conducted through a network of pediatric nephrologists and infection-control practitioners. In addition, eight states review hospital discharge data to ascertain HUS cases. Because of the time required for review of hospital records, this report contains preliminary 2004 HUS data.
During 1996-2005, the FoodNet surveillance population increased from 14.2 million persons (5% of the U.S. population) in five states to 44.5 million persons (15% of the U.S. population) in 10 states. Preliminary incidence for 2005 was calculated using the number of laboratory-confirmed infections and dividing by 2004 population estimates. Final incidence for 2005 will be reported when 2005 population estimates are available from the U.S. Census Bureau.
2005 Surveillance
In 2005, a total of 16,614 laboratory-confirmed cases of infections in FoodNet surveillance areas were identified, as follows: Salmonella (6,471 cases), Campylobacter (5,655), Shigella (2,078), Cryptosporidium (1,313), STEC O157 (473), Yersinia (159), STEC non-O157 (146), Listeria (135), Vibrio (119), and Cyclospora (65). Overall incidence per 100,000 population was 14.55 for Salmonella, 12.72 for Campylobacter, 4.67 for Shigella, 2.95 for Cryptosporidium, 1.06 for STEC O157, 0.36 for Yersinia, 0.33 for STEC non-O157, 0.30 for Listeria, 0.27 for Vibrio, and 0.15 for Cyclospora. Substantial variation occurred across surveillance sites. In 2004, FoodNet identified 44 cases of HUS in children aged <15 years (rate: 0.49 per 100,000 children); 30 (68%) of these cases occurred in children aged <5 years (rate: 0.94).
Of the 5,869 (91%) Salmonella isolates serotyped, six serotypes accounted for 61% of infections, as follows: Typhimurium, 1,139 (19%); Enteritidis, 1,080 (18%); Newport, 560 (10%); Heidelberg, 367 (6%); Javiana, 304 (5%); and a monophasic serotype identified as Salmonella I 4,[5],12:i:-, 154 (3%). Among 109 (92%) Vibrio isolates identified to species level, 59 (54%) were V. parahaemolyticus, and 15 (14%) were V. vulnificus. FoodNet also collected data on 145 STEC non-O157 isolates that were tested for O-antigen determination; 117 (81%) had an identifiable O antigen, including O26 (37 [32%]), O103 (36 [31%]), and O111 (23 [20%]); 28 isolates did not react with the typing antisera used.
In 2005, FoodNet sites reported 205 foodborne disease outbreaks to the national Electronic Foodborne Outbreak Reporting System; 121 (59%) were associated with restaurants. Etiology was reported for 159 (78%) outbreaks; the most common etiologies were norovirus (49%) and Salmonella (18%).
Comparison of 2005 Data With 1996-1998
A main-effects, log-linear Poisson regression model (negative binomial) was used to estimate statistically significant changes in the incidence of pathogens. This model accounts for the increase in the number of FoodNet sites and its surveillance population since 1996 and for variation in the incidence of infections among sites.2 The average annual incidence for 1996-1998 (1997-1998 for Cryptosporidium), the first 3 years of FoodNet surveillance, was used as the baseline period. For HUS surveillance, 2000-2001 was used as the baseline. The estimated change in incidence (relative rate) between the baseline period and 2005 was calculated, along with a 95% confidence interval (CI).
The estimated annual incidence of several infections declined significantly from 1996-1998 to 2005. The estimated incidence of infection with Yersinia decreased 49% (CI = 36%-59%), Shigella decreased 43% (CI = 18%-60%), Listeria decreased 32% (CI = 16%-45%), Campylobacter decreased 30% (CI =25%-35%), STEC O157 decreased 29% (CI = 12%-42%), and Salmonella decreased 9% (CI = 2%-15%). Although Salmonella incidence decreased overall, of the five most common Salmonella serotypes, only the incidence of S. Typhimurium decreased significantly (42% [CI = 34%-48%]). The estimated incidence of S. Enteritidis increased 25% (CI = 1%-55%), S. Heidelberg increased 25% (CI = 1%-54%) and S. Javiana increased 82% (CI = 14%-191%). The estimated incidence of S. Newport increased compared with the baseline, but the increase was not statistically significant. The estimated incidence of postdiarrheal HUS in children aged <5 years decreased 45% in 2004 compared with 2000-2001; whether this trend is significant could not be determined, partly because the limited time span does not provide enough data to evaluate a Poisson regression model. The estimated incidence of Vibrio increased 41% (CI = 3%-92%) compared with the baseline, whereas the estimated incidence of Cryptosporidium infections did not change significantly.
Continue Reading...
