Where Current US Food Safety Activities Fall Short
Over three decades ago, Thomas Kuhn suggested that scientific paradigms remain dominant until they fail to solve new problems (Kuhn, 1962). As the burden of unsolved problems grows, ultimately a new paradigm arises and focuses major shifts in thinking and even changes in institutions. While a new paradigm works well at first, it is likely to work less and less well over time until a new paradigm is needed This model describes our food safety system today.
The recognition that foodborne illnesses and deaths cost this country billions of dollars a year coincides with an apparent lack of public trust in government and gives rise to the suggestion that government is the problem, not a solution. That is a disconcerting depiction and, in one important respect, it is inaccurate. Officials who direct or carry out diverse functions under the multiplicity of statutory mandates are capable and dedicated, as are their state and local counterparts. They perform remarkably well, given their budgetary and statutory constraints, but they operate within an institutional framework that is out of date and poorly designed to accomplish the critical goals that regulation in this field must achieve. The increasing complexity of food production and delivery and the exploding internationalization of the US food supply impose added pressure on the federal regulatory apparatus which was constructed in simpler times.
Given the challenges, the US food safety system has several strengths. Problems are addressed at many points from production to consumption and from many different perspectives (for example, federal or state, public or private, top-down or investigator-initiated, basic or applied, and combined or separated from regulatory pressures). There are significant efforts to improve the current structure through implementation of systems with multiple critical control points to address hazards and increase safety. There is a shared sense of urgency and commitment arising from the current national emphasis on food safety which has resulted in extensive communication and coordination efforts throughout the system (interagency, agency and industry, and state and federal).
There is also substantial private funding in support of food safety research. In spite of these strengths, the current food safety system that the committee studied displays several fundamental weaknesses, which are described on the following pages.
Inadequate Application of Science
Rapid changes in the US food system and spectacular recent scientific advances have only slowly affected food safety policies and regulations. The US Department of Agriculture (USDA), the Food and Drug Administration (FDA), the Environmental Protection Agency (EPA), and the National Institutes of Health (NIH) will collectively spend an estimated $170.8 million on food safety research in 1998 (Appendix E), but there is minimal coordination and little integration of their efforts. The absence of a nationally coordinated research agenda for food safety raises serious concerns about duplication of effort and about the linkage of science to attempt to solve food safety problems of the highest priority.
In some federal departments, such as the USDA, authorities for research and regulation are separated and assigned to different agencies. In other agencies, such as the FDA and the EPA, the regulatory agencies also have authority for research. Separation of research from the constantly changing pressures and priorities of the regulatory environment facilitates science-driven research that requires long-term commitments. The separation is also appropriate for research targeted principally at producers, processors, educators, consumers, and non-regulatory scientists. But combining research and regulation in the same agency facilitates implementation of the research and improves focus on the highest-priority regulatory needs. Such a combination also strengthens the science and technology base of the regulatory agency. However, in spite of growing national concern about foodborne illnesses, both research and its scientific base for the regulatory programs in the FDA were slowly eroding prior to the Food Safety from Farm to Table: A National Food Safety Initiative (Appendix C); this trend appears to be changing with the current actual and proposed increased funding for food safety research (Appendix E).
Both rapid scientific response and long-term research are needed to adequately address food safety issues. Although USDA's Agricultural Research Service (ARS) has the organizational structure to respond quickly, it might not have the scientific expertise or other resources required for a specific research need of USDA's Food Safety and Inspection Service (FSIS). It might not be willing (or in some instances authorized) to redirect resources to an FSIS need if redirection would disrupt its long-term research programs. In short, ARS does not meet all the research needs of FSIS. FSIS is working with other agencies (USDA's Cooperative State Research, Education and Extension Service [CSREES], NIH, and FDA) to persuade them to target more of their research
programs specifically to the FSIS agenda. Communication and coordination of the CSREES research agenda with FSIS does not appear to be well-developed even though both agencies are in the same department.
Research Funding Levels
The committee conducted a public forum in which spokespersons representing several industry and consumer groups cited the need for adequate federal funding of food safety research, and two former FSIS administrators emphasized the need for increased research funding throughout the federal system (Appendix D). President Clinton's fiscal year 1999 budget proposes increased funding for food safety research as part of the National Food Safety Initiative. There is value in having research programs, some of which are responsive to regulatory activities and some to longer term or differently targeted research programs, protected from short-term regulatory pressures. The nation needs epidemiological research on human pathogens on the farm, but the attempts of FSIS to obtain this information are limited by statutory and funding restrictions. Other agencies can do even less at the producer level. Concerns about the adequacy of research funding can be further exemplified by a broader consideration of the issue.
Most federal research efforts directed toward human-pathogen control at the farm-level are in ARS. In 1997, ARS had 18 research projects involving 51 scientists (excluding graduate and postdoctoral students) who were committed to the control of foodborne pathogens in live animals (Panter and Robens, 1997). The recent development of a product to reduce Salmonella enteritidis in chickens (Macriowski et al., 1997) indicates the scientific quality and relevance of that program. However, in the judgment of the committee, the scale and scope of the ARS program are not commensurate with the urgency of the problem, the potential impact of the preharvest approach, or the knowledge gap and scientific complexity of the issues. Understanding the epidemiology, ecology, and molecular mechanisms involved in the array of pathogens confronted and the control procedures needed at the farm level require a larger investment. Funding for extramural programs is not adequate to fill the gap. For example, the total CSREES funding for all food safety research at the 115 extramural institutions participating in CSREES food safety research programs in Fiscal Year 1998 is $9.9 million (Appendix E). Even though those funds are leveraged by state and private investments, they fall far short of the funding required for optimal use of the research resources of the participating institutions in solving the nation's food safety problems. A 1994 National Research Council report emphasizes the need for increased funding of food safety research through the CSREES (NRC, 1994). The relatively modest increases in CSREES funding for food safety research in the last two years were greatly needed but are not adequate for national needs.
Recent Efforts to Improve Research
In an effort to address a portion of the scientific inadequacies of the system, the President directed the Secretaries of the Department of Health and Human Services and the USDA to create a Joint Institute for Food Safety Research that would: (1) develop a strategic plan for conducting food safety research activities consistent with the National Food-Safety Initiative; and (2) efficiently coordinate all federal food safety research, including research conducted with the private sector and academe (Office of the President, 1998). The principal goals of this joint institute is to develop the means to identify foodborne hazards more rapidly and accurately, and to develop effective interventions to prevent food contamination at each step from farm to table. While these actions are significant steps toward improving food safety research, there remain major shortfalls in the application of science in the current system, which include:
- There is no nationally coordinated scientific research agenda among all agencies involved in food safety that stems from a unified mission or centrally focused leadership.
- There is a lack of adequate integration of research efforts among agencies.
- Resources currently identified for research are inadequate to support a science-based system.
Inadequate Use of Risk Assessment
It is widely recognized that eating food entails an inherent risk of illness. The risk of acquiring foodborne illness can vary widely and depends on the type of food and how the food is processed, handled, and prepared. Some foods, such as commercial sterile, retorted canned products, present a very low risk of transmitting foodborne pathogens; others, such as raw oysters, have a well-documented history of disseminating foodborne disease.
Risk assessment determines the probability of illness caused by eating food contaminated with specific foodborne hazards. Critical information needed for risk assessment includes identification of the hazardous agent, data on the prevalence and concentrations of the agent in specific foods, profiles of the consumption of specific foods, and the disease response of people who are exposed to different amounts of the harmful agent. Those data are used in a mathematical calculation to estimate the risk of illness to a specific category of consumers that is caused by a harmful agent in a specific type of food. However, quantitative microbial risk assessment is a new discipline that is in the developmental stages, and refinements are needed before it can be fully implemented.
The limited availability of resources to address food safety issues necessitates that priorities among safety programs be set on the basis of risk
assessment. That approach to assessing the relative safety of different foods enables regulators to estimate the probability of acquiring illness from eating specific foods and thereby allows them to place the greatest emphasis on foods that have the highest risk of causing human illness. Hence, risk assessment is a science-based approach to addressing food safety issues. It is not, however, to be restrictive; dealing with several small risks may be more effective and less expensive than efforts to eliminate a large but intractable problem.
The major shortfall with regard to the use of risk assessment in the current system includes:
- Under the current statutory and budgetary constraints, it is not possible to fully realize the benefits of the valuable and critical tool of risk assessment and its resulting positive impacts on food safety.
A glaring weakness of the current US food safety system is a lack of information. This gap impedes our identification, description, and elimination of problems. There is a lack of knowledge about the seriousness, incidence, and cost of foodborne disease. There is inadequate information on the type of foods that people eat in various regions of the United States, the size of portions, how they are prepared, and how often they are eaten. Information is lacking on the association of pathogens with food animals, and on the prevalence of chemical and physical hazards. Methods and critical control points for reducing or eliminating pathogens in different parts of the food system, especially in production, are insufficient.
The paucity of information to link key players in the food system results in few feedback systems for continuous evaluation and improvement. For instance, mechanisms are lacking for tracing a diseased animal back to point of production on the farm; this impedes objective assessment of the current situation and prevents the ability to reduce the likelihood of future incidents. The committee could not assess with confidence the severity and impact of on farm food-related hazards, because the risk could not be quantified. This lack of information also impedes the direction of new efforts and the measurement of progress associated with changes in private-sector conduct or government policy.
Current understanding of the magnitude of the problem of foodborne disease and the relative importance of the relevant hazards is inadequate and in many cases inaccurate. Furthermore, the scientific resources and structure are lacking to address the inadequacies and inaccuracies. Effective and adequate monitoring, surveillance, and research to characterize risk are required to improve the allocation of resources and to develop the knowledge and technology needed to manage hazards that pose the greatest risk.
The major shortfall regarding available information in the current system can be summarized as:
- There is a serious lack of information to allow appropriate characterization, assessment, and response to the problem of foodborne hazards in the United States.
HACCP Systems and Their Limitations
Federal food safety agencies are shifting to a new conceptual framework based on hazard identification and process control-the science-based Hazard Analysis Critical Control Point (HACCP) program for the prevention of pathogen dissemination. However, for this program to be successful, government must find new ways to deploy its financial and personnel resources. There is serious concern about the lack of sufficient resources for USDA to provide timely implementation of HACCP-based inspection in processing facilities. Most federal inspection resources are committed by statute to carcass-by-carcass inspection of meat and poultry. Current resources are not adequate to meet statutory requirements and to train and hire the federal regulatory personnel required to implement HACCP programs fully throughout the food safety system.
Implementation of HACCP systems throughout the food continuum is a major step toward enhancing food safety, but HACCP systems should not be thought of as preventing all human illnesses resulting from consumption of potentially hazardous foods. Some potentially hazardous foods are produced and processed with HACCP systems that do not involve a treatment that will kill pathogens. The control points used in processing foods like fresh meat and poultry might not kill harmful bacteria, such as Salmonella, but only prevent their growth; refrigeration could be an example of such a control point. Foods with pathogens that have been controlled or reduced, but not eliminated, could still cause human illness. Hence, although HACCP systems properly implemented throughout the food continuum should greatly reduce the incidence of foodborne illnesses, vehicles for contamination and disease may still remain.
The major shortfalls regarding HACCP systems include:
- Under current statutory and budgetary constraints, the benefits of HACCP systems in enhancing the safety of foods cannot be fully realized.
- HACCP systems can minimize risk but cannot assure absolute safety of potentially hazardous foods that do not receive a treatment to destroy harmful microorganisms.
Absence of Focused Leadership
No single federal official can be said to be responsible for the government's food safety efforts. Instead, several officials have responsibility for parts of the system that are organizationally separate and individually funded. Many of the separate programs have other responsibilities as well. They are in different parts of the executive branch and they report to different congressional oversight and appropriation committees. Sometimes they compete for resources and for public attention. None of the heads of these agencies has direct access to the White House, and several report through more than one administrative level. Food Safety from Farm to Table: A National Food Safety Initiative (Appendix C) has given more prominence to the federal government's role generally, but it has not fundamentally or permanently altered the underlying balkanized structure.
Fulfillment of the federal role in protecting the food supply requires central management of now-dispersed efforts. Central management is essential if resources are to be allocated in accord with science-based assessments of risk and potential benefit. It is necessary to assume cooperation among dispersed, sometimes competing, programs. It is important for coordination among the states and between the states and localities and the federal government. It is also the only way to ensure that a focused federal entity is responsible for food safety policy.
The major shortfalls regarding leadership in the current system include:
- There is no single federal entity that is both responsible for the government's efforts and that has the authority to implement policy and designate resources toward food safety activities.
- There is a lack of a unified mission among the various agencies with regard to food safety.
There is no ''food safety" counterpart of such modern federal regulatory laws as the Clean Air Act or the Occupational Safety and Health Act, and there is no comparable legal framework for federal food safety activities. Primary authority for food safety is bifurcated between the USDA and the FDA, and other responsibilities are more widely dispersed. There is no formal structure for coordinating or empowering state regulatory bodies (which in many states are themselves divided along USDA-FDA lines). Federal support for and administration of relevant scientific research are even more widely dispersed and uncoordinated.
Some of the numerous reasons for this state of affairs are obvious. The most important date from the beginning of the century, when Congress made two choices that decisively influence federal regulation to this day. First, it assigned responsibility for ensuring the safety of meat to one agency and responsibility
for other foods to another. That division of authority would have impeded coherent federal regulation even if the threats to food safety had remained unchanged. But they have not remained unchanged, and the result has been to make unmistakably clear the misallocation of regulatory effort that is the product of Congress' second critical decision—its current directive, embedded in statute, that each meat (and later poultry) carcass must be subject to physical inspection. That decision although it may have been appropriate for the hazards present 70 years ago, impedes FSIS efforts to allocate its substantial regulatory resources in ways that correspond to the health risks presented by contemporary sources of food or modern means of food production and processing. In short, the hazards of greatest concern today are microbiological and chemical contamination; and they are not detectable with the traditional inspection by look, sound, smell, and feel. The law's demand has for decades influenced funding of meat and poultry regulation.
The statutory structure impedes coherent, risk-based regulation in other ways. The responsibilities given to FDA emphasize control of the use of chemicals in food production and the addition of chemicals to food. The relevant post-World War II amendments to the Federal Food, Drug, and Cosmetic Act (FDCA) have addressed that apparent threat for conventional foods. The provisions of the FDCA that address food sanitation remain as they were passed 60 years ago, when lawmakers concluded that poor food sanitation poses greater threats.
The discrepancy between regulatory effort and risk is increased because of the way the law's requirements are implemented. Generally speaking, any chemical proposed to be added to food or used in food production must have agency approval before it can be used, unless it is an ingredient in a dietary supplement. The law obligates FDA or EPA to entertain and rule promptly on applications for new food additives. Although the agencies rarely meet statutory deadlines, their legal obligations to rule on new chemicals have led them—and Congress—to allocate major resources to this function and to neglect activities directed toward food sanitation.
Passage of the 1994 Dietary Supplement Health and Education Act provided for the sale of dietary supplements that do not present a significant or unreasonable risk of illness when used as recommended on the label or under ordinary conditions of use. Some consumers believe that if a small amount of these supplements is beneficial, then more will be more beneficial; this poses the risk of adverse health effects due to overconsumption. Current law makes consumer protection against any potential risk difficult.
Neither the Meat Products Inspection Act nor the Poultry Products Inspection Act speaks to how livestock are produced, maintained, or managed. FDA and EPA, respectively, prescribe conditions for food use of animal drugs and feed additives and for pesticides. These conditions are meant to prevent the presence of dangerous amounts of those chemicals in food. However, monitoring of compliance with approved usage is poorly funded and episodic.
State and local authorities have more to say about on-farm practices, but their monitoring capabilities are severely limited.
Most food processors operate in interstate commerce and therefore are subject to FDA standards and inspection. But, as has been noted, the FDCA does not require food processors to register with the agency, so FDA's knowledge of its universe is incomplete.
FDA's shrunken inspection force is seriously over-extended, and FDA appears to have insufficient resources to meet its statutory obligations. For example, numbers of FDA food inspections are declining: premises regulated by The Center for Food Safety and Applied Nutrition are inspected on the average once every 10 years (GAO, 1994b). New technologies, products, and processes continue to need timely regulatory review and decision. As technical people retire or leave the agency, resources must be adequate to attract highly qualified people to fill these public service roles. User fees have been suggested and rejected as a fiscal solution. In the meantime, such reform efforts as the FDA Modernization Act of 1997 increase FDA's workload, at least for the near term. Congress must provide appropriate resources for the tasks demanded of FDA.
Another growing segment of the food industry largely escapes federal oversight: food service establishments—including full-service restaurants, fast-food establishments, grocery store delicatessens, and sidewalk food vendors—whose sales together now account for over 45 percent of the American food dollar (Putnam and Allshouse, 1997). The materials and ingredients that the establishments purchase are theoretically subject to federal regulation, but their food preparation activities are not. It is a matter of debate whether federal law potentially applies to these activities, but it is clear that federal officials leave regulation of them to the states and localities, where monitoring and enforcement are vulnerable to budgetary pressures, jurisdictional disputes, and diverse legal standards.
The major statutory shortfall of the current system is that:
- There are inconsistent, uneven, and at times archaic food statutes that inhibit use of science-based decision-making in activities related to food safety, and these statutes can be inconsistently interpreted and enforced among agencies.
Lack of Coordination
A lack of coordination on several levels seems to be one effect of the lack of strong focused leadership and the lack of a unified mission. The lack of coordination has resulted in a lack of national standards and a lack of focus on food safety. There appear to be no mechanisms to sustain expanding interagency coordination after the current national concern abates and the attention of Congress, the President, and agency leadership is directed to other issues.
Several examples of coordination deficiencies include:
Lack of federal agency coordination. Surveillance information is ultimately communicated among the agencies and organizations involved, but there is (except for the recent creation of FoodNet) no integration of the various programs included in the current structure. Neither routine surveillance programs, special projects, nor emerging issues are addressed in a coordinated interagency manner. There is no comprehensive national strategy or system for surveillance. Human and animal studies and analyses of foods are, for the most part, conducted independently without a common goal or design, even though they may impact the same food safety issues.
Another example is the lack of coordination between FDA and USDA regarding the regulation and clearance of packaging materials. For instance, catalyst systems used to make polymers do not appear in the FDA food additive regulations because they are understood to be proprietary information. However, USDA now requires companies to file a food additive petition with FDA for catalyst systems.
Lack of federal and state coordination. Federal, state, and local authorities must work with varied amounts of resources, skills, and legal authority. Lack of coordination and consistency between federal and state governments is problematic. Some states have initiatives requiring more stringent standards than those required by the federal government. Under California's Proposition 65, warnings may be required for products, including foods, that are not required under federal law. Another example of a lack of federal and state coordination is that food retailers with stores in multiple states must deal with many regulatory entities at the federal, state, and local levels. The committee heard testimony from the Food Marketing Institute (FMI) indicating that one food retailer with stores in several states must report to 88 different regulatory authorities. (Jill Hollingsworth, FMI, personal communication to committee, April 1998; Appendix D). These conflicting requirements create an additional burden on industry and may confuse consumers.
Lack of public and private coordination. As described by the United Food and Commercial Workers Union, implementation of HACCP programs in meat packing plants is often required and attempted without allowing time to perform proper cleaning or to conduct effective employee training (Jackie Nowell, United Food and Commercial Workers Union, personal communication to committee, April 1998; Appendix D).
Lack of international coordination. Currently, sampling of imported foods takes place at the port-of-entry (GAO, 1998). This method of inspection does not allow for the timely identification of potential foodborne hazards. There is a need to identify and correct problems at the point where food is produced and processed, and this requires international government coordination and cooperation.
The major shortfalls regarding coordination in the current system include:
- There is a lack of coordination and integration among the at least 12 agencies that are involved in implementing the 35 primary statues that regulate food safety.
- There is often ineffective or insufficient coordination among federal, state, local, and private entities.
Deficiencies in Regulation of Imported Food
Protecting the safety of domestically produced food is a daunting challenge, but the country's growing reliance on imported food adds several layers of complexity. It is by no means clear that imported food, as a class, poses greater risks than does domestically produced food. What is clear is that federal officials cannot use the same methods in regulating imported food that they use-or that would make sense-in regulating domestically produced food. Methods that rely on production-site monitoring of compliance with safety standards or universal physical inspection of marketed shipments cannot be directly translated overseas.
The laws that FSIS and FDA administer require that imported food meet the same standards as domestic food. But the enforcement approaches of the two agencies to meet this common requirement are quite different. FSIS statutory authority requires meat and poultry food safety systems of exporting countries to be equivalent to the US system (GAO, 1998). FDA lacks the authority to require that imported foods be produced under a system equivalent to the one that it administers domestically; instead, FDA relies primarily on sampling at ports-of-entry to determine whether food imports meet domestic requirements (GAO, 1998). Even if FDA's criteria for sampling and testing were systematically risk-based and its resources were adequate to keep up with a growing demand, sample analysis is not capable of detecting many of the most serious risks to consumer health.
In fact, although both agencies have computerized systems to assist in inspection and tracking, there is no way to determine whether the agencies are focusing their attention on the most important health risks. Both agencies target resources to meet the problems of past violations, in which contamination, processing defects, labeling, and quality were at issue.
The General Accounting Office has reported that FDA lacks the necessary controls over detained and suspect shipments (GAO, 1998). Unscrupulous importers are able to circumvent the system, and are seldom punished in proportion to the seriousness of their violations. Similar concerns center on fish and shellfish inspection as over 50 percent of the fish and shellfish consumed in the United Sates is imported (GAO, 1998). It has been reported that shellfish
alone caused 21 percent of all reported foodborne illnesses from 1978 to 1992 (CDC, personal communication to committee, June 1998).
In an effort to address the challenges of ensuring the safety of imported foods, the President has proposed a variety of measures including hiring additional FDA inspectors to examine the safety of US fruits and vegetables, both domestic and imported. In addition, legislation is being proposed to allow FDA to halt imports of fruits, vegetables, and other food products that do not meet US food safety requirements or that do not provide the same level of protection as is required for US products. Recognizing that sample analysis does not provide a means for detecting many of the most serious risks to consumer health, and without firm knowledge of most significant risks, it is impossible to know whether the proposed actions will adequately address imported food hazards.
Summary Findings: Where the US Food Safety System Falls Short
- Inconsistent, uneven and at times archaic food statutes that inhibit use of science-based decision-making in activities related to food safety, including imported foods;
- a lack of adequate integration among the 12 primary federal agencies that are involved in implementing the 35 primary statutes that regulate food safety;
- inadequate integration of federal programs and activities with state and local activities;
- absence of focused leadership: no single federal entity is both responsible for the government's efforts and given the authority to implement policy and designate resources toward food safety activities;
- lack of similar missions with regard to food safety of the various agencies reviewed;
- inadequate emphasis on surveillance necessary to provide timely information on current and potential foodborne hazards;
- resources currently identified for research and surveillance are inadequate to support a science-based system;
- limited consumer knowledge, which does not appear to have much impact on food-handling behavior; and
- lack of nationwide adherence to appropriate minimum standards.