Good morning. I am Dr. Stephen M. Ostroff, Associate Director for Epidemiologic Science at
the National Center for Infectious Diseases, Centers for Disease Control and Prevention (CDC).
I am accompanied by Dr. Barbara Herwaldt, also of the National Center for Infectious Diseases.
I am pleased to be here this morning to discuss CDC's programs to monitor, prevent, and control
foodborne diseases in the United States. I will provide an overview of CDC's foodborne disease
surveillance systems and describe cyclosporiasis associated with imported raspberries as an
example of our role in outbreak response.
Although the United States has one of the safest food supplies in the world, the public health
burden of foodborne diseases is still substantial. The Council for Agricultural Science and
Technology has estimated that as many as 9,000 deaths and 6.5 to 33 million illnesses in the
United States each year are fool-related. Foodborne disease costs the U.S. economy several
billion dollars annually. A variety of pathogens and toxins have been described as causes of
foodborne disease, and new ones continue to be identified.
In 1997, in response to the growing concern about food safety, the President announced the
National Food Safety Initiative. CDC's collaborative involvement with the Food and Drug
Administration (FDA), the U.S. Department of Agriculture (USDA), and the Environmental
Protection Agency in the ongoing expansion of this initiative responds to the new challenges by
building a national early warning system for hazards in the food supply through enhanced
capacity for surveillance and outbreak investigations at the State and federal levels. Specific
activities of the Initiative include expanding the scope of FoodNet, CDC's active foodborne
disease surveillance system, using it to define the true incidence of many diagnosed foodborne
infections and to assess their sources and potential for control; developing and standardizing new
and rapid diagnostic techniques and molecular subtyping, or fingerprinting, for foodborne
pathogens; and designing and delivering training programs for epidemiologists, laboratorians,
and health professionals.
Foodborne diseases are common and, in principle, preventable. Some of the causes of
foodborne diseases that were formerly problematic are now well controlled by standard
prevention strategies, such as pasteurizing raw milk, appropriately managing the canning of food,
and ensuring that restaurants and other food preparation areas are clean and well maintained.
However, new challenges continue to arise, including the increasing globalization of our food
supply, larger scale production and distribution networks, and changing dietary habits; and new
efforts are required to address these issues.
Preventing foodborne disease requires a coordinated program of risk assessment and risk
management involving Federal, State, and local agencies and non-governmental partners. CDC's
primary role in this coordinated effort is to identify foodborne hazards, characterize the risk of
foodborne disease, and identify strategies that will prevent additional cases. In 1994, CDC
issued a strategic plan, Addressing Emerging Infectious Disease Threats: A Prevention Strategy
for the United States, which emphasized surveillance, applied research, and prevention activities.
An updated version of this plan will be published later this year. As in the 1994 plan, many
aspects of the new version of the plan deal with emerging infectious foodborne diseases. The
plan complements sections of FDA's 1997 Food Code and the Hazard Analysis and Critical
Control Point (HACCP) food safety programs being implemented by FDA and USDA's Food
Safety and Inspection Service (FSIS) and provides a platform from which CDC's role in the
National Food Safety Initiative, which was launched in 1997, can be instituted.
tification of Foodborne Diseases Problems
A person who becomes ill with a foodborne disease may be part of an outbreak or cluster (a
group of patients who all have the same illness after consuming the same food) or may have a
sporadic illness (an illness that may be an isolated occurrence and not part of a recognized
cluster). Usually, investigations of outbreaks can rapidly determine the source and nature of the
illness and identify the control measures needed to prevent additional cases. However, sporadic
illnesses are often not diagnosed or considered to be foodborne. Even if they are recognized as
being foodborne, it is usually impossible, for single cases, to determine which food is the source
of the infection. Because individual sporadic cases are far more common than outbreaks, they
are a prime target for prevention efforts.
Effective public health surveillance is key to identifying and monitoring the prevalence of
foodborne disease. CDC is typically notified of a potential foodborne disease problem by a State
or local health department or by an astute clinician or laboratorian who notices an unusually
large number of cases of a certain disease. Physician-based surveillance is useful for public
health emergencies that require rapid response, such as potentially lethal botulism, where one
case could herald an outbreak and immediate public health action is necessary. Clinical
laboratories help detect foodborne diseases by tracking the number of times they identify a
specific pathogen. Clinical laboratory-based surveillance has identified multiple outbreaks,
including a recent multistate outbreak of Salmonella Agona infection linked to cereal. State
public health laboratories play an important role in further characterizing the strains isolated
from ill people, to see whether there are groups of similar pathogens. Taking advantage of recent
advances in computer technology and molecular biology, CDC has developed PulseNet, a
network of molecular subtyping (fingerprinting) laboratories at State health departments, FDA,
USDA, and CDC, which enhances the ability of laboratory-based surveillance to rapidly identify
clusters of related foodborne infections of certain pathogens, sometimes scattered over large
Another source of data is CDC's FoodNet, which is conducted in CDC's seven Emerging
Infections Program sites developed as part of the 1994 emerging infections plan. The FDA and
the Food Safety and Inspection Service (FSIS) of USDA are providing financial assistance and
are important collaborators with CDC in this system. The seven active surveillance sites cover
about 7.7% of the U.S. population. These sites actively seek out information on foodborne
illnesses identified by clinical laboratories, collect information from patients about their
illnesses, and conduct investigations to determine which foods are linked to specific pathogens.
As data are collected, this surveillance system provides important information about changes
over time in the burden of foodborne diseases and will help the agencies evaluate current food
safety initiatives and develop future food safety activities.
For these surveillance systems to be effective and for an illness to be identified as caused by a
foodborne pathogen, several things must occur. A person who eats contaminated food and
becomes ill must seek medical attention or contact the health department. The patient's clinicians
must obtain appropriate diagnostic tests. The laboratory results must be reported to the health
department. Information must be assessed to recognize a potential outbreak. Often, not all of
these steps occur, and sporadic illnesses and outbreaks are not recognized or reported.
Once an outbreak is detected, the first response is usually from the State or local health
department. When necessary, the State or local health department conducts an outbreak
investigation. Due to limited resources at State and local levels, not all outbreaks can be
investigated and reported. If an outbreak is very large or significant, is thought to involve an
unusual pathogen or unexpected food vehicle, affects multiple states or countries, or when
preliminary investigations do not reveal a source, CDC will often be invited by the State health
departments to participate in the investigation.
When investigating an outbreak of a foodborne illness, public health officials must combine
laboratory diagnostic techniques and epidemiologic investigative methods to determine both the
causative agent of the illness and the vehicle for its transmission. This involves interviews with
patients and comparison of their responses to those of non-ill persons (control subjects) to
determine which foods are implicated. If a food is identified as the source of illness, CDC
collaborates with FDA or USDA on the investigation and control of the outbreak, based upon
which agency regulates the food suspected.
Approximately 400-500 foodborne outbreaks are reported by State health departments to CDC
each year, accounting for 10,000 to 12,000 persons with foodborne illness. CDC summarizes the
information in these reports through its Foodborne Disease Outbreak Surveillance System. The
reports provide useful, detailed information on particular diseases and on the type and severity of
outbreaks that occur in various settings, such as nursing homes or schools. Outbreak
investigations can lead to effective prevention strategies, as they are often critical in identifying
contaminated foods that can then be removed from the marketplace and in elucidating the
problems in food production that lead to disease.
Case Study: Cyclospora cayetanensis
CDC's role in outbreak investigation is well illustrated by the 1996 and 1997 outbreaks of
infection caused by Cyclospora cayetanensis, a recently characterized parasite that causes a
gastrointestinal illness called cyclosporiasis. This illness is typically characterized by watery
diarrhea and other symptoms, such as nausea, abdominal cramps, weight loss, and fatigue. If not
treated, the illness can be severe and prolonged. Before 1996, most of the small number of cases
of cyclosporiasis in the United States occurred in travelers who had been in developing countries,
and only three small U.S. outbreaks had been reported.
When the pattern of Cyclospora infections changed in 1996 and health departments noted cases
of cyclosporiasis in people who had not traveled overseas, CDC was notified promptly. In mid-May of 1996, health departments in Florida and New York informed CDC that sporadic cases of
cyclosporiasis had been identified in their states. At the end of May, health departments in Texas
and Canada informed CDC that some people who had attended specific events, such as a party,
had become ill with cyclosporiasis. Thus, CDC was notified of "clusters" of cases, which
indicated that an outbreak might be occurring. In June, CDC learned of additional sporadic cases
and clusters in the eastern United States and Canada.
Ultimately, 55 clusters with a total of 725 cases of cyclosporiasis were reported to CDC by 14
States, the District of Columbia, and two Canadian provinces. The 55 clusters were associated
with events that occurred May 3 through June 14, 1996. In addition, 740 sporadic cases that
were not associated with identified events or overseas travel were reported, for an overall total of
1,465 cases from 20 States, the District of Columbia, and two provinces. Twenty-two people are
known to have been hospitalized, but no deaths are known to have occurred. Because in most
foodborne outbreaks, particularly those that involve more than one locality, many affected cases
are unrecognized or unreported, the total number of cases of cyclosporiasis that occurred in this
outbreak may have been much larger than the officially reported number.
CDC played many roles in the outbreak investigation, including serving as the national reference
laboratory for identifying Cyclospora in stool specimens and thus confirming that this parasite
caused the outbreak. This role was particularly important because many laboratorians had not
had experience identifying Cyclospora.
Another role CDC played was to help State and local health departments conduct the studies that
ultimately implicated raspberries as the food item that had made people sick. This aspect of the
outbreak investigation focused on the clusters of cases that were associated with specific events.
Health departments interviewed the people who had attended the respective events about what
they had consumed and compared the responses of the sick and the well people to see how they
differed. CDC assisted in various ways for example, by helping to design questionnaires,
conduct data analysis, and identify important issues that needed to be addressed in theinvestigations. CDC also assisted some health departments on site with their local
As more and more clusters of cases were identified, CDC's coordinating role at the national level
became increasingly important. CDC sponsored frequent conference calls and a meeting in July
1996 to discuss the findings to date and to help establish priorities for the investigation and
future research. Whereas the investigators from individual States and localities focused on their
own jurisdictions, staff at CDC repeatedly looked for the patterns that emerged as data from the
individual clusters were compiled and analyzed. Fresh raspberries were found to have been
served at virtually all the events, and a strong statistical association was found between illness
and consumption of raspberries. Although the investigation focused on the clusters of cases,
some studies that compared the exposures of sporadic cases and control subjects were also
conducted and implicated raspberries.
Another important role played by CDC was that of coordinating public communications as the
investigation progressed. CDC helped improve the consistency of the messages that State and
local health departments gave to local media. CDC provided a national perspective about the
outbreak when interviewed by the national media and published articles in CDC's Morbidity and
Mortality Weekly Report to rapidly communicate important findings about the investigation to
the public health and medical communities.
Once it was determined that raspberries were the food item responsible for illness, the next step
was to determine where they had been grown. This traceback process required close
coordination with FDA, State and local agencies, and industry. The first steps of the tracebacks
entailed determining where the various events took place and where the raspberries that were
served had been bought. The raspberries were then tracked from suppliers and distributors back
to importers, exporters, and farms of origin, looking for common themes at each step. The
available traceback data implicated Guatemala as the common source for the raspberries. By the
time Guatemalan raspberries were implicated, Guatemala's spring export season had essentially
Investigators next tried to determine how the raspberries became contaminated. CDC and FDA
sent investigators to Guatemala and to Miami, a major port of entry for imported raspberries, to
explore possible modes of contamination. We were able to observe how raspberries were grown,
picked, sorted, packed, cooled, transported, and inspected. Because no single packing or storage
facility in Guatemala, exporter, type of shipping container, shipment, airline carrier, U.S. port of
entry or cargo clearance area, importer, distributor, retailer, or food handler was linked to all
events for which we had adequate data about the source of the implicated raspberries, we
concluded that some practice or attribute common to multiple farms was the most likely
explanation for the outbreak.
Although the mode of contamination was not determined, one hypothesis under consideration is
that contaminated water may have been used to mix the insecticides, fungicides, and fertilizers
that were sprayed on raspberries. Good laboratory methods for detecting low levels of the
Cyclospora parasite on produce such as raspberries or in water and other environmental samplesare not yet available. By the time the clusters of cases were detected, leftover raspberries from
the events were not available for testing.
Although the precise mechanism by which raspberries became contaminated was unclear, FDA
and CDC provided suggestions to the Guatemalan Berry Commission (GBC) about possible
ways to reduce the risk for contamination. The GBC voluntarily implemented various prudent
measures to improve water quality and sanitary conditions on farms that were going to export to
the United States in subsequent export seasons.
Despite these control measures, another multistate outbreak linked to Guatemalan raspberries
occurred in North America in the spring of 1997. CDC learned of this outbreak in early May
1997, when several health departments informed CDC of clusters of cases that were associated
with April events. Ultimately, 41 clusters with a total of 762 cases were reported, which were
associated with events that occurred April 1 through May 26, 1997, in 13 states, the District of
Columbia, and one Canadian province. In addition, 250 sporadic cases were reported for the
outbreak period, for an overall total of 1,012 cases in 17 states, the District of Columbia, and two
Once again, the investigation, which focused on the clusters of cases, implicated fresh
raspberries, and Guatemala was found to be the major source of the implicated berries. The
outbreak ended shortly after Guatemala voluntarily suspended exportation of fresh raspberries to
the United States at the end of May 1997. The fact that another outbreak occurred despite the
implementation of various control measures suggests either that the control measures may not
have been fully implemented by some farms or that the measures may not have addressed the
true source of contamination of the raspberries.
These outbreaks in 1996 and 1997 highlighted challenges related to the investigation of
outbreaks of foodborne diseases. Many State and local health departments do not have the
necessary infrastructure to conduct outbreak investigations. Also, because Cyclospora is an
emerging pathogen, most laboratorians lacked the experience and expertise to identify
Cyclospora in stool specimens, particularly during the 1996 investigation. CDC is developing
the capacity to use the Internet to assist laboratories in identifying parasites such as Cyclospora
in patient specimens. However, many laboratories do not yet have the necessary equipment to
take advantage of this technology.
New Challenges New Opportunities
As we draw to the close of the 20th century, we face new paradigms for foodborne disease due to
the globalization of the food supply, the large-scale nature of food production and distribution,
and the continuing recognition of new foodborne pathogens. CDC addresses these issues by
harnessing the technology of electronic telecommunications and computer systems, developing
state of the art molecular fingerprinting techniques, integrating its disease prevention and control
activities with food safety programs in FDA and USDA, and building active epidemiologic and
laboratory-based surveillance programs in collaboration with our State and local partners.
However, much work needs to be done to build the necessary architecture for a truly sensitive
and responsive early warning network. Building investigative and laboratory capacity in all of
our State partners, enhancing our collaborative activities with international partners where,
increasingly, some of our food supplies originate, and improving the qualitative and quantitative
understanding of critical food safety problems are important components of CDC's response.
CDC has been working with the Council of State and Territorial Epidemiologists and the
Association of State and Territorial Public Health Laboratory Directors to enhance core
surveillance capacity and to assure that the appropriate architecture exists. A 21st century
system is needed to confront 21st century challenges.
In conclusion, strong Federal, State, and local public health surveillance networks are the
foundation for rapid identification and investigation of infectious disease threats, including those
illnesses that are caused by foodborne pathogens. Foodborne diseases remain a challenge for
public health. To address this challenge will require continued investments in our public health
infrastructure and strong partnerships among State and local health departments and Federal
Thank you for the opportunity to discuss the surveillance of foodborne disease. We will be
happy to answer questions you or other members of the Subcommittee may have.