Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 62
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment 4 Preparation Module The purpose of the Preparation Module is to estimate the occurrence and extent of E. coli O157:H7 contamination in consumed ground-beef servings. The approach involves determining the frequency of exposure of consumers in different age groups to E. coli O157:H7 in ground beef served at and away from home. Six primary steps are evaluated: grinding of beef, ground-beef storage during processing or by the retailer or distributor; transportation to the home or to hotels, restaurants, and institutions (HRI); storage at home or in HRI; cooking; and consumption. Consumption patterns are modeled as being dependent on the age of the consumer and the location of the meal. Ground beef is consumed in many forms, but the FSIS draft risk assessment focuses on hamburger patties and on ground beef used as a major ingredient in beef-based foods (such as meatballs and meatloaf). The model does not include ground beef as a granulated ingredient (as in commercial meat sauce for spaghetti). CROSS CONTAMINATION A central issue for the committee in its review of the draft Preparation Module was the factoring in of the contributing influence of cross contamination on human illness. Cross contamination during preparation results when E. coli O157:H7 is transmitted from contaminated ground beef to such vehicles as other foods, food preparation and processing surfaces, and food handlers. Because of the highly infectious nature of the pathogen, which has an estimated low infectious dose of under 100 cells, vehicles cross-contaminated through direct or indirect exposure to E. coli
OCR for page 63
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment O157:H7-tainted raw ground beef are likely to be important sources of human illness (Buchanan and Doyle, 1997). A case-control analysis of sporadic infection with E. coli O157:H7 by Mead et al. (1997) substantiates that notion. It determined that most ill persons in question had eaten hamburgers prepared at home and that the primary risk factors associated with infection were food preparers who had not washed their hands or work surfaces after handling raw ground beef. The investigators concluded that in many instances hamburgers were not the direct vehicle of transmission of E. coli O157:H7, but rather that transmission occurred more commonly when the food preparers’ hands, contaminated by raw ground beef, were allowed to cross-contaminate other meal items or utensils. In a multistate outbreak of E. coli O157:H7 infection in 1995, cross contamination from raw ground beef was identified as the likely contributing factor associated with eating cooked ground-beef sandwiches prepared at fast-food restaurants of a specific chain (CDC, 1996). Although they did not address the issue of E. coli O157:H7 directly, two studies released while the draft risk assessment was under development support the notion that cross contamination during food preparation is an important risk factor for foodborne illness in general. Audits International (2001) published a study of food-preparation practices that identified cross contamination (25% of failures) as the third most-common critical violation1 of good hygienic practices in the home. Previous Audits International studies had ranked it as the most common critical violation, with a frequency of 71% in 1997 and 31% in 1999. Another study researching commercial and institutional food operations was prepared by the Food and Drug Administration (FDA, 2000). Researchers at FDA found that 15% of fast-food restaurants and 44% of full-service restaurants examined were out of compliance with one or more items in the category “contaminated equipment/protection from contamination”. Those items included whether raw animal foods were separated from one another, whether raw and ready-to-eat foods were separated, and whether surfaces and utensils were cleaned or sanitized. The Food Safety and Inspection Service (FSIS) of the US Department of Agriculture (USDA) itself identifies cross contamination during preparation as a significant factor in food safety. Two of the four steps in USDA’s Fight BAC!2 campaign—”Clean—wash hands and surfaces often” and “Separate—don’t cross-contaminate”—address interventions intended to minimize it. 1 Critical violations are defined as conditions or actions that by themselves can cause foodborne illness. 2 Where “BAC” refers to bacteria.
OCR for page 64
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment However, the FSIS draft risk assessment indicates that cross contamination in the preparation stage is outside the scope of the analysis (p. 3). Later (p.74), it states: Currently, quantitative modeling of cross-contamination in foods is hampered by a dearth of evidence. Furthermore, cross-contamination pathways are potentially complex, and each pathway may require as much data regarding growth dynamics and cooking effect as the primary product of interest. The model, however, can serve as a starting point for analyzing the effects of cross-contamination on human exposure to E. coli O157:H7. The committee understands and respects the decision of the modelers to establish reasonable bounds on the reach of their work; it is a necessary part of any risk assessment. It observes, however, that cross contamination during preparation is an established, important risk factor; that the lack of data on its effects is no more severe than the lack of data for some other parts of the draft model; and that further attention to cross contamination will help to lay the groundwork for an analysis and better identify the data gaps that need to be filled by future research efforts. The value of the risk assessment in informing public health policy and supporting regulatory interventions will be increased if it is able to factor in the effect of cross contamination on E. coli O157:H7 infections and perhaps address the influence of interventions. Just as important, the committee is concerned that the draft risk assessment may foster the inaccurate and misleading impression that proper cooking of ground beef will prevent all associated E. coli O157:H7 infections. If the model is used to simulate the effects of various interventions on human health outcomes, omission of this major route of infection could produce ambiguous results and potentially deficient policy decisions. The committee recognizes that data are lacking on the extent to which various forms of exposure—whether direct (through contact with contaminated beef itself) or indirect (through contact with meat drippings or with surfaces that have previously been in contact with contaminated drippings or beef)—to E. coli O157:H7-tainted raw ground beef during storage, transportation, and meal-making affect infection. However, that is not the only circumstance in the FSIS draft model in which there is a dearth of information. As noted elsewhere in the chapter, for example, some estimates of the amount of raw ground beef consumed in subpopulations are derived from rather scanty data, and simplifying assumptions or conjectures are used in lieu of data in several steps of the Slaughter Module. The committee also acknowledges that it may not now be possible to model cross contamination at a level of detail that would permit informed
OCR for page 65
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment analysis of the efficacy of specific interventions. However, it points out that the ability to specify the particulars of the myriad scenarios by which cross contamination with raw ground beef can occur is not a prerequisite for accounting for this risk factor in the model. As noted elsewhere in this review, the process of constructing a risk assessment necessarily results in the identification of critical data gaps. With a better understanding of what information would be needed to perform more-sophisticated modeling, USDA will be in a better position to define a research agenda. In summary, disregarding the contribution of cross contamination of E. coli O157:H7-tainted raw ground beef to human illness weakens the draft risk assessment. The committee suggests that consideration be given to factoring in cross contamination as an additional step. If that is not possible, it recommends that the final FSIS risk assessment highlight more clearly the role of cross contamination in E. coli O157:H7 infection and emphasize the limitations in the model engendered by a decision to not factor it in. MODELING IN THE PREPARATION MODULE Data Selected for Use and Means of Analysis Have Weaknesses An important limitation in modeling in the Preparation Module is the paucity of adequate or validated data regarding some components of the preparation steps. It leads to diminished confidence in estimates derived from these data. One example is the modeling of storage times. There are no data that directly document the length of time that ground beef is stored at refrigeration temperature. The draft uses storage temperature data of Audits International (1999) (Table 3-16 in the draft risk assessment) for home and HRI storage (Step 4). However, those data were obtained from a super-market study in which temperatures were monitored from the retail distribution channel into the home, and it is inappropriate to extrapolate them to the whole of the HRI industry, because the vast majority of ground beef distributed through the food-service segment of HRI is distributed frozen. Such ground beef is processed into patties that are transported frozen and cooked from the frozen state. Hence, it is important to recognize that the Audits International data are relevant only to retail products and a minor portion of HRI ground beef. The vast majority of ground beef used in HRI is stored frozen, so the storage-temperature profiles of the product would be much different from those of fresh ground beef stored in a home setting. The FSIS draft risk assessment does attempt to model the effects of freezing ground beef on E. coli O157:H7 cell numbers during storage and
OCR for page 66
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment distribution (p. 83), assuming a uniform distribution of 20–80% of ground beef is produced frozen. That estimate, though, is so broad as to be uninformative. The committee suggests that expert opinion be sought regarding a more precise estimate and distribution and, if it is found useful, that it be documented in the text and used in the model until data become available. Any revised estimate should recognize that most ground-beef products used by the food-service sector are stored frozen. The committee recommends, in general, that more precise information regarding the percentage of ground beef that is stored and distributed frozen and cooked from the frozen state be obtained and used for determining estimates associated with frozen ground beef, especially that used by fast-food restaurants. A trade association, such as the American Meat Institute, could be a source of this information. Differences in Cooking Practices Based on Location Are Not Appropriately Considered Practices for cooking ground beef in the home, at fast-food restaurants, and in other HRI facilities vary considerably; those of major chain fast-food restaurants are well defined and validated to kill pathogens, whereas those used in the home are based largely on the appearance of the cooked product and may result in pathogen survival. A 2002 case-control study conducted by the Centers for Disease Control and Prevention to identify risk factors associated with sporadic E. coli O157 infections determined that eating hamburgers cooked in the home was a major risk factor (Kennedy et al., 2002), whereas an earlier case-control study based on data obtained through the same FoodNet system identified eating hamburgers served at table-service restaurants—but not restaurants of major fast-food chains—as a major risk factor (Kassenborg et al., 1998). The committee recommends that each location—the home, fast-food restaurants, and the remainder of HRI facilities—where ground beef is cooked be modeled separately. That would necessitate that data on internal temperatures of cooked ground-beef patties be obtained or estimated for the three general locations. The Risk Characterization chapter in the draft correctly notes that “data on variability in food preparation behavior between consumers (home) and food preparers (HRI) are lacking” (p. 141) but this does not necessarily preclude modeling. The committee notes, for example, that most fast-food restaurants that cook patties from the frozen state would not encounter the wide variation in pretreatment storage conditions that was used in the draft to model cooking of ground beef and that variability in pretreatment storage conditions would more likely occur in ground beef cooked in homes. Caution should be used in applying to the model the data of Jackson
OCR for page 67
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment et al. (1996) regarding the mean reduction in E. coli O157:H7 in grilled ground-beef patties because some of their results (summarized in the draft’s Table 3-20) are counterintuitive. There are several observations where greater or equivalent E. coli O157:H7 populations were killed at 62.8°C (145°F) than at 68.3°C (155°F). It is well established that the higher the temperature (above the maximal growth temperature), the greater the number of bacteria killed. Furthermore, pretreatment by freezing may increase the sensitivity of pathogens like E. coli O157:H7 to thermal inactivation. The Jackson et al. data contradict that: more E. coli O157:H7 were inactivated at equivalent cooking temperatures in patties previously held refrigerated at 3°C for 9 hours than in patties held frozen at 18°C for 8 days. The committee recommends that, until more reliable data become available, D values3 established for E. coli O157:H7 inactivation in ground beef be used to model the effect of pretreatment storage conditions on rates of E. coli O157:H7 inactivation. The analysis should account for the varied fat content of ground beef used in the home, fast-food restaurants, and other HRI environments. Estimates of Amount of Raw Ground Beef Consumed Are Flawed The draft model calculates that “cooking” does not yield any log reduction of E. coli O157:H7 in 4–8% of ground beef servings. The explanation—described in a footnote (on p. 89)—is that the USDA’s Continuing Survey of Food Intakes by Individuals (CSFII) data used as the sole reference reported that four people (three 25–64 years old and one less than 5 years old) consumed “raw” ground beef. For modeling purposes, the servings were considered to be a subset of servings that had no log reduction in E. coli O157:H7 during cooking (for example, grossly undercooked servings). That information is critical to understanding the rationale for the relatively high occurrence of no-log-reduction ground-beef servings. Because of the importance of log-reduction information for interpreting calculated estimates, the committee recommends that the material in the footnote be moved to the text after the estimates that are presented as having no log reduction. More important, simple extrapolation of data from the 1994–1996 and 1998 CSFII surveys for estimating the annual number of raw ground-beef servings is scientifically unfounded because of the small number of obser 3 D (or decimal reduction) value is the amount of time in minutes required to reduce the number of organisms of a particular bacterium by 90% at a specified temperature. A 90% reduction—from 106 to 105 colony-forming units, for example—is equivalent to a 1-log decrease.
OCR for page 68
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment vations available in some subsets. Table 3-24 in the draft indicates that children 0–5 years old eat an estimated 522,315 servings of raw ground beef annually away from home but none at home. That calculation is based on a single observation, and confidence intervals are not included. The committee recommends that FSIS acknowledge that it lacks adequate information on the consumption of raw ground beef in the United States. Linear scaling of observations from one or a small number of individuals to the entire US population is statistically inappropriate. The committee believes that in this circumstance expert judgment, with appropriate accounting for uncertainty, may be superior to using extant data and suggests that FSIS solicit such input in the short term.4 For the longer term, the committee suggests that better data on raw-meat consumption be gathered and that research account for the fact that some groups of individuals consume raw ground beef in traditional dishes or in keeping with cultural traditions. USDA’s Agricultural Marketing Service or industry sources may have additional information bearing on this question. Human Exposure to E. coli O157:H7 in Ground Beef Fails to Address Potentially Important Variables The primary outputs of the FSIS draft Preparation Module are estimates of distributions that describe the prevalence of E. coli O157:H7 in ground-beef servings prepared during the seasons in which E. coli O157:H7 is more and less prevalent in cattle at slaughter. The Preparation Module relies solely on outputs of the Slaughter Module related to seasonal differences rather than using FSIS ground-beef sampling data. In addition, considering the major differences in handling of frozen ground beef and in cooking ground beef between fast-food restaurants and the home, there may be substantial differences in distributions of E. coli O157:H7 in ground-beef servings, depending on the location where the meat is prepared, cooked, and consumed. The committee recommends that FSIS ground-beef sampling data be used to determine seasonal differences in E. coli O157:H7 contamination of ground beef and that inputs into the model be further differentiated on the basis of location of ground-beef preparation and consumption. Insight into another potentially important variable is provided by the CSFII. The 1994–1996 data regarding ground-beef consumption, repro 4 Faced with a similar problem in their risk assessment of Shiga-producing E. coli O157 in steak tartare, Dutch researchers convened an expert solicitation workshop to estimate values for parameters for which no data were found (Nauta et al., 2001).
OCR for page 69
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment TABLE 4-1 Ground-beef quantity (g) consumed per eating occasion by age and sex (2-day sample) Age, years 2–5 6–11 12–19 20–39 40–59 60 and above M|F M|F M F M F M F M F Sample Size 2,109 1,432 696 702 1,543 1,449 1,663 1,694 1,545 1,429 Mean 31 41 66 52 80 52 82 57 73 62 Percentile 5th 1 1 1 1 3 1 3 2 3 2 10th 1 2 2 2 6 3 7 3 8 12 25th 10 16 28 23 38 22 40 24 36 31 50th 24 34 58 44 71 45 70 48 68 54 75th 42 59 88 77 102 77 110 84 96 87 90th 68 85 131 101 155 102 169 112 125 113 95th 85 104 168 127 194 122 200 129 176 140 Source: Smiciklas-Wright et al., 2002 (based on USDA, 2000).
OCR for page 70
Escherichia Coli O157:H7 In Ground Beef: Review of a Draft Risk Assessment duced below in Table 4-1, suggests that both sex and age are important in serving size. Age is factored in by the model, but sex is not accounted for in the characterization of the quantity of ground-beef products consumed. The gain in precision from including sex is likely to be small compared with other elements for which data are weak or absent. The committee suggests that the final risk assessment at least note the possible role of sex for completeness and future reference. REFERENCES Audits International. 1999. Audits International Home Food Safety Survey. http://www.foodriskclearinghouse.umd.edu/audits_international.htm, accessed June 10, 2002. Audits International. 2001. Audits International 2000 Home Food Safety Study. Buchanan RL, Doyle MP. 1997. Foodborne disease significance of Escherichia coli O157:H7 and other enterohemorrhagic E. coli. Food Technology 51:69–75. CDC (Centers for Disease Control and Prevention). 1996. Outbreak of Escherichia coli O157:H7 infection Georgia and Tennessee June 1995. Morbidity and Mortality Weekly Report 45(12):249–251. FDA (Food and Drug Administration). 2000. Report of the FDA Retail Food Program Database of Foodborne Illness Risk Factors. FDA Retail Food Program Steering Committee. August 10, 2000. Jackson, TC, Hardin MD, Acuff GR. 1996. Heat resistance of Escherichia coli O157:H7 in a nutrient medium and in ground beef patties as influenced by storage and holding temperatures. Journal of Food Protection 59:230–237. Kassenborg H, Hedberg C, Evans M, Chin G, Fiorentino T, Vugia D, Bardsley M, Slutsker L, Griffin P. 1998. Case-control study of sporadic Escherichia coli O157:H7 infections in 5 FoodNet sites. Abstracts of the International Conference on Emerging Infectious Diseases, p. 50. Kennedy, MH, Rabatsky-Ehr T, Thomas SM, Lance-Parker S, Mohle-Boetani J, Smith K, Keene W, Sparling P, Hardnett FP, Mead PS, and the EIP FoodNet Working Group. 2002. Risk factors for sporadic Escherichia coli O157 infections in the United States: A case-control study in FoodNet sites, 1999–2000. Abstracts of the International Conference on Emerging Infectious Diseases, p. 169. Mead PS, Finelli L, Lambert-Fair MA, Champ D, Townes J, Hutwagner L, Barrett T, Spitalny K, Mintz E. 1997. Risk factors for sporadic infection with Escherichia coli O157:H7. Archives of Internal Medicine 157(2):204–208. Nauta MJ, Evers EG, Takumi K, Havelaar AH. 2001. Risk assessment of Shiga-toxin producing Escherichia coli O157 in steak tartare in the Netherlands. Rijksinstituut voor Volksgezondheid en Milieu (RIVM). Report 257851 003. Smiciklas-Wright H, Mitchell DC, Mickle SJ, Cook AJ, Goldman JD. 2002. Foods Commonly Eaten in the United States: Quantities Consumed per Eating Occasion and in a Day, 1994– 1996. US Department of Agriculture NFS Report No. 96-5, prepublication version. http://www.barc.usda.gov/bhnrc/foodsurvey/Products9496.html, accessed July 10, 2002. USDA (US Department of Agriculture), Agricultural Research Service. 2000. Continuing Survey of Food Intakes by Individuals 1994–96, 1998. National Technical Information Service. CD-ROM. NTIS Accession no. PB2000-500027.
Representative terms from entire chapter: