Redesigning Food Safety

Controversy over genetically modified foods has helped put food safety in the headlines, but that issue, like others we read about–mad cow disease, Listeria and Salmonella outbreaks, chemical contamination–needs to be understood and addressed in the broader context of how we protect consumers from all foodborne hazards. This broader perspective is obscured, however, by the fragmented and in many ways outdated legal and organizational framework for food safety in the United States. Food safety law is a patchwork of many enactments that, all told, lack a coherent, science-based mandate for regulators and that split food jurisdiction among a dozen or more agencies, most prominently the Food and Drug Administration (FDA), the Department of Agriculture (USDA), and the Environmental Protection Agency (EPA).

The potential impact of this framework on the safety of biotech foods is important, but there is a broader and more fundamental public health question about the effectiveness of the current system in protecting consumers from foodborne illness. The Centers for Disease Control and Prevention (CDC) recently issued new, more reliable estimates of the persistently high incidence of foodborne illness in the United States: an estimated 5,000 deaths, 325,000 hospitalizations, and 76,000,000 illnesses annually, most of which are preventable.

In 1998, an Institute of Medicine/National Research Council (IOM/NRC) committee studied the current framework and called for a comprehensive statutory and organizational redesign of the federal food safety system. In its report, Ensuring Safe Food from Production to Consumption, the committee documented how the century-old accumulation of food safety laws and fragmented agency structure are impeding the efforts of regulators to reduce the risk of foodborne illness. The committee recommended a science-based, integrated food safety regulatory system under unified and accountable leadership; a system that would be better able to deploy resources in the manner most likely to reduce risk.

The IOM/NRC recommendations make common sense, but this does not mean that they will be readily adopted. The statutory and organizational status quo in Washington is politically difficult to change, which is why most major reforms in public health and environmental laws have occurred in response to some galvanizing event or crisis. Fortunately for current health, if not policy for the future, the U.S. food safety system is not in crisis. It remains, in many respects, the strongest in the world, and it has made important strides in recent years toward more effective regulatory policies that properly emphasize preventive process control to reduce significant hazards.

The food safety system is, however, under serious stress, largely because of rapid change in the food system. Many of the cases of foodborne illness reported by the CDC are linked to new and emerging microbial pathogens, changing U.S. eating habits, and an aging population. The system is also challenged by new agricultural and food technologies, such as genetically engineered food crops; by an increasingly globalized food supply, which makes European and Latin American food safety problems potential problems for the United States; and by intense public and media scrutiny of issues such as mad cow disease and biotech foods. Regrettably, chronically strained food safety budgets have seriously eroded the government’s scientific staffing and inspection resources even as the food safety job has become more difficult.

In response to these stresses, and with an eye on lessons learned in Europe concerning the fragility of public confidence in food safety, U.S. lawmakers and nongovernmental organizations are showing growing interest in modernizing our food safety laws and structures along the lines contemplated by the IOM/NRC committee. Consumer groups that have been pushing for such reform have recently been joined by some food industry associations and scientific organizations. On Capitol Hill, Sens. Richard J. Durbin (D-Ill.) and George Voinovich (R-Ohio) recently wrote to President Bush calling for a bipartisan effort to combine the food safety functions of the FDA, the USDA, and the EPA into a single food safety agency. The Senate Agriculture Committee is also showing interest in the subject, with its chairman, Sen. Tom Harkin (D-Iowa), supporting the single agency concept.

The most compelling reason to modernize the food safety laws and unify the agencies is to allow, indeed mandate, science-based deployment of the government’s food safety resources in the manner most likely to contribute to reducing foodborne illness. This means, among other things, prioritizing the opportunities for reducing risk by means of government intervention.

The government’s role

The overarching purpose of food safety regulation and other government food safety interventions is to minimize the risk of foodborne illness. An effective food safety system provides an array of other important social and economic benefits, including maintenance of public confidence in the safety of the food supply and support for the export of U.S. food and agricultural products, but these benefits flow from success in minimizing food safety risk. The core public expectation, put simply, is that those involved in producing food and overseeing food safety are doing every0

.thing reasonably possible to make the food safe.

Food safety is first and foremost the responsibility of food producers, processors, and others throughout the food chain, including consumers. The government obviously does not produce food and cannot, by itself, make food safe or unsafe. The government does, however, play two important roles in the effort to minimize food safety risk.

The first and broadest role is to set and enforce food safety standards through laws, regulations, inspections, and compliance actions. Such standards range from general statutory prohibitions of adulterated food to specific limits on permissible levels of various chemical residues in food. Most of the government’s food safety resources are devoted to setting and enforcing these standards, with the majority of those resources going to food inspection. This role fulfills the uniquely governmental function of ensuring that commercial firms involved in the food system have accountability to the public for meeting basic food safety standards. The USDA’s recently adopted Hazard Analysis and Critical Control Points (HACCP) system for meat and poultry plants is an example of a food safety standard that has had measurable benefits in reducing harmful contamination and the risk of foodborne illness.

The government’s second role in minimizing food safety risk is to mount initiatives to tackle food safety problems that are beyond the control of any individual participant in the food chain and that require more than a regulatory solution. For example, the pathogen E. coli 0157:H7, which poses a significant hazard when present in any raw or undercooked food, originates primarily in the gut of cattle and is spread via manure through the environment to contaminate water and fresh produce. Through other pathways, it also contaminates beef during the slaughter process. Tackling this and many other food safety problems requires a strong research base; development of effective control measures; and collaboration among growers, animal producers, food processors, retailers, and consumers. The government has an essential leadership role to play in fostering research and collaboration on such issues.

Opportunities to reduce risk

In both of its primary roles, the government has substantial opportunities to improve performance through a more risk-based allocation of its food safety resources. The improvement would come from more systematic prioritization of risks and risk reduction opportunities and better allocation of resources in accordance with those opportunities.

Under current law, the FDA is authorized to inspect food establishments but is not required to do so. With about 50,000 processing and storage facilities under FDA’s jurisdiction and with resources to conduct about 15,000 inspections per year, many plants under FDA’s jurisdiction go years without inspection. Even plants rated by the FDA as “high risk” may be inspected only once a year or less. In contrast, the USDA has a statutory mandate to inspect every carcass passing through slaughter establishments and to inspect every meat and poultry processing plant every day, without regard to the relative riskiness of the operations in these plants.

These approaches to inspection, which reflect fundamental differences in statutory mandates and modes of regulation between the FDA and USDA, skew the allocation of resources in ways that may not be optimal for public health and the government’s ability to contribute to risk reduction. For example, USDA’s budget for regulating meat and poultry is about $800 million per year. FDA’s budget for all the rest of the food supply is less than $300 million. USDA employs about 7,600 meat and poultry inspectors, whereas the FDA has a total field staff of 1,700 for all of its food programs, including inspectors, laboratory technicians, and administrative staff. This is despite the fact that there are more reported cases and outbreaks of foodborne illness associated with FDA-regulated products than with USDA-regulated products. About 3,000 USDA inspectors are assigned to the statutorily mandated carcass-by-carcass inspection program in poultry plants alone, a largely visual process that primarily serves to address product quality rather than food safety concerns and thus makes a fairly minor contribution to food safety. Yet this poultry slaughter inspection program costs about $200 million per year.

The potential to improve this situation through risk-based priority setting and resource allocation is apparent. According to the IOM/NRC report, the agencies should be free to allocate their inspection and other resources across the entire food supply to “maximize effectiveness,” which requires “identification of the greatest public health needs through surveillance and risk analysis.”

Within the existing statutory framework, USDA has some limited flexibility to adjust its inspection models, so potentially it could redeploy resources to reduce risk more directly, such as through enforcement of HACCP and pathogen-reduction performance standards as well as oversight of distribution, storage, and retail facilities. The FDA legally has complete discretion to allocate its resources as it sees fit. Both agencies are making an effort to consider risk in making resource allocations. For example, USDA is developing new inspection models that would permit redeployment of some of its resources to oversee higher risk activities, and the FDA has traditionally attempted to target its limited inspection resources on plants that it judges to be high risk or likely to be committing safety violations.

Both agencies are severely constrained, however, by the current system. In USDA’s case, the statutory inspection mandate commits most of the available resources to activities that are not planned primarily around risk. The FDA’s food safety program is so severely underfunded that it cannot even afford to analyze risk priorities systematically. Thus, as things stand today, neither agency is able to establish risk-based priorities for its inspection program or allocate resources accordingly. For these and other reasons, the IOM/NRC committee recommended that Congress change the law so that resources could be allocated and inspection and enforcement could be based on “scientifically supportable risks to public health.”

The government can also be more effective in reducing risk by setting risk-based priorities for its initiatives that go beyond the core function of establishing and enforcing basic food safety standards. Such initiatives could include research, collaborative efforts with the food industry, targeted regulatory interventions, and consumer education. These efforts require significant money, staff time, and management attention, but they are necessary to bring about the change in practices and behavior that are required to reduce the risk of foodborne illness. In recent years, for example, the FDA and USDA have carried out initiatives to reduce the risk of illness posed by Salmonella enteriditis in eggs. These efforts have resulted in a decline in outbreaks and cases, but only after a significant investment of time and energy.

Risk-based priority setting is critical in deciding which initiatives to pursue and in managing those initiatives. For example, the CDC, through its FoodNet active surveillance program, now reports on cases of illness associated with nine specific bacterial and parasitic pathogens. These pathogens, which are the most significant known sources of foodborne illness, enter the food supply through a range of foods and at different stages of the food production process. If the government is to make the best use of its food safety resources, it should assess and compare the risks posed by various pathogen/food combinations and prioritize opportunities for reducing these risks through targeted food safety initiatives.

Likewise, the presence in food of environmental contaminants, such as mercury, lead, and dioxin, continues to be a matter of public health concern. The government has had success in the past with initiatives to reduce the levels of such contaminants, lead being a notable example. Through risk analysis, the government can identify opportunities for further risk reduction and mount initiatives accordingly.

Improving the role of risk analysis

The statutory, organizational, and resource constraints on risk-based priority-setting and resource allocation would have to be addressed through legislative action. However, there is also much that natural and social scientists can do to improve the risk analysis tools required to design and manage a more risk-based food safety system. These tools include the biological and statistical assessment of particular risks; risk comparison and ranking (in terms of public health significance); and prioritization of risk-reduction opportunities (taking into account feasibility, cost, and social considerations).

In the past, only one component of risk analysis–the risk assessment–has played an important role in food safety regulation, and that was limited to providing the basis for food safety decisions about specific substances. Today, there are much broader roles for risk analysis at the level of system design and management, but this will require improvement in the data and methods available to carry out such analyses.

Comparison and ranking of food safety risks according to public health significance are inherently complicated because of the diversity of risks and health outcomes of concern. Chemical risks range from the acute to the chronic, vary significantly with exposure, sometimes affect age groups differently, and often are predictable only with great uncertainty. Microbiological risks are also diverse, ranging from minor intestinal infections to permanently disabling disease and death, and vary among age groups. But unlike chemical risks, microbiological risk assessments are typically grounded in epidemiological data on actual illnesses in humans. How can these factors be taken into account when comparing and ranking food safety risks? There is a need for public health experts and social scientists to collaborate in developing methods to value risks so that they can be compared and ranked.

The ultimate objective of risk analysis is not risk comparison and ranking for their own sake or to provide the basis for concluding that some food safety risks are unimportant. In the daily activities of people who produce, market, and consume food, any significant risk of harm is important and should be prevented to the extent reasonably possible. For the government, however, the question is how best to allocate finite resources to reduce the risk of foodborne illness. This requires building on risk comparison and ranking to prioritize opportunities for risk reduction. It means not stopping with an understanding of the relative magnitude of food safety risks but examining how the government can make the best use of its resources to reduce risk.

With respect to standard setting and inspection, for example, which segments of the food supply or which specific food/ pathogen combinations pose significant risks that are most amenable to reduction through government intervention? This analysis should start with the magnitude of the risk but also should consider the tools available to government and industry (standards, inspection, testing, new preventive controls) to reduce the risk, the feasibility and cost of reducing the risk in relation to other risk-reduction opportunities, and the value the public places on reducing the risk, as reflected, for example, in willingness to pay to reduce it. With respect to research, education, and other nonregulatory initiatives, where would government interventions have the greatest impact on risk reduction? There is currently no accepted model for considering these and other relevant factors in resource allocation and priority setting for the government’s food safety program. Such a model should be developed.

According to the IOM/NRC committee report, “the cornerstone of a science-based system of food safety is the incorporation of the results of risk analysis into all decisions regarding resource allocation, programmatic priorities, and public education activities.” We agree. Achieving this goal requires statutory and organizational reform, so that the results of risk analysis can be fully implemented in program design and management. It also requires significantly greater investment to improve the data and methods available for risk analysis. With these changes, the regulatory system can most effectively reduce the risk of foodborne illness and, in turn, maintain public confidence in the food supply and preserve our international leadership role on food safety.