Regulatory Comments
Submitted by the National Restaurant Association
To: Food and Drug Administration
Date: July 17, 2001
Topic: Listeria Risk Among Certain Ready-to-Eat Foods

July 17, 2001

Dockets Management Branch, HFA-305
Docket No. 99N-1168
Food and Drug Administration
5630 Fishers Lane, Room 1060
Rockville, MD 20852

Dear Sir or Madam:

The National Restaurant Association appreciates the opportunity in this written statement to submit its views on the Food Safety and Inspection Service's (FSIS) and the Food and Drug Administration's (FDA) "Draft Assessment of the Relative Risk to Public Health from Foodborne Listeria monocytogenes Among Selected Categories of Ready-to-Eat Foods" and a proposed risk management action plan for L. monocytogenes as referenced in the Federal Register on January 19, 2001. As the leaders of the hospitality industry, we have had a long-standing commitment to food safety and the protection of our customers, and are deeply interested in further improvements in the nation's food supply.

Founded in 1919, the National Restaurant Association is the leading business authority for the nation's $408 Billion restaurant industry, which has more than 945,000 restaurant locations. Our 46,000 members represent more than 235,500 individual fullservice restaurants, quickservice units and cafeterias, institutions, hospitals, universities and military clubs. We have made multibillion dollar investments in improving restaurant food safety and developing state-of-the-art food safety education programs during the past 80 years. In cooperation with state and local health officials, the National Restaurant Association has been a partner in the development of numerous state food safety regulations, food safety educational programs and informational materials based upon industry research, current science and Hazard Analysis Critical Control Point (HACCP).

We would like to commend the FSIS and the FDA for their efforts to solicit comments from the industry and the public on the very important issue of Listeria monocytogenes.
Listeria is a foodborne illness and responsible actions should be taken to reduce the incidences of Listeria infections in consumers' homes and at the retail level. However, we believe that careful consideration should be given before taking possibly costly, ineffective and counterproductive actions. For this reason, the National Restaurant Association believes that the current Listeria control measures developed by the processing and manufacturing industries should be fully supported by FDA and FSIS in order to support industry-directed improvements from the start.

Foodborne listeriosis has been recognized for decades. The Listeria problem gained prominence in the 1980's after several large foodborne illness outbreaks, and again regained recognition with a large multistate outbreak in late 1999. As a result of this attention, many segments of the food industry have implemented Listeria intervention strategies. In fact, because of the major efforts put forth by industry, during the 1990's, we saw a 50% reduction in the incidence of listeriosis. Clearly, the striking 50% reduction is indicative that intervention strategies incorporated at various steps by the industry and local officials worked well.

The control of Listeria monocytogenes throughout the food chain from farm-to-fork must incorporate a broad-based multipronged approach that begins with intervention at the production level. A further reduction in the level of Listeria contaminated products and corresponding incidences of Listeria cannot be expected unless introduction of the microbe at the production level is severely restricted. Consequently, such production-level restriction practices must include principles based on sanitary standard operating procedures (SSOPs), good agricultural practices/good manufacturing practices (GAPs/GMPs), and/or other interventions that may be developed. The National Restaurant Association believes that voluntary SSOPs in conjunction with GAPs/GMPs will present the most feasible intervention strategy for all producers, regardless of size.

We believe that the draft risk assessment developed by the FDA and the FSIS inaccurately characterizes the risk to public health due to L. monocytogenes in restaurants and other retail establishments. The assessment speculates that the move to increased consumption of foods prepared outside the home or ready-to-eat foods has slowed the further reduction of foodborne listeriosis. It further notes that foods bought already prepared from retail establishments, grocery stores, and deli counters are of concern because "adequate food safety measures may not be in place to control or prevent L. monocytogenes contamination." However, if one reviews the current epidemiological data related to foodborne illness outbreaks attributed to listeriosis, the data clearly show that the majority of cases do not appear to be part of recognized foodborne illness outbreaks. Therefore, they most likely are not foodservice related.

Outbreaks associated with retail establishments and restaurants have most often occurred because foods arrived at the facilities already contaminated during manufacturing and processing. For example, outbreaks of listeriosis have been linked to failure to protect a frankfurter processing line from environmental contamination, use of defective processing equipment in the production of chocolate milk, and inadequate pasteurization of milk used to make cheese. Yet, the document is "primarily based on contaminated foods at retail." Directing the risk assessment to only those foods contaminated at the retail level to the exclusion of other venues such as manufacturing and processing plants is wholly inappropriate and misleading.

Additionally, the assessment omits review of homes, daycare centers, schools, and other places where food is served or prepared outside of retail. While it is understandable that these types of environments are a challenge to study, and no real data may exist on Listeria illnesses resulting from home or other exposures, it is unacceptable to assume that they do not need to be incorporated into the assessment.

Assumptions and Surrogate Data Sources
The assessment attempts to describe the current state of scientific knowledge related to human listeriosis associated with ready-to-eat foods. However, repeatedly, the document acknowledges that there were many instances where data were lacking or were "less than ideal." For example, foods were grouped into 20 categories in part because of "lack of adequate contamination and growth rate data to model the specific contamination concentration distributions of individual foods at consumption." Additionally, storage times for foods purchased from retail establishments and held in the home were estimated "because no specific data were available." In another instance when data was lacking on the rates of symptomatic versus asymptomatic L. monocytogenes infections in humans, "a combination of epidemiological data and data acquired from studies using animal models," were assessed. Such multiple speculative assumptions do not constitute a valid scientific basis for the conclusion that Listeria monocytogenes contamination is a significant risk in retail establishments.

Exposure Assessment
The purpose of the exposure assessment was to "estimate how often consumers eat food contaminated with L. monocytogenes and the number of bacteria likely to be in that food." But, unlike other risk assessments, this particular document does not attempt to consider how a food became contaminated. The reason cited for this omission is that the contamination data used were primarily from food samples collected in the marketplace. Since the overwhelming majority of outbreaks are associated with processing or manufacturing contamination, we strongly believe that this method of assessment was misguided. The exposure assessment further assumes that the food categories contribute to listeriosis through direct consumption. Epidemiological data suggest that cross contamination could play a major role in transmission of the disease, but because data were lacking, this possibility was not considered. To meet the purpose of the exposure assessment, it is critical that all possible types of contamination be considered. Without doing so, the degree of certainty of the exposure assessment is severely compromised.

General Comments
This risk assessment was the basis for the joint response plan developed by the FDA and the FSIS to further reduce illnesses caused by foodborne Listeria in ready-to-eat foods that are intended to be eaten as purchased. The plan denotes eight action steps. These steps focus on those foods and facilities identified in the risk assessment as warranting additional control measures or data. It is clear that the risk assessment was severely lacking data. For this reason, we strongly believe that several of the action steps are unwarranted. For example, action item 5 addresses new regulations and revisions to existing regulations. The FDA plans to reduce the storage temperature for certain ready-to-eat foods based on the risk assessment which was done with little data coordinated to actual illness and inappropriately focused on retail establishments. Hence, it seems that solutions are being derived without fully identifying any real problems. In fact, the risk assessment recognizes that "the nature of the data and the structure of the models used in this risk assessment lead to conclusions with varying degrees of certainty."

As noted in the assessment, "the accumulation of additional data from research and surveillance, as well as industry-collected data, could potentially reduce the uncertainty associated with the risk assessments predictions." Moving forward, the Agencies must reassess the document and work to fill in the knowledge-gaps that currently exist to make the document viable. Furthermore, the assessment must be redirected to focus on those venues most associated with Listeria contamination that has been correlated to foodborne illnesses. Until that time, an acceptable, scientifically justified response plan cannot be developed.

Specific Questions
The draft assessment requests in the preface that commenters provide responses to several specific questions. The following are the National Restaurant Association's recommendations concerning several of the questions.

1a. Noting that foods were grouped into 20 categories in part because of lack of adequate contamination and growth rate data to model the specific contamination concentration distributions of individual foods at consumption, the agencies are asking if the grouping is appropriate for modeling purposes. We believe that the groupings are, in fact, inappropriate. The grouping leaves a large degree of variability and uncertainty around the model's predictions as noted in the assessment. The assessment must center on those foods most associated with Listeria monocytogenes that have led to foodborne illnesses, not food categories. This information will be the most useful when focusing on control strategies as the controls will be directly related to foodborne illness.

1d. Storage times for foods in homes before consumption were estimated because no specific data were available and the Agencies are asking if this data should be included. We believe it is critical that data on storage times be included. Simply estimating does not provide a true reflection of what is occurring during home storage of ready-to-eat foods. Accurate data describing the actual holding times and temperatures is an essential piece. As noted, it is essential that these types of data and knowledge gaps are ascertained and filled before a valid assessment can be completed.

2. This question addresses the issue of the current and quantitative contamination data that were used in developing the assessment. The Agencies recognize that the assessment primarily relied on published data. However, through their own admission, there were instances when unpublished data were used to make definitive conclusions. This is unacceptable because it has not been peer reviewed and may not be defensible. All data included in the assessment must be published. Additionally, much of the data was attained from studies more than 10 years old. Over the last decade, the food industry has made great strides in developing effective Listeria controls. Through food safety systems such as HACCP, GMPs and GAPs, we have seen a striking reduction in the number of Listeria illnesses attributed to foods. As such, we strongly believe that current data must be used as the basis for the assessment and further control strategies. Again, the ultimate focus must be to fill the gaps in the knowledge that we currently have before pushing forward and devising further unjustified controls.

3c. The dose-response models relied on animal studies to describe the range of variability in L. monocytogenes virulence and host susceptibility. The relationships developed were then anchored to relate the resulting distributions to human surveillance reports. We strongly believe that an alternative approach should be developed for the dose-response models. The risk assessment tells us that there is no quantification of the number of organisms consumed by both ill and exposed subjects during outbreak investigations associated with Listeria. Therefore, extrapolation of data from animal surrogates (particularly mice) to humans was incorporated into the risk assessment.

The extreme differences between mice and humans make it virtually impossible to adequately characterize human dose requirements that correlate to illness resulting from ingestion of L. monocytogenes. The assessment acknowledges this fact and notes that several factors were overlooked, such as gastrointestinal endpoint of listeriosis in humans, because of the inherent differences. Furthermore, key assumptions and assertions were made when extrapolating animal data to humans. The Agencies must do further studies into human cases of Listeria monocytogenes to develop defensible dose-response data that can be used in developing Listeria control measures.

The National Restaurant Association welcomes increased cooperative efforts to address Listeria monocytogenes involving industry, retail, academia, and government officials and feels that revised actions should be justified and based on appropriate data warranting such changes. We appreciate the opportunity to comment on this draft risk assessment and action plan. Should you have any questions, please feel free to contact the Health and Safety Regulatory Affairs Department at (202) 973-5384.

Sincerely,

Steven F. Grover, R.E.H.S.
Vice President, Health and Safety Regulatory Affairs

Cc:
Steven C. Anderson, President and Chief Executive Officer
Lee Culpepper, Senior Vice President of Government Affairs and Public Policy
Gay Westbrook, Legislative Representative