Many of us have experienced the frustration of learning of a foodborne disease outbreak and wondering why didn’t we know about it sooner, why we don’t know all the details we wanted to know, why the investigation wasn’t successful or why we weren’t informed of the final details.
There are challenges to foodborne disease outbreak, surveillance, detection and response. This article is not a long list of excuses, but rather some insights into the challenges that investigators face every day.
Government and industry have been investigating foodborne illness and outbreaks for decades. These investigations occur at multiple organizational levels and they involve multiple components of organizations.
The complexity of the investigations and the multiple players involved invariably lead to coordination challenges. In government alone you can have local, state, federal and international organizations involved and at each level you can have epidemiologists, clinical laboratories, food laboratories and regulatory agencies all with overlapping roles and responsibilities.
Investigators need to quickly identify the vehicle (contaminated food) and alert the public to minimize further cases of illness and at the same time they need to be right.
In the U.S., the 1996 Cyclospora outbreak first linked to strawberries and later linked to raspberries and the 2008 Salmonella Saintpaul outbreak first linked to tomatoes and later to Jalapeno peppers are well-known examples of what can happen if an outbreak is incorrectly linked to a food.
Being wrong can cause serious economic harm to industry, reduce public and industry confidence in government and undermine support and trust for future investigations and findings. These issues are further complicated by a growing public and political distrust of government and science in general.
Foodborne illness outbreaks can generate tremendous public, media and political interest. Media and political demands for information can distract agencies struggling to complete their investigations. Because investigating agencies also must ensure the information provided has been vetted and confirmed, this often means the information is not provided as soon as others want it.
Congressional hearings after some past national outbreaks did help to raise awareness of foodborne disease, but the “theater” aspect of such hearings also can distort understanding of information that is available.
Although many people only hear about foodborne disease illness when it is linked to a nationally distributed food, that view is a very limited one. Most reported foodborne illness cases are so-called sporadic cases where no specific risk factor for illness is determined.
Approximately 1 – 5 percent of all reported sporadic cases of foodborne illness agents are ever linked to an outbreak. The victims themselves self-identify and report most foodborne disease outbreaks to state or local agencies. These outbreaks are often linked to local restaurants, investigated locally and often not reported in the media. Such outbreaks are not always reported to the Centers for Disease Control’s voluntary foodborne illness surveillance system, either. Many of these investigations are inconclusive, just like the multi-state investigations readers are more aware of.
The CDC’s PulseNet laboratory-based surveillance system has been successful in identifying many diffuse outbreaks since its inception in 1995. Many of these outbreaks involve illness cases that are widely dispersed geographically and likely would not have been linked as being an outbreak without PulseNet.
These outbreak investigations have often resulted in greater public awareness of the food contamination due to public alerts and recalls. Difficult processes to develop and issue them often precede these alerts. Multiple agencies at local, state and federal levels can see message priorities differently and they can work under different rules for what they can say, all of which creates challenges in developing and providing consistent public information.
Our Government System
The men who created the U.S. Constitution distrusted central authority/power in government. Therefore, they created a federalist form of government with much authority and responsibility, including public health, residing in the states. Just look at our name. We are the “United States” of America, not America. Similarly, many states have home rule constitutions and other similar provisions such as commonwealth constitutions that call for government services and authority to be provided at the local level.
As a result, there are approximately 3,000 state and local agencies with some regulatory responsibility for food safety in the U.S. Many of these agencies also have their own epidemiology/disease surveillance programs and public health laboratories as well.
When it comes to foodborne disease surveillance, detection, investigation and response, there often is a lack of clarity over roles and responsibilities and decision-making authority. Staff qualifications and training levels vary and staffing levels have been declining.
Staffs who conduct these investigations usually have many responsibilities in addition to foodborne disease. Epidemiologists are conducting surveillance for many diseases and often have multiple investigations they are juggling at the same time, nurses who interview illness cases may also be providing immunization clinics and visiting homebound patients. Regulatory staffs have inspection quotas to meet.
Federal or state epidemiologists may need a follow-up interview on a sporadic case of illness that is a PFGE match to other sporadic cases around the country. Local nurses may have to juggle that request with their need to visit patients and conduct clinics.
Think back to the confusion during the government response to Hurricane Katrina. In multi-state foodborne outbreak investigations, government agencies at multiple levels are trying to work in a coordinated way with people they will never meet and over multiple time zones. The scale of such outbreak investigations is not as large as the hurricane response, but the coordination challenges can be similar.
Some of the smaller agencies that do not conduct many foodborne disease outbreak investigations suddenly find themselves thrust into a situation they are inadequately staffed or prepared for. Federal food safety agencies depend on under-funded state and local agencies to identify and investigate outbreaks linked to products that they regulate.
Nevertheless, staffs at all levels of government are invariably dedicated to public service and public health. This asset is drawn on in every investigation. When we talk about organizations that are more successful than others, this often reflects a few key individuals who are really making the difference.
Government at all levels is facing serious challenges with the loss of funding that translates into vacancies not being filled, supplies/equipment not being purchased or replaced, staff salaries and benefits being cut and travel being restricted.
Experienced staff is leaving for retirement before benefits are lost or to take better paying jobs elsewhere. New staff may not come with sufficient education/training for their functions and there is little or no funding to train them once they are on the job. Cuts in training, salary and benefits for new government hires will result in a different workforce in the future. We cannot predict what imp
acts this different workforce will have on government foodborne disease activities.
A discussion about government foodborne outbreak investigation challenges would not be complete with out mentioning information/data sharing.
All parties involved in outbreak investigations are prohibited from sharing patient identifiers, regulatory agencies are sometimes prohibited from sharing commercial confidential or proprietary information (processing methods, customer lists) as well as investigational findings that might be used in any future enforcement actions.
Some agencies have more institutional resistance to sharing information than others. These laws, regulations and policies can slow down investigations and lead to friction between investigation and response organizations over access to information.
Similarly, the food industry is looking for information about what is going on to help them inform government investigations, focus and speed up recalls and to rapidly put in place interventions to prevent future outbreaks. Consumer groups and the public want actionable information quickly to protect themselves and their families. Government agencies are looking for ways to be more transparent, but progress has been slow so far.
Typical foodborne disease outbreak epidemiological investigations involve analytic studies to determine statistically significant associations between certain food exposures and illness. This is an iterative process, because the contaminated food is not known at the beginning of the investigation and may not be identified by the initial epidemiological studies.
Further, initial laboratory and environmental information may not be sufficient to help identify the vehicle. The investigations move as quickly as possible to identify the food and thereafter prevent additional exposures and illnesses. These investigations usually do identify the correct food, but mistakes have been made. Every investigation faces the same dilemma of needing to be fast to prevent additional illness and needing to be right to prevent identifying the incorrect food.
One of the key jobs of an epidemiologist is to collect and analyze disease data. The data needs to be accurate if the conclusions drawn from it are to be accurate. Much of epidemiologists’ time, therefore, is spent trying to improve the quality of the data they collect and trying to find ways to get more and better data.
The skills to perform these tasks are most likely gained with a graduate degree in epidemiology and yet many epidemiology offices have few if any staff trained at this level. Most epidemiologist’s, and public health laboratory staff, who conduct foodborne disease surveillance at the state and local level are funded by CDC grants, not the state or local agency they work for. That funding has been shrinking.
This loss of disease surveillance capacity could pose a longer-term dilemma along with the obvious loss of immediate outbreak detection capacity. The impact of the 2011 Food Safety Modernization Act on food safety will not be measurable if state and local agencies do not have the capacity to detect any changes in disease incidence and prevalence resulting from implementation of the act. Would any hoped for reduction in reported sporadic cases of illness and outbreaks mean that food is safer or that surveillance is less robust?
An article “Food Safety Epidemiology Capacity in State Health Departments – United States, 2010” in the December 23, 2011 edition of the CDC’s “Morbidity and Mortality Weekly Report” listed barriers state health departments reported to their foodborne disease surveillance programs.
The barriers are listed below with the number of states reporting each one:
– Delayed notification of outbreaks (41)
– Lack of foodborne safety staff (29)
– Lower prioritization of investigations (27)
– Lack of ability to pay overtime (20)
– Lack of adequate epidemiology expertise (12)
– Difficulties working with in-state agencies (8)
– Constraints in administrative support (8)
– Difficulties working with other state of federal agencies (5)
So-called stealth or hidden ingredients as vehicles are being identified more often in foodborne disease outbreaks. These outbreaks are also referred to as ingredient driven outbreaks. No doubt they have always been there, we just were not finding them. They may help to explain why so many outbreak investigations are inconclusive.
Two of the best-known examples of stealth ingredient outbreaks have already been mentioned. The 1996 Cyclospora outbreak first linked to strawberries and later to raspberries hinged on whether the chef who baked wedding cakes topped them with one type of berry or another and that in turn depended on price and availability of berries.
The 2008 multi-state Salmonella Saintpaul outbreak was first associated with tomatoes and later shown to be linked to Jalapeno peppers. That initial vehicle identification hinged on what questions were asked of early victims of the outbreak. Victims were asked if they had eaten tomatoes and peppers among many other foods. It is much easier to remember that you ate tomatoes in some form than that you ate Jalapeno peppers served in fresh salsa. Also, Jalapeno peppers had no history of being a Salmonella vehicle and tomatoes had been a vehicle for Salmonella many times in the past.
These examples and more recent ones involving ground pepper used as a spice in restaurants (Salmonella Rissen) and on processed meats (Salmonella Montevideo) point to how nuanced it can be to identify vehicles from epidemiological studies.
Investigators are looking for additional tools such as product testing and ingredient tracebacks to identify vehicles that case control or cohort epidemiological studies alone cannot tease out.
Similar problems have been vexing investigators in outbreaks where multiple ingredient foods like tomatoes and lettuce are served in the same dish (salad, sandwich). Mexican-style foods have been involved in several outbreaks in which tomatoes, lettuce, cheese and ground meat were served together in tacos and other similar dishes. All of these ingredients have been identified as food vehicles in the past, so how do outbreak investigators discover which one was contaminated when they are served and eaten together?
Laboratories are too often the forgotten partners in foodborne disease surveillance systems, even though a “gold standard” outbreak investigation includes laboratory confirmation of the agent in patient and food samples.
Laboratory surveillance enables the identification of sporadic case clusters that can lead to identifying otherwise unrecognized outbreaks. In some ways, the laboratory-based PulseNet system has become a victim of its own success. So many disease clusters are identified by this CDC led system that the epidemiologists and laboratorians have to triage what clusters they have the resources to investigate.
Thirty to 40 disease clusters are on their radar at any point in time. Unfortunately, the Pul
seNet system and public health laboratories in general are suffering from budget cuts that are threatening to severely limit the effectiveness of this vital disease surveillance system.
Laboratories can be left out of efforts to analyze the data that they have generated and from planning to develop new surveillance systems. Public health labs can have a very hard time keeping up with new technologies as they lack the resources to conduct proof of principle studies to demonstrate the value of moving to those technologies.
Laboratories also can struggle to maintain standardization of methods. Under the best of circumstances, they are not always going to be able to recover an agent from a sample even if it is there. Laboratories can be left out of planning for ongoing investigations as well as be left out of planning and implementation of interventions after an investigation.
Laboratorians are often only seen as working in their labs. Laboratorians need to be part of field investigation teams to bring their knowledge to bear when looking for places to collect samples and to ensure that samples are properly collected and submitted.
In a somewhat ironic development, newly available rapid test methods for foodborne pathogens being used in the clinical setting have meant that fewer bacterial cultures are being forwarded to public health laboratories for further characterization, such as serotyping and PFGE testing.
These further tests are what enable public health laboratorians and epidemiologists to identify possible disease clusters and link otherwise sporadic cases to ongoing outbreaks. If public health agencies can’t identify outbreaks, then they won’t be investigated, additional exposures will not be prevented and food safety in general will suffer.
Environmental Health / Food Regulatory Challenges
This part of the foodborne disease outbreak system is all about what are we going to do about the outbreak / contamination? A bad thing has happened; food may have made people sick. Who did this and what are we going to do about it? Are our objectives, to prevent more illnesses or to take some kind of regulatory or punitive action against the wrong doer? Can we have it both ways?
The historical approach to environmental health/food regulatory investigations in response to food outbreaks has been to conduct regulatory inspections at locations implicated: food service, retail, manufacturing and production and to take regulatory action of some sort to “remedy” the problem and punish a responsible party.
Regulatory actions might include food seizures, citations for violations of regulations, recalls of food, public alerts and even closure of a firm. Investigators are expected to get all of this done quickly and because they are regulators to do so following the appropriate procedures and legal restrictions on how they conduct their activities.
Firm operators are expected to cooperate with the regulatory investigations in the face of allegations that they have made customers ill, that they likely will face bad publicity, loss of money, possible government and or private legal action and loss of reputation.
Some operators are public-health minded enough to cooperate in this setting. Others do not.
What kind of outcome do we expect to reach given this approach to these investigations? Do such investigations prevent future outbreaks/food contamination at the immediate source of contamination or in the wider industry? Does a successful regulatory or civil legal action or even the threat of one result in safer food in the future? Are investigations that are directed toward possible legal actions even collecting all of the information needed to inform future preventive activities?
Many government environmental health/food regulatory agencies are beginning to conduct environmental assessments/root cause analyses, a systems analysis approach, in response to foodborne disease outbreaks and other food contamination events rather than or in addition to regulatory inspections.
These investigations are intended to identify the contributing factors, what went wrong like inadequate cooking or cross contamination, that lead to the outbreak and the environmental antecedents (root causes) that lead to the contributing factors happening. Environmental antecedents are findings such as inadequate equipment, inadequate employee training and supervision, lack of a food safety culture, lack of a HACCP system.
Environmental assessments mean a very different kind of investigation needs to be conducted and a very different relationship needs to exist between an operator and investigator. Some of the most important findings may not even constitute regulatory violations. These findings often necessitate working out of the bounds of a historical regulatory inspection.
In an environmental assessment investigators are not there to find violations of regulations, they are looking at what happened, regardless of the regulation. Ideally regulatory violations are not even cited. To be fully successful an environmental assessment needs to be a cooperative investigation between an operator and an investigator with full transparency from both parties. They are trying to find out what went wrong in the system, not what legal action will result.
A regulatory investigation understandably can be a cat and mouse game, as an operator is in damage-control mode. In such a setting, it is much less likely that the contributing factors and environmental antecedents will be found. Many operators do not see it to be in their best interest to help to identify conditions that could expose them to greater jeopardy and liability.
The Food Safety Modernization Act directs FDA to put more emphasis on implementation of preventive controls, which are developed from an understanding of conditions and practices that can lead to contamination and subsequent foodborne illness. Regulatory inspections are much less likely to provide information to inform preventive controls than are environmental assessments.
The question facing us all is what do we want out of investigations of foodborne disease outbreaks and what are we willing to give up to get us there? An emphasis on the legal aspects of an outbreak may not lead to a true systems analysis of what went wrong.
Historical investigations limit themselves to violations at one or more settings, not to root causes for the outbreak. They also do not look at the wider food safety system in industry and government and how all of those aspects need to improve if future outbreaks are to be prevented. Implementation of environmental assessment will be a significant first step in improving our overall food safety system and food safety culture.
Federal agencies depend on timely and complete state and local epidemiology, laboratory and environmental data to inform their investigations going forward. Incomplete, inaccurate, and untimely information has slowed federal investigations and yet the state and local agencies federal agencies depend on often have inadequate levels of staff. Some of those staff have received little or inadequate training and have little experience in conducting foodborne disease outbreak investigations.
They also face many competing priorities in their respective agencies. Communications issues and lack of clar
ity over roles and responsibilities plague many investigations. The same challenges for federal agencies also impact those local surveillance and response systems and the state/local outbreaks that they investigate.
One problem that every environmental investigation faces is the time that has passed between the contamination of the food and the beginning of the investigation into how that contamination occurred. At best it can be a few days, but in some occasions months have passed before the investigation gets to this location. By this time left over foods and ingredients may all be gone. Key employees may be gone.
In outbreaks involving fresh produce, the field where the produce was grown may be plowed and fallow, or another crop may be growing there. Foodborne outbreak investigations can not be considered successful if investigators have not been able to identify the contributing factors that caused the outbreak and the environmental antecedents that lead to the contributing factors.
Very few outbreak investigations that identify a vehicle also report contributing factors and even fewer report environmental antecedents. This lack of information seriously limits everyone’s ability to identify future preventive controls.
Put another way, we fail to learn from others mistakes.
What Is Being Done and What Works?
Government agencies at all levels recognize these problems and are working to address them. Agencies are accomplishing this through development of new and enhancing existing partnerships, providing training including cross training with partners, setting and implementing program standards, piloting new approaches and developing protocols to improve the system from surveillance through response.
Partnerships are being developed among government agencies, with universities and with industry organizations. One example of the partnerships is the Council to Improve Foodborne Outbreak Response (CIFOR). This organization of representatives of government agencies and disciplines at the local, state and federal level has produced the “CIFOR Guidelines to Improve Foodborne Outbreak Response” and the CIFOR Guidelines “Toolkit.” These two documents are being used by government agencies to develop more effective and efficient systems.
FDA has created its new foodborne disease response team CORE (Coordinated Outbreak Response and Evaluation) to help improve internal management of foodborne outbreak surveillance and response, as well as improve coordination with other government agencies.
FDA is also funding several Rapid Response Teams made up of members from state food safety regulatory agencies, public health agencies and the associated laboratories, along with area FDA District Office members.
In cooperation with state and local agencies, FDA has also developed retail food and manufactured food model regulatory program standards, and works with state and local food safety regulatory agencies to assist them in implementing these standards. Standard number five for both of these sets of standards sets out recommended practices for foodborne disease surveillance and response programs.
CDC has been promoting and funding improved public health laboratory and epidemiology capacity in state and local public health agencies for decades. One of the newest efforts has been the funding of several pilot outbreak surveillance teams known as FoodCORE sites. These public health agencies are working to improve laboratory and epidemiology surveillance for foodborne diseases as well as improved collaboration with environmental health/food regulatory partners.
CDC has funded the Environmental Health Specialist – Network, also know as EHS – Net, for more than 10 years. EHS – Net has funded state agencies to study the causes of foodborne illness linked to retail food establishments and to develop improved procedures for conducing environmental assessments at locations linked to outbreaks.
The Department of Homeland Security has funded the updating of the “Food Emergency Response Plan” (FERP). This model emergency response plan is focused on intentional contamination of the food supply, but since many of the same players are also involved in unintentional contamination the document can be useful for agencies preparing for those events as well.
The Public Health Accreditation Board has developed a set of standards that state, local and tribal public health departments need to meet of they wish to be accredited. Those standards include many provisions that directly or indirectly impact the effectiveness of foodborne disease surveillance, detection, response and prevention.
The Food safety Modernization Act contains a number of provisions that direct FDA and CDC to improve foodborne disease surveillance and response. Both agencies are to improve capacity at state and local agencies and CDC is required to establish Centers of Excellence.
There are no quick fixes that will make dramatic improvements in the U.S. foodborne disease surveillance and response system. The system is just too complex, government funding is too tight and the challenges are too numerous for quick fixes.
Looking for someone to blame will not make things better any sooner either. Government agencies recognize they need to do better. They will need to clarify roles and responsibilities, improve communications and coordination, share findings more freely and increase capacity.
There are many ongoing efforts working towards these improvements, but we will continue to be frustrated by the consequences of the many challenges we face.
John J. Guzewich (Jack) retired from the FDA Center for Food Safety and Applied Nutrition in August of 2011. He was Senior Advisor for Environmental Health in the Office of Food Defense, Communication and Emergency Response.