The body of this report contains the principal comments of the committee regarding the content of the draft US Department of Agriculture E. coli O157:H7 risk assessment. This appendix contains a number of additional comments regarding citations, typographic errors, and the like that the committee wishes to bring to the attention of the draft’s authors. The comments are listed by page number.
Page 12: In the list of outputs of the Production Module, insert the words “use in manufacturing” (or a comparable descriptive phrase) after the word “for” in “prior to slaughter for ground beef.”
Page 25: The study by Brackett (not Brachett) et al. (1994) was of decontamination of meat cuts, not carcasses as implied.
Page 32: The word “proportion” may be better than the word “share” in describing amounts of imported and domestic ground beef consumed, which could explain the findings.
Page 33: “Test sensitivity is a complex parameter that incorporates variability in sample collection and handling and in the biological properties of the sample”: the meaning of the statement is unclear. It should simply be stated that sample variability may affect apparent test sensitivity.
Page 33: “The slaughter plant intake segment considers the effect of clustering cattle as they enter the slaughter plant”: this would apply to cows and bulls, which may have different sources and be commingled; steers and heifers from feedlots, however, are not mixed with cattle from other groups.
Page 39: “Although evidence is limited, it suggests that dairy cow
calf herds are similar with respect to E. coli O157:H7”: it is not clear what regarding O157:H7 is similar.
Page 51: “Five studies provide evidence on apparent within-feedlot”: insert the word “prevalence” after that phrase.
Page 51: “…this protocol is assumed to be 100% sensitive”: a reason should be provided for this assumption.
Page 53: The age of breeding cattle is more likely 3, instead of 2, years; most feedlot cattle should be 1–2 years old.
Page 56: Truckloads of cattle from different feedlots are not usually mixed at slaughter.
Page 65: The draft risk assessment cites Sheridan et al. (1992) as having identified such equipment as knives, gloves, and aprons as reservoirs of bacteria in the slaughterhouse. That is correct, but the reference also indicates that “the level of contamination varied with different cuts of meat,” which may affect the extent of contamination of trim. There is also a typographic error in the citation: Meat Science 32:185–194, rather than 32:155–164.
Page 66: Contrary to what is stated in the draft, no “excess fat is trimmed away from each side” of the carcass at splitting (before washing), although blood-soiled tissue may be trimmed.
Page 67: The draft asserts that distilled water and chlorine are occasionally sprayed on carcasses in chillers (Step 6). Carcasses are spraychilled with water but not with distilled water or chlorine (which causes corrosion). Lactic acid may be in the initial stages of finding some use in this application.
Page 67: “FSIS regulations require chilling deep muscle (6 inches) to 10.0°C (50.0°F) within 24 hours and 7.2°C (45.0°F) within 36 hours (NACMCF, 1993)”: this may be done in practice but to the committee’s knowledge, is not required by the Food Safety and Inspection Service; if it is, a more direct reference should be provided.
Page 67: “Dorsa (1997) found a 1.2 log CFU/cm2 increase in E. coli O157:H7 on carcasses stored for 2 days in the chiller at 5.0°C (41.0°F)”: It appears that the correct reference may be Dorsa et al. (1997), not Dorsa (1997). Instead of carcasses, the study evaluated inoculated beef-carcass tissue samples that were decontaminated and packaged, thus simulating retail products rather than carcasses. In addition, it is questionable whether E. coli O157:H7 would grow at 5°C.
Page 68: “Prasai et al. (1995) found no difference in concentrations of E. coli O157:H7 between hot deboning and cold deboning”: it appears that the Prasai et al. (1995) reference listed does not deal with E. coli O157:H7 or hot or cold deboning.
Page 70: The carcass surface areas estimated to end up in ground beef
seem high (75–90%; citing McAloon, 1999), given that much of the external carcass surface fat is trimmed away during fabrication and is used in rendering. One major concern is the issue of estimating surface area of trim, considering the extensive cutting that takes place during fabrication.
Page 77: Combo bins contain trim, not ground beef.
Page 77: Newer data may be available for ground-beef proportions used at retail and in hotels, restaurants, and institutions (HRI); retail grinding may have decreased. The draft apparently does not, but should, consider coarse ground-beef chubs.
Page 79/81: Palumbo (1997), which is cited in the text, is not in the reference list.
Page 82, 83: Citations of the unpublished Vose 1999 manuscript should be updated to a published reference if at all possible.
Page 95: The report needs to specify how combinatorial mathematics was used to compute the convolutions instead of Monte Carlo simulation.
Page 100: The Gill (1996) reference may be incorrectly listed.
Page 124: MPN is not CFU. Most probable number and colony forming units are measures of microbial density measured differently.
Page 142: The draft report’s correlation analysis states that “the size of the E. coli O157:H7-contaminated carcass surface was the only factor correlated (coefficient = 0.33) with the number of E. coli O157:H7 organisms in steer/heifer combo bins (Table 5-3).” It should be noted that there is no evidence (and it is probably impossible to obtain evidence) on the size of the area of a carcass that is contaminated. In addition, contamination is not expected to occur uniformly on a given carcass area (whatever the area); contamination usually occurs as clumps of cells in microscopic environments.
Page 166: How is it possible that deterministic values “come from a distribution” but “do not change as a result of Monte Carlo iteration”?
Page 170: Here and elsewhere in Appendix C there are some apparent inconsistencies in the definition of the term deterministic. For instance, in Equation 3.3, how can Hsens be deterministic if one of the variables used to calculate it— pi —is stochastic?
Page 183: The last word in the description of Equation 3.15 should be “evisceration” rather than “dehiding.”
Page 206: Here and elsewhere in Appendix C, what does it mean when the space reserved to declare a variable as deterministic or stochastic is left blank?
Appendix D, page 3: Minus signs, rather than parentheses, should be used to denote negative values for consistency with the rest of the draft.
The temperature scale (Celsius) used in the regression should be mentioned. The words incorporating and ste-y should be defined or explained in the text.
REFERENCES FROM THE DRAFT RISK ASSESSMENT THAT NEED TO BE CORRECTED
Cassin MH, Lammerding AM, Todd EC, Ross W, McColl RS. 1998. Quantitative risk assessment of Escherichia coli O157:H7 in ground beef hamburgers. International Journal of Food Microbiology 41:21–44.
Codex Alimentarius Commission. 1999. Principles and Guidelines for the Conduct of Microbiological Risk Assessment. CAC/GL-30. Food and Agricultural Organization of the United Nations.
FDA/USDA 2001. Draft Assessment of the Relative Risk to Public Health from Foodborne Listeria monocytogenes Among Selected Categories of Ready-to-Eat Foods. FDA/Center for Food Safety and Applied Nutrition, USDA/Food Safety and Inspection Service, Centers for Disease Control and Prevention, January 2001. http://www.foodsafety.gov/~dms/lmrisk.html.
ILSI Risk Science Institute. 2000. Revised Framework for Microbial Risk Assessment—An ILSI Risk Science Institute Workshop report. Washington, DC: International Life Sciences Institute.
Marks HM, Coleman ME, Lin CTJ, Roberts T. 1998. Topics in risk assessment: Dynamic flow tree process. Risk Analysis 18:309–328.
Smeltzer TI, Peel B, Collins G. 1979. The role of equipment that has direct contact with the carcase in the spread of Salmonella in a beef abattoir. Australian Veterinary Journal 55:275–277.
WHO (World Health Institute), 1995. Application of Risk Analysis to Food Standards IssuesThe report of the Joint FAO/WHO Expert Consultation, 13–17 March 1995 (WHO/ FNU/FOS/95.3).