Food Safety: Information on Foodborne Illnesses (Letter Report, 05/08/96,
GAO/RCED-96-96).
GAO reviewed the extent of foodborne illnesses caused by microbal
contamination, focusing on: (1) the frequency, health consequences, and
economic impacts of these illnesses; and (2) the extent of information
available to develop effective control strategies.
GAO found that: (1) between 6.5 million and 81 million cases of
foodborne illness and as many as 9,100 related deaths occur each year;
(2) the risk of foodborne illness is increasing due to changes in food
supply and consumption, recognition of new causes of foodborne
illnesses, new modes of transmission, increased resistance to
long-standing food-processing and storage techniques, and emerging
virulent strains of well-known bacteria; (3) while foodborne illnesses
are most often brief and do not require medical care, a small percentage
cause long-term disability or even death; (4) foodborne illness may cost
billions of dollars every year in medical costs and lost productivity;
(5) the current voluntary reporting system does not provide sufficient
data on the prevalence and sources of foodborne illnesses; (6) efforts
are under way to collect more and better data on the prevalence and
sources of foodborne illnesses; and (7) more uniform and comprehensive
data on the number and causes of foodborne illnesses could lead to more
effective control strategies.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: RCED-96-96
TITLE: Food Safety: Information on Foodborne Illnesses
DATE: 05/08/96
SUBJECT: Contaminated foods
Health hazards
Safety regulation
Safety standards
Food industry
Consumer protection
Infectious diseases
Public health research
Biomedical research
IDENTIFIER: E. coli Bacteria
Salmonella Enteritidis Bacteria
Listeria monocytogenes Bacteria
Campylobacter jejuni/coli Bacteria
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to Congressional Committees
May 1996
FOOD SAFETY - INFORMATION ON
FOODBORNE ILLNESSES
GAO/RCED-96-96
Information of Foodborne Illnesses
(150639)
Abbreviations
=============================================================== ABBREV
CDC - Centers for Disease Control and Prevention
ERS - Economic Research Service
FDA - Food and Drug Administration
FSIS - Food Safety and Inspection Service
HUS - Hemolytic Uremic Syndrome
USDA - U.S. Department of Agriculture
Letter
=============================================================== LETTER
B-270753
May 8, 1996
Congressional Committees
Over the past decade, a number of widely reported outbreaks of
foodborne illnesses caused by microbial contamination--the most
commonly identified cause of such outbreaks--have heightened the
public's concern about the safety of food. While such outbreaks have
brought the issue to the public's attention, public health and food
safety officials agree that the largest number of foodborne illnesses
occur as isolated cases, not as part of a publicized outbreak.
Given the scale of the U.S. food production and distribution system
and the lack of control over how consumers handle food after they
purchase it, there is little likelihood that foodborne illnesses can
be completely eliminated. The most that can be hoped for is that all
reasonable steps are taken to ensure that the problem is minimized.
In this context, we have attempted in this report to provide a
factual picture on the extent of foodborne illness in the United
States today. Specifically, we obtained the best available
information on the (1) frequency of foodborne illness, (2) health
consequences and economic impacts of foodborne illnesses, and (3)
adequacy of knowledge about foodborne illnesses to develop effective
control strategies.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
Data on the extent of foodborne illnesses and related deaths are
incomplete and may understate the extent of the problem because most
cases go unreported. However, according to the best available
estimates by the Centers for Disease Control and Prevention and other
public health and food safety experts, millions of illnesses and
thousands of deaths each year in the United States can be traced to
contaminated food. Moreover, public health officials believe that
the risk of foodborne illnesses has been increasing over the last 20
years.
Although foodborne illnesses generally cause temporary disorders of
the digestive tract, they can also lead to more serious consequences.
While the precise cost of foodborne illnesses is unknown, recent
estimates range from over $5 billion to more than $22 billion
annually. According to U.S. Department of Agriculture estimates,
the cost of medical treatment and lost productivity related to
foodborne illnesses from seven of the most harmful bacteria may have
been as much as $9.4 billion during 1993.
Public health and food safety officials believe that current data on
foodborne illnesses do not provide a complete picture of the risk
level and do not depict the sources of contamination and the
populations most at risk in sufficient detail. More uniform and
comprehensive data on the number and causes of foodborne illnesses
could form the basis of more effective control strategies. In 1995,
federal and state agencies began steps to collect such data in five
areas across the country. Federal officials expressed some concern
about whether, in this era of budget constraints, they would be able
to continue funding this discretionary effort for the number of years
needed to collect meaningful trend data.
BACKGROUND
------------------------------------------------------------ Letter :2
The Centers for Disease Control and Prevention (CDC) is the federal
agency primarily responsible for monitoring the incidence of
foodborne illness in the United States. In collaboration with state
and local health departments and other federal agencies, CDC
investigates outbreaks of foodborne illnesses and supports disease
surveillance, research, prevention efforts, and training related to
foodborne illnesses. CDC coordinates its activities concerning the
safety of the food supply with the Food and Drug Administration (FDA)
in the Department of Health and Human Services and those concerning
the safety of meat, poultry, and eggs with the Food Safety and
Inspection Service (FSIS) in the U.S. Department of Agriculture
(USDA). FDA and FSIS, which are the primary federal agencies
responsible for overseeing the safety of the food supply, maintain
liaison with CDC in Atlanta, Georgia.
CDC monitors individual cases of illness from harmful bacteria,
viruses, chemicals, and parasites (hereafter referred to collectively
as pathogens) that are known to be transmitted by foods, as well as
foodborne outbreaks,\1 through reports from state and local health
departments, FDA, and FSIS. CDC does not have the authority to
require states to report data on foodborne illnesses. In practice,
each state determines which diseases it will routinely report to CDC.
In addition, state laboratories voluntarily report the number of
positive test results for several diseases that CDC has chosen to
monitor. However, these reports do not identify the source of
infection and are not limited to cases of foodborne illness. CDC
also investigates a limited number of more severe or unusual
outbreaks when state authorities request assistance. (For a
description of the data that CDC relies on to monitor foodborne
illnesses, see app. I.)
At least 30 pathogens are associated with foodborne illnesses. For
reporting purposes, CDC categorizes the causes of outbreaks of
foodborne illnesses as bacterial, chemical, viral, parasitic, or
unknown pathogens. (See app. II. for information on these
pathogens and the illnesses they cause.) Although many people
associate foodborne illnesses primarily with meat, poultry, eggs, and
seafood products, many other foods, including milk, cheese, ice
cream, orange and apple juices, cantaloupes, and vegetables, have
also been involved in outbreaks during the last decade.
Bacterial pathogens are the most commonly identified cause of
outbreaks of foodborne illnesses. Bacterial pathogens can be easily
transmitted and can multiply rapidly in food, making them difficult
to control. CDC has targeted four of them--E. coli O157:H7,
Salmonella Enteritidis, Listeria monocytogenes, and Campylobacter
jejuni--as those of greatest concern. (See app. III.) CDC is also
concerned about other bacterial pathogens, such as Vibrio vulnificus
and Yersinia enterocolitica, which can cause serious illnesses, and
Clostridium perfringens and Staphylococcus aureus, which cause less
serious illnesses but are very common. The chemical causes of
foodborne illnesses are primarily natural toxins that occur in fish
or other foods but also include heavy metals, such as copper and
cadmium. Viral pathogens are often transmitted by infected food
handlers or through contact with sewage. Only a few viral pathogens,
such as the Hepatitis A and Norwalk viruses, have been proven to
cause foodborne illnesses. Finally, parasitic pathogens, such as
Trichinella--found in undercooked or raw pork--multiply only in host
animals, not in food. CDC officials believe that viral and parasitic
pathogens are less likely than bacterial pathogens to be identified
as the source of an outbreak of foodborne illness because their
presence is more difficult to detect.
--------------------
\1 For most pathogens, CDC defines an outbreak of foodborne illness
as two or more persons experiencing a similar illness for which a
common food was implicated.
FOODBORNE ILLNESSES ARE
BELIEVED TO BE A GROWING
PROBLEM
------------------------------------------------------------ Letter :3
The existing data on the extent of foodborne illnesses have
weaknesses and may not fully depict the extent of the problem.
Public health experts believe that the majority of cases of foodborne
illness are not reported because the initial symptoms of most
foodborne illnesses are not severe enough to warrant medical
attention, the medical facility or state does not report such cases,
or the illness is not recognized as foodborne. However, according to
the best available estimates, based largely on CDC's data, millions
of people become sick from contaminated food each year, and several
thousand die. In addition, public health and food safety officials
believe that the risk of foodborne illnesses is increasing for
several reasons. For example, as a result of large-scale food
production and broad distribution of products, those products that
may be contaminated can reach a great number of people in many
locations. Furthermore, new and more virulent strains of previously
identified harmful bacteria have been identified in the past several
decades. Also, mishandling or improper preparation can further
increase the risk.
AVAILABLE ESTIMATES SHOW
THAT MILLIONS ARE AFFECTED
BY FOODBORNE ILLNESSES
---------------------------------------------------------- Letter :3.1
Between 6.5 million and 81 million cases of foodborne illness and as
many as 9,100 related deaths occur each year, according to the
estimates provided by several studies conducted over the past 10
years. Table 1 shows the range of estimates from four studies cited
by food safety experts as among the best available estimates on the
subject. The table also identifies the data on which these estimates
are based. While various foods have been implicated as vehicles for
pathogens in foodborne illnesses and related deaths, the available
data do not allow a precise breakdown by specific foods. In general,
animal foods--beef, pork, poultry, seafood, milk, and eggs--are more
frequently identified as the source of outbreaks in the United States
than non-animal foods. USDA, which regulates meat and poultry
products, has estimated that over half of all foodborne illnesses and
deaths are caused by contaminated meat and poultry products.
Table 1
Estimates of Foodborne Illnesses and
Related Deaths
Study or report/ Estimated number of Estimated number of
authors (year) illnesses per year deaths per year Basis for estimates
---------------------- -------------------- -------------------- --------------------
The Incidence and Cost 24 million to 81 No estimate provided National survey of
of Foodborne Diarrheal million or more physicians conducted
Disease in by National Center
the United States, for Health
Archer and Kvenberg Statistics (1977-
(1985)\a 78), supplemented
with 1983 data on
illnesses from
specific pathogens
Closing the Gap: The 6.5 million annually 9,100 Published and survey
Burden of Unnecessary data from National
Illness, Bennett Center for Health
et. al. (1987)\b Statistics, Center
for Infectious
Diseases, and the
Center for
Prevention Services
Preliminary Estimates 12.6 million 523 Median of four
of Costs of Foodborne estimates based on
Disease in the United (1) CDC outbreak
States, Todd (1989)\c data; (2) 1987
Bennett study data;
(3) Salmonella
underreporting; and
(4) Canadian disease
rates extrapolated
to U.S. population
Foodborne Pathogens: 6.5 million to 33 Up to 9,000 Review of past
Risks and million studies
Consequences, Council
for Agricultural
Science and Technology
(1994)\d
----------------------------------------------------------------------------------------
\a Journal of Food Protection, Vol. 48, No. 10 (Oct. 1985), pp.
887-894.
\b New York: Oxford University Press.
\c Journal of Food Protection, Vol. 52, No. 8 (Aug. 1989), pp.
586-601.
\d No. 122, Sept. 1994. Ames, Iowa: Council for Agricultural
Science and Technology.
The wide range in the estimated number of foodborne illnesses and
related deaths is due primarily to the considerable uncertainty about
the number of cases that are never reported to CDC and the
methodology used to make the estimate. Public health and food safety
officials believe that many of these illnesses are not reported
because the episodes are mild and do not require medical treatment.
For example, CDC officials believe that many intestinal illnesses
that are commonly referred to as the stomach flu are caused by
foodborne pathogens. According to these officials, people do not
usually associate these illnesses with food because the onset of
symptoms occurs 2 or more days after the contaminated food was eaten.
In other cases, a foodborne illness may contribute to the death of an
already ill person. In these cases, a foodborne illness may not be
reported as the cause of death. In the absence of more complete
reporting, researchers can only broadly estimate the number of
illnesses and related deaths.
Furthermore, most physicians and health professionals treat patients
who have diarrhea without ever identifying the specific cause of the
illness. In severe or persistent cases, a laboratory test may be
ordered to identify the responsible pathogen. However, some
laboratories may not have the ability to identify a given pathogen.
Finally, physicians may not associate the symptoms they observe with
a pathogen that they are required to report to the state or local
health authorities. For example, a CDC official cited a Nevada
outbreak in which no illnesses from E. coli O157:H7 had been
reported to health officials, despite a requirement that physicians
report such cases to the state health department.\2 Nevertheless, 58
illnesses from this outbreak were identified after public service
announcements alerted the public and health professionals that
contaminated hamburger had been shipped to restaurants in a specific
area of the state.
--------------------
\2 These cases were part of the 1993 outbreak in the western United
States.
EXPERTS BELIEVE THAT THE
RISK OF FOODBORNE ILLNESSES
IS INCREASING
---------------------------------------------------------- Letter :3.2
Food safety and public health officials believe that the risk of
foodborne illnesses is increasing. Several factors contribute to
this increased risk. First, the food supply is changing in ways that
can promote foodborne illnesses. For example, as a result of modern
animal husbandry techniques, such as crowding a large number of
animals together, the pathogens that can cause foodborne illnesses in
humans can spread throughout the herd. Because of broad
distribution, contaminated products can reach individuals in more
locations. Mishandling of food can also lead to contamination. For
example, leaving perishable foods at room temperature increases the
likelihood of bacterial growth, and improper preparation, such as
undercooking, reduces the likelihood that bacteria will be killed and
can further increase the risk of illness. There are no comprehensive
data to explain at what point pathogens are introduced into foods.
Knowledgeable experts believe that although illnesses and deaths
often result after improper handling and preparation, the pathogens
were, in many cases, already present at the processing stage.
Furthermore, the pathogens found on meat and poultry products may
have arrived on the live animals.
Second, because of demographic changes, more people are at greater
risk of contracting a foodborne illness. Certain populations are at
greater risk for these illnesses: people with suppressed immune
systems, children, and the elderly. In addition, children are more
at risk because group settings, such as day care centers, increase
the likelihood of person-to-person transmission of pathogens. The
number of children in these settings is increasing, as is the number
in other high-risk groups, according to CDC.
Third, three of the four pathogens CDC considers the most important
were unrecognized as causes of foodborne illness 20 years
ago--Campylobacter, Listeria, and E. coli O157:H7.
Fourth, bacteria already recognized as sources of foodborne illnesses
have found new modes of transmission. While many illnesses from E.
coli O157:H7 occur from eating insufficiently cooked hamburger, these
bacteria have also been found more recently in other foods, such as
salami, raw milk, apple cider, and lettuce. Other bacteria
associated with contaminated meat and poultry, such as Salmonella,
have also been found in foods that the public does not usually
consider to be a potential source of illness, such as ice cream,
tomatoes, melons, alfalfa sprouts, and orange juice.
Fifth, some pathogens are far more resistant than expected to
long-standing food-processing and storage techniques previously
believed to provide some protection against the growth of bacteria.
For example, some bacterial pathogens, such as Yersinia and Listeria,
can continue to grow in food under refrigeration.
Finally, according to CDC officials, virulent strains of well-known
bacteria have continued to emerge. For example, one such pathogen,
E. coli O104:H21, is another potentially deadly strain of E. coli.
In 1994, CDC found this new strain in milk from a Montana dairy.
FOODBORNE ILLNESSES CAN BE
DEBILITATING AND COSTLY
------------------------------------------------------------ Letter :4
While foodborne illnesses are often temporary, they can also result
in more serious illnesses requiring hospitalization, long-term
disability, and death. Although the overall cost of foodborne
illnesses is not known, two recent estimates place some of the costs
in the range of $5.6 billion to more than $22 billion per year. The
first estimate, covering only the portion related to the medical
costs and productivity losses of seven specific pathogens, places the
costs in the range of $5.6 billion to $9.4 billion. The second,
covering only the value of avoiding deaths from five specific
pathogens, places the costs in the range of $6.6 billion to $22
billion.
PATHOGENS CAUSE DISABLING
HEALTH EFFECTS IN SOME
PEOPLE
---------------------------------------------------------- Letter :4.1
While foodborne illnesses are often brief and do not require medical
treatment, they can also result in more serious illnesses and death.
In a small percentage of cases, foodborne infections spread through
the bloodstream to other organs, resulting in serious long-term
disability or even death. Serious complications can also result when
diarrhetic infections resulting from foodborne pathogens act as a
triggering mechanism in susceptible individuals, causing an illness
such as reactive arthritis to flare up. In other cases, no immediate
symptoms may appear, but serious consequences may eventually develop.
The likelihood of serious complications is unknown, but some experts
estimate that about 2 to 3 percent of all cases of foodborne illness
lead to serious consequences. For example:
-- E. coli O157:H7 can cause kidney failure in young children and
infants and is most commonly transmitted to humans through the
consumption of undercooked ground beef. The largest reported
outbreak in North America occurred in 1993 and affected over 700
people, including many children who ate undercooked hamburgers
at a fast food restaurant chain. Fifty-five patients, including
four children who died, developed a severe disease, Hemolytic
Uremic Syndrome, which is characterized by kidney failure.
-- Salmonella can lead to reactive arthritis, serious infections,
and deaths. In recent years, outbreaks have been caused by the
consumption of many different foods of animal origin, including
beef, poultry, eggs, milk and dairy products, and pork. The
largest outbreak, occurring in the Chicago area in 1985,
involved over 16,000 laboratory-confirmed cases and an estimated
200,000 total cases. Some of these cases resulted in reactive
arthritis. For example, one institution that treated 565
patients from this outbreak confirmed that 13 patients had
developed reactive arthritis after consuming contaminated milk.
In addition, 14 deaths may have been associated with this
outbreak.
-- Listeria can cause meningitis and stillbirths and has a fatality
rate of 20 to 40 percent.\3 All foods may contain these
bacteria, particularly poultry and dairy products. Illnesses
from this pathogen occur mostly in single cases rather than in
outbreaks. The largest outbreak in North America occurred in
1985 in Los Angeles, largely in pregnant women and their
fetuses. More than 140 cases of illness were reported,
including at least 13 cases of meningitis. At least 48 deaths,
including 20 stillbirths or miscarriages, were attributed to the
outbreak. Soft cheese produced in a contaminated factory
environment was confirmed as the source.
-- Campylobacter may be the most common precipitating factor for
Guillain-Barre syndrome, which is now one of the leading cause
of paralysis from disease in the United States. Campylobacter
infections occur in all age groups, with the greatest incidence
in children under 1 year of age. The vast majority of cases
occur individually, primarily from poultry, not during
outbreaks. Researchers estimate that 4,250 cases of
Guillain-Barre syndrome occur each year and that about 425 to
1,275 of these cases are preceded by Campylobacter infections.
--------------------
\3 Fatalities as a percentage of all cases of illness.
FOODBORNE ILLNESSES CAN COST
BILLIONS OF DOLLARS IN
MEDICAL EXPENSES AND LOST
PRODUCTIVITY ANNUALLY
---------------------------------------------------------- Letter :4.2
While the overall annual cost of foodborne illnesses is unknown, the
studies we reviewed estimate that it is in the billions of dollars.
The range of estimates among the studies is wide, however,
principally because of uncertainty about the number of cases of
foodborne illness and related deaths. (See app. IV.) Other
differences stem from the differences in the analytical approach used
to prepare the estimate. Some economists attempt to estimate the
costs related to medical treatment and lost wages (the
cost-of-illness method); others attempt to estimate the value of
reducing the incidence of illness or loss of life (the
willingness-to-pay method). Two recent estimates demonstrate these
differences in analytical approach.
In the first, USDA's Economic Research Service (ERS) used the
cost-of-illness approach to estimate that the 1993 medical costs and
losses in productivity resulting from seven major foodborne pathogens
ranged between $5.6 billion and $9.4 billion.\4 Of these costs, $2.3
billion to $4.3 billion were the estimated medical costs for the
treatment of acute and chronic illnesses, and $3.3 billion to $5.1
billion were the productivity losses from the long-term effects of
foodborne illnesses. Medical expenses ranged from more modest
expenses for routine doctors' visits and laboratory tests to more
substantial expenses for hospital rooms and kidney transplants.
Productivity losses included expenses such as lost wages from
long-term disabilities and deaths caused by foodborne illnesses.
Table 2 provides information on the costs associated with each of the
seven pathogens.
Table 2
ERS' Estimated Medical Costs and
Productivity Losses From Illnesses
Caused by Seven Foodborne Pathogens
During 1993
(Dollars in billions)
Productivi Total
Medical ty costs and
Pathogens costs losses\a losses
---------------------------------- ---------- ---------- ----------
Campylobacter jejuni or coli $0.6-0.8 $0.0\b- $0.6-1.0
0.2
Clostridium perfringens 0.1\c Not 0.1
available
E. coli O157:H7 0.0\a-0.1 0.2-0.5 0.2-0.6
Listeria monocytogenes 0.1 0.1-0.2 0.2-0.3
Salmonella 0.4-2.0 0.2-1.5 0.6-3.5
Staphylococcus aureus 1.2\c Not 1.2
available
Toxoplasma gondii 0.0\b 2.7 2.7
======================================================================
Total $2.3\d- $3.3\d- $5.6-9.4
4.3 5.1
----------------------------------------------------------------------
\a The productivity losses for E. coli O157:H7, Listeria
monocytogenes, and Toxoplasma gondii included costs related to
long-term disabilities and deaths. Productivity losses for
Campylobacter and Salmonella were calculated for deaths only.
\b These amounts are less than $50 million.
\c Based on average costs and included some productivity losses in
addition to medical costs.
\d Does not add due to rounding.
Source: ERS and FSIS.
CDC, FDA, and ERS economists stated that these estimates may be low
for several reasons. First, the cost-of-illness approach generates
low values for reducing health risks to children and the elderly
because these groups have low earnings and hence low productivity
losses. Second, this approach does not recognize the value that
individuals may place on (and pay for) feeling healthy, avoiding
pain, or using their free time. In addition, not all of the 30
pathogens associated with foodborne illnesses were included.
In the second analysis,\5 ERS used the willingness-to-pay method\6 to
estimate the value of preventing deaths for five of the seven major
pathogens (included in the first analysis) at $6.6 billion to $22.0
billion in 1992. The estimate's range reflected the range in the
estimated number of deaths, 1,646 to 3,144, and the range in the
estimated value of preventing a death, $4 million to $7 million.
Although these estimated values were higher than those resulting from
the first approach, they may have also understated the economic cost
of foodborne illnesses because they did not include an estimate of
the value of preventing nonfatal illnesses and included only five of
the seven major pathogens included in the first analysis.
--------------------
\4 "Pathogen Reduction, Hazard Analysis and Critical Control Point
(HACCP) Systems," Federal Register (Feb. 3, 1995).
\5 Aldrich, Lorna. "Food-Safety Policy: Balancing Risk and Costs,"
Food Review. USDA, Economic Research Service, Vol. 17, Issue 2,
May-Aug. 1994, pp. 10-11.
\6 ERS also refers to the willingness-to-pay approach as the value of
reducing the risk of illness or death.
BETTER DATA COULD LEAD TO MORE
EFFECTIVE CONTROL STRATEGIES
------------------------------------------------------------ Letter :5
While current data indicate that the risk of foodborne illnesses is
significant, public health and food safety officials believe that
these data do not identify the level of risk, the sources of
contamination, and the populations most at risk in sufficient detail.
More uniform and comprehensive data on the number and causes of
foodborne illnesses could form the basis of more effective control
strategies. Beginning in 1995, federal and state agencies took steps
to collect such data in five areas across the country. While this
effort will provide additional data, CDC officials believe that
collecting data at more locations and for other pathogens would
provide even more representative data and identify more causes of
foodborne illnesses.
CURRENT DATA DO NOT PROVIDE
SUFFICIENT DETAIL ON THE
RISK POSED BY FOODBORNE
ILLNESSES
---------------------------------------------------------- Letter :5.1
According to public health and food safety officials, the current
voluntary reporting system does not provide sufficient data on the
prevalence and sources of foodborne illnesses. There are no specific
national requirements for reporting on foodborne pathogens.
According to CDC, states do not (1) report on all pathogens of
concern, (2) usually identify whether food was the source of the
illness, or (3) identify many of the outbreaks or individual cases of
foodborne illness that occur.
Consequently, according to CDC, FDA, and FSIS, public health
officials cannot precisely determine the level of risk from known
pathogens or be certain that they can detect the existence and spread
of new pathogens in a timely manner. They also cannot identify all
factors that put the public at risk or all types of food or
situations in which microbial contamination is likely to occur.
Finally, without better data, regulators cannot assess the
effectiveness of their efforts to control the level of pathogens in
food.
According to public health and food safety officials, a better system
for monitoring the extent of foodborne illnesses would actively seek
out specific cases. Such a system would require outreach to
physicians and clinical laboratories. CDC demonstrated the
effectiveness of such an outreach effort when it conducted a
long-term study, initiated in 1986, to determine the number of cases
of illness caused by Listeria. This study showed that a lower rate
of illness caused by Listeria occurred between 1989 and 1993 during
the implementation of food safety programs designed to reduce the
prevalence of Listeria in food.
EFFORTS TO IMPROVE
INFORMATION ON FOODBORNE
ILLNESSES ARE UNDER WAY
---------------------------------------------------------- Letter :5.2
In July 1995, CDC, FDA, and FSIS began a comprehensive effort to
track the major bacterial pathogens that cause foodborne illnesses.
These agencies are collaborating with state health departments in
five areas across the country to better determine the incidence of
infection with Salmonella and E. coli O157:H7 and other foodborne
bacteria and to identify these sources of diarrheal illness from
Salmonella and E. coli O157:H7.\7 Initially, FDA provided $378,000
and FSIS provided $500,000 through CDC to the five locations for 6
months. The agencies believe that this effort should be a permanent
part of a sound public health system. For fiscal year 1996, FSIS is
providing $1 million and FDA is providing $300,000. CDC provides
overall management and coordination and facilitates the development
of technical expertise at the sites through its established
relationships with the state health departments.
The project consists of three parts:
-- a survey of the local population in the five locations and
interviews with local health professionals to estimate the
number of diarrheal illnesses and determine the number of
illnesses for which medical attention was sought and laboratory
samples were taken;
-- a survey of laboratories to determine the microbiological
testing procedures and processes used to identify foodborne
illnesses and an audit of the participating laboratories' test
results to determine what proportion of cases were detected; and
-- statistical studies to determine, among other things, the risks
associated with different foods.
CDC and the five sites will use the information to identify emerging
foodborne pathogens and monitor the incidence of foodborne illness.
FSIS will use the data to evaluate the effectiveness of new food
safety programs and regulations to reduce foodborne pathogens in meat
and poultry and assist in future program development. FDA will use
the data to evaluate its efforts to reduce foodborne pathogens in
seafood, dairy products, fruit, and vegetables.
According to CDC, FDA, and FSIS officials, such projects must collect
data over a number of years to identify national trends and evaluate
the effectiveness of strategies to control pathogens in food.
Funding was decreased slightly for this project in 1996, and these
officials are concerned about the continuing availability of funding,
in this era of budget constraints, to conduct this discretionary
effort over the longer term.
--------------------
\7 The areas are (1) the greater metropolitan area of Atlanta, (2) an
area comprised by two northern California counties, (3) an area
comprised by two Connecticut counties, (4) the state of Minnesota,
and (5) the state of Oregon.
AGENCY COMMENTS
------------------------------------------------------------ Letter :6
We provided copies of a draft of this report to CDC, FSIS, and FDA
for their review and comment. We met with the Director, Division of
Bacterial and Mycotic Diseases, CDC; the Associate Administrator,
FSIS; and other relevant officials from both agencies. These
officials generally agreed with the information discussed and
provided some clarifying comments that we incorporated into the
report. FDA's Office of Legislative Affairs notified us that FDA
generally agreed with the contents of the report and provided several
technical comments that we incorporated.
SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :7
To conduct this review, we spoke with, and obtained studies, data,
and other information on foodborne illnesses from, officials at CDC,
ERS, FDA, and FSIS. We met with these officials at their
headquarters in Atlanta, Georgia, and Washington, D.C. To examine
the frequency of foodborne illness, we met with agency officials to
identify and discuss the most widely recognized studies on the
incidence of foodborne illness in the United States and obtained
documentation. To examine the health consequences of foodborne
illnesses, we relied primarily on discussions with medical experts at
CDC and articles that have appeared in professional journals obtained
from CDC officials and our literature review. To examine the
economic impacts of foodborne illnesses, we reviewed the analytical
approaches used to estimate the costs of foodborne illnesses and
recent examples of such estimates and spoke with economists at CDC,
ERS, and FDA. To examine the adequacy of knowledge about foodborne
illnesses to develop effective control strategies, we spoke with the
project managers from CDC, FDA, and FSIS and other agency officials
associated with a joint effort with five state health departments
recently undertaken to improve their knowledge about foodborne
illnesses and collected agency documents.
We reviewed but did not independently verify the accuracy of the data
available on the number of reported cases of foodborne illness, the
overall estimates of incidence, or the estimates of costs from
specific pathogens because this effort would have required the
verification of multiple databases and other information from state
and federal agencies and other sources. This verification process
would have required a large commitment of additional resources. We
did not review data on the incidence of foodborne illness in other
countries because comparable data were not readily available and the
data that are available have some of the same limitations as the data
on U.S. foodborne illnesses.
We conducted our review from June 1995 through April 1996 in
accordance with generally accepted government auditing standards.
---------------------------------------------------------- Letter :7.1
We are sending this report to you because of your role in overseeing
the activities and funding of the agencies responsible for the issues
discussed. If you or your staff have any questions about this
report, I can be reached at (202) 512-5138. Major contributors to
this report are listed in appendix V.
Robert A. Robinson
Director, Food and
Agriculture Issues Area
List of Congressional Committees
The Honorable Richard G. Lugar
Chairman
The Honorable Patrick J. Leahy
Ranking Minority Member
Committee on Agriculture, Nutrition,
and Forestry
United States Senate
The Honorable Mark O. Hatfield
Chairman
The Honorable Robert C. Byrd
Ranking Minority Member
Committee on Appropriations
United States Senate
The Honorable Pat Roberts
Chairman
The Honorable E (Kika) de la Garza
Ranking Minority Member
Committee on Agriculture
House of Representatives
The Honorable Bob Livingston
Chairman
The Honorable David R. Obey
Ranking Minority Member
Committee on Appropriations
House of Representatives
SOURCES OF DATA ON ILLNESSES AND
DEATHS FROM FOODBORNE PATHOGENS
=========================================================== Appendix I
To monitor, control, and prevent foodborne illnesses, the Centers for
Disease Control and Prevention (CDC) relies primarily on four types
of data from local and state health departments, according to CDC
officials. These four types of data are shown in table I.1.
Table I.1
Data CDC Relies on to Monitor, Control,
and Prevent Foodborne Illnesses
Description of Frequency of
Type data collection Purpose Limitations
---------------- ---------------- ---------------- ---------------- ----------------
Outbreak data Reports of Reported Identify new Do not include
foodborne annually for trends in all outbreaks or
outbreaks from most outbreaks illnesses and isolated cases
local and state (more frequently modes and for most
health for outbreaks of vehicles of pathogens;
departments E. coli O157:H7 transmission of often, the
and Salmonella pathogens pathogen and the
Enteritidis) route of
infection are
not identified;
delay in
compiling data
Laboratory data Computer-based Usually weekly, Analysis of Can only
reporting on but varies; can trends and rapid identify cases
positive test be daily or more detection of for which tests
results for four frequently outbreaks were performed;
pathogens from if needed limited to four
local and state pathogens; do
laboratories not identify
source of
infection
Outbreak Data CDC As requested Provide specific Only account for
investigations collects when it information on a portion of the
provides the pathogen and total number of
assistance at source of outbreaks
the request of infection
local and state
health
authorities
Active Comprehensive Ongoing Estimate the Until recently,
surveillance data collected incidence of few networks had
data from from a network illness and been established
laboratories, of determine the for collecting
physicians, and representative source of data on
patients sites, illness and the foodborne
laboratories, populations illnesses
physicians, and affected because such
patients networks require
substantial
resources
----------------------------------------------------------------------------------------
Source: Officials at the National Center for Infectious Diseases,
CDC.
As table I.1 notes, each type of data has limitations, particularly
the outbreak and laboratory data, which have been CDC's primary
monitoring tools. More specifically, in about half of the outbreaks
as shown in figure I.1, the data do not identify the agent that
caused the outbreak. Furthermore, these data generally do not
provide information about the cause of a new trend. One or more
factors can account for a new trend: a change in consumption
behavior, such as a preference for turkey over red meat; a reporting
bias, such as an increase in the number of laboratories testing for
the disease; or a change in the nature of the disease, such as the
emergence of a new strain. Finally, there is a delay from the time
these data are reported to CDC until they are compiled into annual
summaries. At the time of our review, complete annual summaries of
data were only available through 1991.
Figure I.1: Reported Cases of
Illness From Foodborne
Outbreaks by Cause, 1988-91
(See figure in printed
edition.)
Note: Other causes include chemicals, viruses, and parasites.
Source: CDC.
Furthermore, CDC's laboratory data, from its Public Health Laboratory
Information System, represent only a fraction of the cases of
illnesses that occur from four pathogens that CDC tracks.\8 For
example, only one confirmed case of infection was cited in the
laboratory data that the Georgia Health Department reported to CDC
during an outbreak caused by contaminated ice cream products in 1994.
However, on the basis of a survey of home delivery customers that it
conducted, CDC estimated that 11,404 cases occurred in Georgia alone
(products were distributed in 48 states). Finally, these data do not
include information about the source of the illness.
In addition to its program activities to monitor, control, and
prevent foodborne illnesses, CDC collects national data on a range of
pathogens and illnesses from a variety of data sources. These
sources include the National Notifiable Diseases Surveillance
System,\9 the National Hospital Discharge Survey, the National
Ambulatory Medical Care Survey, the National Health Interview Survey,
and the National Vital Statistics System. Researchers use these data
to estimate the number of foodborne illnesses, their severity, and
their costs. But these data have major limitations for understanding
foodborne illnesses, primarily because they rarely identify the
specific pathogen or indicate the method of transmission. For
example, illnesses, such as those caused by E.coli O157:H7,\10 cannot
always be distinguished from other similar illnesses. Researchers
may supplement national data with data from health maintenance
organizations or community health studies. Such studies provide more
detailed information about foodborne illnesses but are limited to
small samples and have only been done occasionally.
--------------------
\8 The four pathogens are Salmonella, Shigella, Campylobacter, and E.
coli O157:H7. At least 30 pathogens are associated with foodborne
illnesses.
\9 Data from this system are the basis of CDC's publications on the
diseases reported to local and state public health officials each
week.
\10 E. coli O157:H7 is the strain of bacteria responsible for the
1993 outbreak from undercooked hamburgers at a chain of restaurants
in the western United States.
ILLNESSES FROM FOODBORNE PATHOGENS
AND CHEMICALS
========================================================== Appendix II
Although foodborne illnesses are often short term and do not require
medical treatment, in some cases, these illnesses can involve other
organs, resulting in serious complications. In other cases,
foodborne illnesses may not result in immediate symptoms but
ultimately may produce serious health problems.
CDC has classified the causes of foodborne illnesses into the
following four categories:
-- Bacterial pathogens are microorganisms that can be seen with a
microscope but not with the naked eye. Some bacterial pathogens
are infectious themselves or can produce toxins. Furthermore,
bacteria can multiply rapidly in food, making them difficult to
control and can be transmitted through person-to-person contact.
Some bacteria, such as Clostridium botulinum, which causes
botulism, can form spores in food that can resist some food
preservation treatments, including boiling.
-- Chemical agents are primarily naturally occurring toxins that
can enter the food supply. Paralytic shellfish poisoning and
mushroom poisoning are caused by such chemicals. Heavy
metals--such as cadmium, copper, iron, tin, and zinc--are also
included in this category. These pathogens can cause a variety
of gastrointestinal, neurologic, respiratory, and other
symptoms.
-- Viral pathogens are too small to be seen with a conventional
microscope. Only a few viral pathogens, such as the Hepatitis A
and Norwalk viruses, have been proven to cause foodborne
illnesses. Viral pathogens are often transmitted by infected
food handlers or through contact with sewage.
-- Parasitic pathogens are larger than bacterial pathogens and
include protozoa (one-celled microorganisms) and multicelled
parasites. They multiply only in host animals, not in food.
Protozoa form cysts that are similar to spores but less
resistant to heat. Cysts can be transmitted to new hosts
through food that has been eaten. Multicelled parasites, such
as Trichinella spiralis, which causes trichinosis, occur in
microscopic forms, such as eggs and larvae. Thorough cooking
will destroy larvae.
While the likelihood of serious complications from foodborne
illnesses is unknown, some researchers estimate that about 2 to 3
percent of all cases of foodborne illness lead to serious
consequences. Although anyone can suffer from foodborne illnesses,
certain populations are more at risk from them or their complications
than others: pregnant women, children, those with compromised or
suppressed immune systems, and the elderly. These groups are more at
risk because of altered, underdeveloped, damaged, or weakened immune
systems.
Table II.1 provides information on several foodborne pathogens, the
serious complications they may result in, and some of the foods in
which they have been found.
Table II.1
Some Foodborne Pathogens That Can Cause
Serious Illnesses
Serious illnesses that can Foods in which pathogens
Foodborne pathogen result have been found
---------------------------- ---------------------------- ----------------------------
Bacteria
Campylobacter Arthritis, blood poisoning, Poultry, raw milk, and meat
Guillain-Barre syndrome
(paralysis), chronic
diarrhea, meningitis, and
inflammation of the heart,
gallbladder, colon, and
pancreas
E. coli O157:H7 HUS,\a which is associated Meat, especially ground
with kidney failure and beef; raw milk; and produce
neurologic disorders; and
other illnesses
Listeria Meningitis, blood poisoning, Soft cheese, other dairy
stillbirths, products, meat, poultry,
and other disorders seafood, fruits, and
vegetables
Salmonella Reactive arthritis, blood Poultry, meat, eggs, dairy
poisoning, products, seafood, fruits,
Reiter's disease and vegetables
(inflammation of joints,
eye membranes, and urinary
tract) and inflammation of
the pancreas, spleen, colon,
gallbladder, thyroid, and
heart
Shigella Reiter's disease, HUS, Salads, milk and dairy
pneumonia, blood poisoning, products, and produce
neurologic disorders, and
inflammation of the spleen
Vibrio vulnificus Blood poisoning Seafood
Yersinia enterocolitica Reiter's disease, pneumonia, Pork and dairy products
and inflammation of
vertebrae, lymphatic glands,
liver, and spleen
Parasites
Toxoplasma gondii Central nervous system Meat, primarily pork
disorders
Trichinella spiralis Heart and neurologic Pork
disorders
----------------------------------------------------------------------------------------
\a Hemolytic-uremic syndrome.
Source: CDC, Food and Drug Administration, U.S. Department of
Agriculture, and Council for Agricultural Science and Technology.
ESTIMATED NUMBER OF ILLNESSES AND
DEATHS FROM FOUR MAJOR PATHOGENS
========================================================= Appendix III
In 1990, the Public Health Service identified E. coli O157:H7,
Salmonella, Listeria monocytogenes and Campylobacter jejuni as the
four most important foodborne pathogens in the United States because
of the severity and the estimated number of illnesses they cause.
According to CDC officials, illnesses caused by E. coli O157:H7 and
Listeria monocytogenes are generally more deadly than illnesses
caused by other foodborne pathogens. In contrast, illnesses caused
by Salmonella and Campylobacter jejuni are less likely to be deadly
but are more common. This appendix discusses the estimated number of
cases of foodborne illness caused by these pathogens.
E. COLI O157:H7
----------------------------------------------------- Appendix III:0.1
E. coli O157:H7 has emerged as an important cause of outbreaks of
foodborne illness in the United States since 1982. (See fig.
III.1).
Figure III.1: Reported Cases
of Illness from Outbreaks of E.
coli O157:H7, 1982-95
(See figure in printed
edition.)
Notes: 1995 data are preliminary. Approximately 66 percent of the
outbreaks from 1982-85 were foodborne. The remaining outbreaks
include those from other routes of infection, such as
person-to-person contact, or were unidentified and may also be
related to contaminated food.
Source: CDC.
Because few laboratories in the United States routinely test for E.
coli O157:H7, the actual number of illnesses caused by this pathogen
is unknown, but CDC officials estimate that this pathogen causes
approximately 21,000 illnesses annually. As shown in figure III.2,
only 33 states required reporting of such illnesses through the end
of 1994, according to information provided by CDC.
Figure III.2: Number of States
Requiring Reporting of E. coli
O157:H7, 1987-94
(See figure in printed
edition.)
Source: CDC.
Figure III.3 provides estimates of the percentage of people who
recover, remain ill, or die from E. coli O157:H7.
Figure III.3: Estimated
Percentage of People Who
Recover, Remain Ill, or Die
From E. coli O157:H7
(See figure in printed
edition.)
Source: Estimated by USDA's Economic Research Service on the basis
of information from CDC and other sources.
SALMONELLA
----------------------------------------------------- Appendix III:0.2
On the basis of population-based studies, CDC officials estimate that
between 800,000 and 4 million illnesses from the more than 2,000
strains of Salmonella occur each year in the United States. In 1994,
one strain, Salmonella Enteritidis, accounted for more than 25
percent of all reported infections from Salmonella. Confirmed
laboratory reports of the Salmonella Enteritidis strain increased
from 3,322 to 10,009 between 1982 to 1994. While the number of
outbreaks from Salmonella Enteritidis has declined since 1989, over
5,000 people, more than in any other year, became ill from the 44
outbreaks reported in 1994.
Figure III.4 shows the estimated percentage of people who recover or
die from all strains of Salmonella.
Figure III.4: Estimated
Percentage of People Who
Recover or Die From Salmonella
(See figure in printed
edition.)
Source: Estimated by USDA's Economic Research Service on the basis
of information from CDC and other sources.
LISTERIA MONOCYTOGENES
----------------------------------------------------- Appendix III:0.3
CDC estimates that the number of illnesses and deaths caused by
Listeria monocytogenes declined between 1989 and 1993, from 1,965
cases and 481 deaths to 1,092 cases and 248 deaths. CDC attributes
this downward trend to prevention efforts implemented by the food
industry and regulatory agencies. Figure III.5 shows the estimated
percentages of people who recover, remain ill, or die from Listeria
monocytogenes.
Figure III.5: Estimated
Percentage of People Who
Recover, Remain Ill, or Die
From Listeria monocytogenes
(See figure in printed
edition.)
Source: Estimated by USDA's Economic Research Service on the basis
of information from CDC and other sources.
CAMPYLOBACTER JEJUNI
----------------------------------------------------- Appendix III:0.4
According to CDC, Campylobacter jejuni is the most common bacterial
cause of diarrhea in the industrialized world. An estimated 2
million to 4 million cases occur each year in the United States,
according to population-based studies. Although the number of
Campylobacter jejuni cases confirmed by laboratory reports represents
only a small proportion of the total number of illnesses that are
estimated to occur from Campylobacter jejuni, the reported number
more than doubled from 3,947 in 1982 to 7,970 in 1989. Most cases of
illness occur sporadically and not as part of an outbreak. Illness
can occur from contact with raw foods (often poultry) during food
preparation. Figure III.6 shows the estimate of the percentage of
people who recover or die from Campylobacter jejuni.
Figure III.6: Estimated
Percentage of People Who
Recover or Die From
Campylobacter jejuni
(See figure in printed
edition.)
Source: Estimated by USDA's Economic Research Service on the basis
of information from CDC and other sources.
ESTIMATED COSTS OF FOODBORNE
ILLNESSES
========================================================== Appendix IV
This appendix provides information on the cost of foodborne illnesses
using both the cost-of-illness and the willingness-to-pay methods.
The range of estimates is wide, however, principally because of
uncertainty over the number of cases of foodborne illness and deaths.
Table IV.1 provides the estimated number of illnesses and deaths in
1993 used to calculate the cost-of-illness estimate. As the table
indicates, food was the most frequent source of contamination for
five of the seven pathogens the U.S. Department of Agriculture's
(USDA) Economic Research Service examined. CDC has targeted four of
these seven pathogens as the most threatening foodborne pathogens.
Table IV.1
Estimated Number of Illnesses and Deaths
for Seven Major Pathogens From All
Sources and From Food in 1993
Estimated Estimated Estimated Estimated
number of number of Estimated number of number of
Pathogens illnesses deaths percentage illnesses deaths
------------------- ------------ ------------ ------------ ------------ ------------
Campylobacter 2,500,000 200-730 55-70 1,375,000- 110-511
jejuni or coli 1,750,000
Clostridium 10,000 100 100 10,000 100
perfringens
E. coli O157:H7 10,000- 200-500 80 8,000- 160-400
20,000 16,000
Listeria 1,795-1,860 445-510 85-95 1,526-1,767 378-485
monocytogenes
Salmonella 800,000- 800-4,000 87-96 696,000- 696-3,840
4,000,000 3,840,000
Staphylococcus 8,900,000 7,120 17 1,513,000 1,210
aureus
Toxoplasma gondii 4,111 82 50 2,056 41
=========================================================================================
Total 12,225,906- 8,947- Not 3,605,582- 2,695-6,587
15,435,971 13,042 applicable 7,132,823
-----------------------------------------------------------------------------------------
Source: Economic Research Service, USDA.
Table IV.2 presents cost-of-illness estimates for all foodborne
illnesses and illnesses from meat and poultry. Contaminated meat and
poultry are believed to be among the most common sources of foodborne
illness from these pathogens.
Table IV.2
"Cost-of-Illness" Estimates for Seven
Major Pathogens From All Foods and From
Meat and Poultry During 1993
(Dollars in billions)
Estimated Estimated Estimated
Pathogens costs percentage costs
---------------------------- ------------ ------------ ------------
Campylobacter jejuni or coli $0.6-1.0 75 $0.5-0.8
Clostridium perfringens 0.1 50 0.1
E. coli O157:H7 0.2-0.6 75 0.2-0.5
Listeria monocytogenes 0.2-0.3 50 0.1-0.2
Salmonella 0.6-3.5 50-75 0.3-2.6
Staphylococcus aureus 1.2 50 0.6
Toxoplasma gondii 2.7 100 2.7
======================================================================
Total $5.6-9.4 Not $4.5-7.5
applicable
----------------------------------------------------------------------
Source: Economic Research Service and Food Safety and Inspection
Service, USDA.
ERS also used the willingness-to-pay method to estimate the value of
preventing deaths for five of the seven major pathogens. The results
of this analysis are shown in table IV.3.
Table IV.3
"Willingness-to-Pay" Estimate for Five
Major Pathogens in 1992
(Dollars in billions)
Estimated number
of
deaths from Implied value of
Pathogens foodborne sources preventing deaths
------------------------------ ------------------ ------------------
Salmonella 960-1,920 $3.8-13.4
Campylobacter jejuni or coli 120-360 .48-2.5
E. coli O157:H7 146-389 .58-2.7
Listeria monocytogenes 378-433 1.5-3.0
Toxoplasma gondii 42 .17-.29
======================================================================
Total 1,646-3,144 $6.6-22.0
----------------------------------------------------------------------
Source: Economic Research Service, USDA.
MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V
Edward M. Zadjura, Assistant Director
Jay Cherlow, Assistant Director for Economic Analysis
Daniel F. Alspaugh, Project Leader
Carol Herrnstadt Shulman
Jonathan M. Silverman
*** End of document. ***