Printer Friendly

A summary of health outcomes: multistate foodborne disease outbreaks in the U.S., 1998-2007.


U.S. consumers have become used to hearing about foodborne disease outbreaks (FBDOs), and multistate outbreaks in particular garner some of the news coverage. Multistate outbreaks, or outbreaks that either spread to other states or originate from the same vector in multiple states at the same time, pose a particular threat to public health. The continued growth and expansion within the food industry and the transport of food across multiple state lines have made identifying the sources of and subsequently tracking multistate FB DOs more complicated (Allos, Moore, Griffin, & Tauxe, 2004), increasing the likelihood of more cases of foodborne illness linked to a particular outbreak. Additionally, outbreaks caused by fresh produce transported across state lines have become more prominent; the proportion of outbreaks attributed to fresh produce has increased from <1% of all reported FBDOs in the 1970s to 6% during the 1990s (Sivapalasingam, Friendman, Cohen, & Tauxe, 2004). Approximately 12% of all FBDOs occurring between 1990 and 2003 were associated with produce (Dewaal et al., 2006).

Although the incidence of FBDOs has decreased since 1990 (Silver & Bassett, 2008), major outbreaks continue to occur. In 2006, 1,270 FBDOs were reported, resulting in 27,634 reported individual cases of illness and 11 deaths (Ayers, Williams, Gray, Griffin, & Hall, 2009). In summer 2010 a multistate foodborne disease outbreak of human Salmonella Enteritidis was attributed to shell eggs and led to a voluntary nationwide recall. As of December 2010 the outbreak had affected 1,939 people in 11 states and was eventually linked to an unintentional contamination of animal feed (Centers for Disease Control and Prevention [CDC], 2010). Multistate FBDOs are particularly concerning because of the difficultly in tracking and following up on cases after outbreaks have crossed state lines, leading to a potential increase in negative health outcomes.

In addition to the impact on health outcomes, the economic costs of FBDOs are substantial as well. The costs associated with FBDOs total just under $152 billion each year to U.S. residents, including direct and indirect costs, with the average case totaling $1,851 (Scharff, 2010). Loss in productivity ranges between $377 and $924 per case (Scharff, 2010). FBDOs have been associated with increased spending on response including vaccinations, medical care, hospitalizations, and administrative costs (Dalton, Haddix, Hoffman, & Mast, 1996). The costs associated specifically with multistate outbreaks, however, have not been calculated. These findings indicate the need for an analysis of the human costs associated with multistate outbreaks separate from intrastate ones in order to improve upon how public health officials identify and respond to these types of food safety hazards.

In addition to cost, it is also necessary to consider factors associated with response to a multistate FBDO. Taylor and co-authors (2010) identified three factors that contribute to poor response to multistate FBDOs in particular, including "1. delayed response due to discrepancies in available resources and expertise at state and local levels, 2. inadequate communication between stakeholders and agencies, and 3. poor traceability capabilities (Taylor, Kastner, & Renter, 2010)." The absence of a coordinated response system to identify and mitigate the effects of multistate FBDOs poses a particular threat to the ability of public health officials to limit the negative health consequences associated with a multistate outbreak. The existence of a comprehensive emergency response plan, such as those recommended in the Food and Drug Administration's (FDAs) Food Protection Plan (2007) and the Council to Improve Foodborne Outbreak Response's (CIFOR's) Guidelines for Foodborne Disease Outbreak Response (2009), allows for effective communication and coordination between affected states as well as potentially affected states and may result in fewer negative health outcomes. Having a horizontally integrated system of communication in place may reduce the burden on labor, time, and resources required by states to notify one another and the Centers for Disease Control and Prevention (CDC) of foodborne illness.

The public health and economic effects of recent multistate FBDOs signal potential vulnerabilities in current practices of food safety and food defense. Previous research has reported that the incidence of multistate outbreaks has increased (Crutchfield & Roberts, 2000); further investigation of the outcomes associated with this increase is warranted and may help to identify areas for improvement of the current food safety and defense system in the U.S. Our study sought to describe the incidence and health outcomes of multistate FBDOs occurring in the U.S. between 1998 and 2007 and to highlight patterns associated with cases that may help to identify potential vulnerabilities in the current food safety and defense system. (At the time of data collection, data for 1998-2007 were all that were made available by CDC.)


CDC's OutbreakNet Foodborne Outbreak Online Database was used to access information on FBDOs attributed to unintentional food contamination that had been reported to and confirmed by CDC between the years 1998 and 2007. CDC maintains three surveillance systems that collect data on foodborne illness, including ePulseNet, FoodNet, and the National Outbreak Reporting System (Selman, 2010). Information on foodborne illness is voluntarily submitted to CDC by territorial, local, and state health departments using the Foodborne Outbreak Reporting System and is classified as "the occurrence of two or more cases of a similar illness resulting from the ingestion of a common food (CDC, 2011)." The database was released for public use in fall 2009. All states may not report all incidences; however, the system is comprehensive and likely the most complete source of public data available. The database provides information on year of outbreak; location (which varies from home family gatherings to nursing homes); type of location in which the infected food was reported to have been served; specific strain or type of virus (bacterial, parasitic, or chemical agent); and status of the vector as the cause of the outbreak (agent had been determined a confirmed, suspected, or non-applicable culprit). Using SAS 9.1 for the UNIX environment, the index function was used to classify the vectors into 29 general agents (e.g., all Salmonella strains were classified as "Salmonella"). Additionally, only outbreaks consisting of three or more confirmed cases were included. The purpose of this was to allow for more robust analyses and a more relevant interpretation of the outbreaks.

Outbreaks with vectors confirmed by CDC affecting patients in the 50 states were included, as well as those occurring in Guam, the District of Columbia, and Puerto Rico. FBDOs resulting in ill parties in more than one state were included as a separate attribute "multistate"; these outbreaks were exclusive of those in which victims were contained to a single state. The variable "state" (which included the state where FBDO cases were confirmed) was recoded into a dichotomous variable "multistate" (multistate FBDO = 1 and intrastate outbreak = 0). All aforementioned variables were initially collected in October 2009 and included in analysis conducted between August and October 2010.

To account for population growth over time, U.S. Census projections for total population by state were added to the dataset for the years 1998-2007 (U.S. Census Bureau, 2011) and adjusted incidence rates were calculated to more accurately assess the linear trend of FBDOs. Satterthwaite t-tests were run on all outbreaks to compare means of total illnesses, total hospitalizations, and total deaths between multistate and intrastate outbreaks, and logistic regression was applied to predict the class of outbreaks most associated with these outcomes. Univariate analysis was conducted on outcomes of illness, hospitalization, and death by month, year, food product, and vector associated with both intrastate and multistate outbreaks. ANOVA tested for significant changes from year to year in the outcomes of illness, hospitalization, and death resulting from multistate outbreaks; least significance difference post-hoc tests were performed.


Over 4,600 separate outbreaks occurred between 1998 and 2007, resulting in 143,260 ill (mean [bar.x] = 31.10, SD = 68.65), 6,385 hospitalizations ([bar.x] = 1.67, SD = 5.58), and 158 deaths ([bar.x] = 0.04, SD = 0.52). The maximum persons ill, hospitalized, or dead associated with a single outbreak were 1,644, 129, and 21, respectively, with Campylobacter confirmed as the vector. In 398 outbreaks only two people were reported ill; these outbreaks made up 17.28% of the overall sample and were not included in the regression. The most frequent months for deaths associated with outbreaks were May (n = 27) and October (n = 32). The t-tests revealed a significant difference between intrastate and multistate FBDOs for the outcomes of illness, hospitalization, and death (Table 1).

Logistic regression modeled multistate versus intrastate outbreaks as the dependent variable to predict total illnesses, total hospitalizations, and total deaths in a given outbreak. Results indicated that hospitalization and death were more likely to be associated with a multistate FBDO than an interstate one (hospitalization: p < .001, log odds ratio [logOR] = 1.206; death: p < .05, logOR = 1.284).

The following information summarizes the data on multistate outbreaks. Results provided are not state specific but are a composite of data collected from all the states impacted in particular multistate FBDOs. Of all outbreaks occurring during the 10-year period, 107 (2.54%) were multistate outbreaks and accounted for 5.53%, 17.68%, and 36% of total illnesses, hospitalizations, and deaths, respectively. In 2007 (the last year included in analysis), multistate FBDOs made up 3.59% of all outbreaks (see Figure 1 for the increase in proportion of multistate to intrastate outbreaks over time). In terms of multistate outbreak occurrences, the range was 5-18 total over the 10-year period, with a slight overall increase over time (Figure 2).

The average illnesses, hospitalizations, and deaths for multistate outbreaks were 74.075 (SD = 106.24), 14.11 (SD = 23.23), and 0.826 (SD = 2.88), respectively. See Table 2 for totals by year. The year with the greatest total of multistate FBDOs was 2007 (n = 18), with the fewest multistate FBDOs occurring in 2001 (n = 5). While the total number of people ill in a multistate outbreak did not experience a statistically significant change from year to year, significant changes were seen in total hospitalizations (p < .05); Table 2 summarizes the outcomes associated with multistate FBDOs and indicates significant changes in totals from year to year.

Months of the year during which incidence of multistate food-related emergencies were higher were July and November, with 17 and 13 total outbreaks occurring, respectively, over the 10-year period of study; the fewest outbreaks were reported in December (n = 2). July (n = 1,857), June (n = 1,127), and February (n = 1,043), however, were the months with the highest total illnesses; December again had the lowest total (n = 97). The frequency of outbreaks and total illnesses were correlated (r = .75). The highest rate of negative heath consequences associated with a single multistate outbreak was a 2002 case in which Salmonella made 510 consumers ill.

The vectors most often identified as the cause of the FBDOs in this sample were Salmonella (n = 57) and E. coli (n = 30), making up 81% of all multistate outbreaks. Table 3 reports the total illnesses, hospitalizations, and deaths associated with each vector confirmed in a multistate outbreak.


The findings in our study add to the literature highlighting the persistence of multistate FBDOs as a threat to public health (CIFOR, 2009; Sobel, Griffin, Slutsker, Swerdlow, & Tauxe, 2002) by describing morbidity and mortality over time. The proportion by which multistate outbreaks made up total outbreaks between 1998 and 2007 increased slightly as well. The Public Health Security and Bioterrorism Preparedness and Response Act (2002) includes several provisions intended to provide for the safety and protection of the U.S. food supply. Since 2003, the total multistate FBDOs occurring in a given year have totaled 10 or more, an increase over the previous decade. More needs to be done to address the threat of multistate food safety incidents as a matter of national security.

It is important to note that totals presented in this article are likely conservative estimates: these numbers capture only a portion of the FBDOs. Many people who are directly affected by a foodborne disease do not go to the hospital for treatment or are not admitted. In 2007, the Agency for Healthcare Research and Quality recorded 6,495 discharges from hospitals in the U.S. in which food poisoning (ICD9 code 005.9) was listed as one of the diagnoses (Agency for Healthcare Research and Quality, 2011); however, analysis of the OutbreakNet foodborne outbreak online database data revealed a total of 753 hospitalizations, a much smaller number, signaling that all hospitalizations, and perhaps outbreaks, are not being reported to CDC. Further investigation into states' current strategies for surveillance and response to multistate FBDOs as well as communication of outcomes to the CDC are needed to ensure an accurate assessment of illness and appropriate magnitude of response.

Statistically significant changes in outcomes were seen from year to year, and several instances occurred in which this difference approached statistical significance (e.g., the surge in hospitalizations between 2005 and 2006). Furthermore, the total number of cases attributed to Salmonella and E. coli indicate that prevention mechanisms should pay particular attention to these bacteria; Salmonella, Campylobacter, E. coli, Shigella, and Clostridium poisoning have been linked to bacterial gastroenteritis (Meehan, Atkeson, Kepner, & Melton, 1992), which can result in chronic conditions such as irritable bowel disease (Ruigomez, Garcia Rodriguez, & Panes, 2007). Moreover, as we see particular disease vectors more closely associated with multistate FBDOs, it provides an opportunity to focus on prevention efforts for that particular vector and perhaps the production of more sensitive testing to detect food contamination prior to human consumption.

As the likelihood of hospitalization and death has significantly increased over time in multistate FBDOs, the direct and indirect costs of health outcomes associated with this class of FBDOs must be investigated. Although multistate FBDOs are more difficult to detect due to limitations with surveillance and communications (Lynch, Tauxe, & Hedberg, 2009), an accurate interpretation of health outcomes and direct and indirect economic costs of these outbreaks is imperative to understand the true impact on food safety and defense maintenance. Furthermore, encouraging cooperation and input from industry (producers and distributors), as has been done with increased encouragement of the development of food defense plans by the Food Safety Inspection Service, might enhance coordination within the food industry and with government agencies at the local, state, and national levels. This coordination could lead to a more collaborative and productive relationship, rather than just the government driving food safety and defense initiatives. For example, Georgia and Idaho now target industry communication and inspections, respectively (Zhang, 2009). In April 2010, the Florida State Senate passed the Tomato Food Safety law, which revises and increases safety standards for tomatoes, and includes inspections and administrative fines (Tomato Food Safety Standards, 2010).

Several limitations to our analysis should be noted. Only foodborne illness reported to the CDC that affected three or more people were included in our analysis. Among the multistate FBDOs in the database, no specific information existed on exactly how many states were affected in each outbreak or the specific states. While this information would have proven meaningful, our recommendations suggest review of current practices among states regarding outbreaks that affect more than one state overall. Finally, the nature of FBDO reporting, as well as rates of uninsurance and underinsurance and therefore decreased likelihood of presentation at a medical facility, lend themselves to underreporting and make the results of our study highly conservative.


Our study analyzed changes in multistate FBDOs over time and interpreted the effects on morbidity and mortality. Our findings add to the literature citing a gradual increase in the incidence of multistate FBDOs, suggesting that this is a problem shared among states, not merely one individual state, and therefore plans for response should be coordinated across states. The variability of incidence and outcomes highlights a potential vulnerability in food defense and illustrates a need for states to restructure--or develop--systems by which industry and government can communicate information about risks to the public and each other effectively and coordinate appropriate mechanisms for response across jurisdictions. This is particularly important among contiguous jurisdictions or states that receive goods from the same source or by similar means. The CIFOR Guidelines for Foodborne Disease Outbreak Response (2009) provides a template for states to follow and should be adopted and modified to formally create the most relevant plans at the state level, whether food originates from a domestic or international source (CIFOR, 2009). Furthermore, results from our study suggest that public education encouraging safe food handling and preparation may need to be increased during the summer and fall, in preparation for major American holidays of which food plays an important role (Independence Day and Thanksgiving); previous research has indicated that this method is effective in increasing knowledge and awareness around this issue (Ratnapradipa, Quilliam, Wier, & Rhodes, 2011). The higher incidence of outbreak and resulting outcomes during the month of February should be investigated further as well.

More comprehensive analyses are needed to investigate the outcomes of multistate FBDOs and will be possible as more data are released. For example, investigation of reporting trends and consistency in reporting of FBDOs are needed to truly understand the effect on health outcomes. Further research investigating multistate FBDOs will allow us to better understand the implications of current practices on health outcomes and food defense initiatives and interpret appropriate measures to improve relevant economic and policy developments.


Agency for Healthcare Research and Quality. (2011). Healthcare cost and utilization project (HCUP). [Software]. Retrieved from

Allos, B.M., Moore, M.R., Griffin, P.M., & Tauxe, R.V (2004). Surveillance for sporadic foodborne disease in the 21st century: The FoodNet perspective. Clinical Infectious Diseases, 38(Suppl. 3), S115-S120.

Ayers, L.T., Williams, I.T., Gray, S., Griffin, P.M., & Hall, A.J. (2009). Surveillance for foodborne disease outbreaks--United States, 2006. Morbidity and Mortality Weekly Report, 58(22), 609-615.

Centers for Disease Control and Prevention. (2010). Investigation update: Multistate outbreak of human Salmonella Enteritidis infections associated with shell eggs. Retrieved from http://www.cdc. gov/salmonella/enteritidis/Centers for Disease Control and Prevention. National Outbreak Reporting System. (2011). Retrieved from

Council to Improve Foodborne Outbreak Response. (2009). Guidelines for foodborne disease outbreak response. Retrieved from http://

Crutchfield, S.R., & Roberts, T. (2000). Food safety efforts accelerate in the 1990s. Food Review, 23(3), 44-49.

Dalton, C.B., Haddix, A., Hoffman, R.E., & Mast, E.R. (1996). The cost of a food-borne outbreak of hepatitis A in Denver, Colorado. Archives of Internal Medicine, 156(9), 1013-1016.

Dewaal, C.S., Hicks, G., Barlow, K., Alderton, L., & Vegosen, L. (2006). Food associated with foodborne illness outbreaks from 1990 through 2003. Food Protection Trends, 26(7), 466-473.

Food and Drug Administration. (2007). Food protection plan: An integrated strategy for protecting the nation's food supply. Retrieved from FoodProtectionPlan2007/ucm132565.htm

Lynch, M.F, Tauxe, R.V., & Hedberg, C.W. (2009). The growing burden of FBDO due to contaminated fresh produce: Risks and opportunities. Epidemiology and Infection, 137(3), 307-315.

Meehan, P.J., Atkeson, T., Kepner, D.E., & Melton, M. (1992). A foodborne outbreak of gastroenteritis involving two different pathogens. American Journal of Epidemiology, 136(5), 611-616.

Public Health Security and Bioterrorism Preparedness and Response Act of 2002 [section] 107, U.S.C. [section] 3448 (2002).

Ratnapradipa, D., Quilliam, D., Wier, L., & Rhodes, D.L. (2011). Food safety education: Child-to-parent instruction in an immigrant population. Journal of Environmental Health, 73(6), 70-75.

Ruigomez, A., Garcia Rodriguez, L.A., & Panes, J. (2007). Risk of irritable bowel syndrome after an episode of bacterial gastroenteritis in general practice: Influence of comorbidities. Clinical Gastroenterology and Hepatology, 5(4), 465-469.

Scharff, R.L. (2010). Health-related costs from foodborne illness in the United States. Retrieved from Related-Foodborne-IllnessCosts-Report.pdf-1.pdf

Selman, C.A. (2010). Improving foodborne disease prevention. Journal of Environmental Health 73(2), 28-29.

Silver, L., & Bassett, M.T. (2008). Food safety for the 21st century. Journal of the American Medical Association, 300(8), 957-959.

Sivapalasingam, S., Friendman, C.R., Cohen, L., & Tauxe, R.V. (2004). Fresh produce: A growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. Journal of Food Protection, 67(10), 2342-2353.

Sobel, J., Griffin, P.M., Slutsker, L., Swerdlow, D.L., & Tauxe, R.V. (2002). Investigation of multistate foodborne disease outbreaks. Public Health Reports, 117(1), 8-19.

Taylor, E., Kastner, J., & Renter, D. (2010). Challenges involved in the Salmonella Saintpaul: Outbreak and lessons learned. Journal of Public Health Management Practice, 16(3), 221-231.

Tomato Food Safety Standards, Florida Public Chapter 2010-25, Stat. (2010).

U.S. Census Bureau. (2011). U.S. population projections. Retrieved from

Zhang, J. (2009, May 12). Hoping to make food safer, states decide to go it alone. Wall Street Journal. Retrieved from http://online.wsj. com/article/SB124208733123008901.html

Fanta W. Purayidathil, MPH

Jennifer Ibrahim, MA, MPH, PhD

College of Health Professions and Social Work

Temple University

Corresponding Author: Fanta W. Purayidathil, PhD Candidate, Temple University, 1301 Cecil B. Moore Ave., 9th Floor, Philadelphia, PA 19122. E-mail:

Summary of All Foodborne Disease Outbreak (FBDO) Outcomes,

FBDO Outcome Intra- Multi- Significance Cl (a) (95%)
 state state
 Average Average
 (Two- Lower Upper

Total 38.808 74.075 0.00 -61.739 -20.795
Total 1.5043 14.113 0.00 -17.934 -7.282
Total deaths 0.0276 0.8261 0.024 -1.490 -0.107

(a) Cl = Confidence interval.

* The Satterthwaite method was applied because Levene's
statistic found unequal variances.


Summary of Multistate (MS) Foodborne Disease Outbreak (FBDO)
Outcomes, 1998-2007

Year Total Total # MS Illnesses Hospitalizations
 FBDOs Outbreaks ([bar.x], SD) ([bar.x], SD)(a)

1998 369 8 949 (118, 127 (42.33,
 158.08) 50.89)
1999 370 10 670 (67, 19* (4.75,
 119.91) 3.77)
2000 448 13 1279 (98.38, 108 (10.8,
 145.17) 11.17)
2001 461 5 175 (35, 28 (5.6,
 10.05) 4.66)
2002 482 7 774 (110.57, 36 (12, 7)
2003 419 13 499 (38.38, 82* (6.83,
 37.43) 5.68)
2004 532 10 1002 (100.2, 196 (28,
 126.51) 45.43)
2005 402 13 569 (43.77, 84 (7,
 40.42) 4.02)
2006 621 10 979 (97.9, 248 (24.8,
 80.48) 31.9)
2007 502 18 1035 (57.5, 201 (14.36,
 89.36) 27.23)

Year Deaths p-Value
 ([bar.x], SD) (b)

1998 24 (6, --
1999 0 (0, 0) .039 (a),
 .003 (b)
2000 7 (0.88, .663 (a),
 2.47) .6 (b)
2001 2 (0.67, .686 (a),
 1.15) .9 (b)
2002 10 (2, .709 (a),
 3.39) .5 (b)
2003 4 (0.44, .033 (a),
 0.72) .3 (b)
2004 0 (0, 0) .061 (a),
 .7 (b)
2005 1 (0.1, .063 (a),
 0.31) .9 (b)
2006 6 (0.67, .080 (a),
 1.65) .6 (b)
2007 3 (0.23, .284 (a),
 0.83) .7 (b)

(a) Hospitalizations.

(b) Deaths.

* Bold and asterisk indicate a statistically significant
change in hospitalizations (a) or deaths
(b) from the previous year (p < .05).


Multistate (MS) Foodborne Disease Outbreaks (FBDOs) by Vector,

Vector Total # MS FBDOs Illnesses
 Total ([bar.x], SD)

Campylobacter 2 105 (52.5, 48.79)
Clostridium 2 12 (6, 2.82)
E. coli 30 1132 (37.7, 43.9)
Hepatitis A 2 71 (35.5, 4.94)
Listeria 5 208 (41.6, 37.4)
Norovirus 3 544 (181.3, 146.7)
Other chemical 1 11 (N/A, N/A)
Salmonella 57 4653 (81.6, 105.7)
Shigella 3 990 (330, 277.1)
Vibrio 2 200 (50, 38.18)

Vector Hospitalizations Deaths
 Total ([bar.x], SD) Total ([bar.x], SD)

Campylobacter 0 (0, N/A) 0 (0, N/A)
Clostridium 12 (6, 2.82) 1 (0.5, 0.7)
E. coli 363 (14.52, 22.03) 6 (0.3, 1.12)
Hepatitis A 18 (9, 8.48) 0 (0, N/A)
Listeria 143 (47.66, 46.85) 47 (9.25, 8.42)
Norovirus 3 (1.5, 2.12) 0 (0, N/A)
Other chemical 1 (1, N/A) 0 (0, N/A)
Salmonella 574 (13.66, 24.53) 13 (0.37, 0.84)
Shigella 13 (13, N/A) 0 (0, N/A)
Vibrio 2 (2, N/A) 0 (0, N/A)


Multistate Foodborne Disease Outbreaks, 1998-2007
(Adjusted for Population Growth)

Year Total Multistate

1998 8.00
1999 9.91
2000 12.45
2001 4.74
2002 6.57
2003 12.11
2004 9.23
2005 11.89
2006 9.06
2007 16.15

Note: Table made from line graph.


Proportion of Multistate Foodborne Disease Outbreaks, 1998-2007

Year Percentage

1998 2.70%
1999 2.16%
2000 2.90%
2001 1.00%
2002 1.50%
2003 3.10%
2004 1.90%
2005 3.20%
2006 1.60%
2007 3.60%

Note: Table made from line graph.
COPYRIGHT 2012 National Environmental Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Purayidathil, Fanta W.; Ibrahim, Jennifer
Publication:Journal of Environmental Health
Date:Nov 1, 2012
Previous Article:Crisis communication: apocalypse now or apocalypse not.
Next Article:Outbreak of cryptosporidiosis associated with a man-made Chlorinated lake--Tarrant County, Texas, 2008.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |