PREVENTING FOODBORNE DISEASE FROM ENVIRONMENTAL SOURCES: FROM RETAIL AND PROCESSING PLANTS TO K – 12
EDUCATION
A Dissertation Presented to the Faculty of the Graduate School
of Cornell University In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
by Courtenay Simmons
August 2016

© 2016 Courtenay Simmons

PREVENTING FOODBORNE DISEASE FROM ENVIRONMENTAL SOURCES: FROM RETAIL AND PROCESSING PLANTS TO K – 12 EDUCATION
Courtenay Simmons, Ph. D. Cornell University 2016
Listeria monocytogenes is widely dispersed throughout natural and urban environments and commonly persists in the food processing environment under a range of conditions. Food processing plants contain a number of surfaces and areas that serve as environmental reservoirs of L. monocytogenes and other L.spp.; however, limited science-based data describes L. spp. ecology in Ready-to-Eat (RTE) facilities that handle post-heat treated products.
In the following studies, we investigated several approaches to control L. spp. in areas relevant to food safety. Specifically, in the work described here we: (i) used Pulsed-Field Gel Electrophoresis (PFGE), a molecular-based subtyping method to study the ecology and transmission of L. monocytogenes in 30 U.S. full-service retail delicatessens across 3 geographical regions of the U.S., (ii) developed an assessment tool that will help food industry professionals design their Pathogen Environmental Monitoring (PEM) programs for L. monocytogenes, and (iii) translated research into a food safety module for K – 12 education.
Overall, we observed 9.5% (427/4503) and 5.3% (237/4503) L. monocytogenes and L. spp. prevalence rates in retail delis respectively. We found L. monocytogenes persistence in a number of facilities (12/30) including sporadic contamination events in others (2/30) and observed cross-contamination between non-food contact surfaces (NFCS) and food-contact surfaces (FCS). Characterization of 245 putative L. spp. isolates identified L. innocua as the
iii

predominant L. spp. aside from L. monocytogenes. Prompted by research findings, we developed an assessment tool that can be used for the development and implementation of scientifically supported PEM programs. The assessment tool may have a broader impact in the food industry as it may provide a starting point for similar efforts targeting other pathogens.
Further inspired by research findings, we developed a mini-course for middle grade science classes to (i) teach food safety fundamentals and (ii) stimulate interest in food science and food safety. We found that the level of student engagement and content retention increased as familiarity with content increased. We recommend the frequent translation of food safety research into age-appropriate mini-courses or modules to promote science literacy, encourage broader public understanding, and demonstrate concepts outlined in the Next Generation Science Standards.
iv

BIOGRAPHICAL SKETCH Courtenay Simmons was born in South Carolina where she attended Claflin University and received a B.S. degree in Biochemistry. As an undergraduate student, Courtenay had the opportunity to visit the Agricultural Research Service (ARS) in Athens, Georgia. While visiting the ARS, Courtenay became interested in Food Science and would later focus her attention on Food Microbiology. During her undergraduate career, she conducted research in a chemical engineering lab and an analytical chemistry lab but her fascination with Food Science never waned. In the summer of 2008, Courtenay participated in the Food Science Summer Scholars Program at Cornell University and worked with Dr. Martin Wiedmann in the Food Safety Lab. She decided to pursue her Ph.D. and began as a doctoral student in 2009 in the Food Science and Technology Department under the direction of Dr. Martin Wiedmann. Courtenay is excited to use her background to focus on food safety in federally administered child nutrition programs.
v

DEDICATED TO JAMYRA SIMMONS vi

ACKNOWLEDGMENTS I thank Dr. Martin Wiedmann for his unwavering support and mentorship during my time at Cornell, and also my minor advisors, Dr. Andrew Novakovic and Dr. Chang Yong Lee. I also thank collaborators for their help with research and the administration staff, particularly, Janette, Marin, and Erin for their patience, professionalism, and diligence. In addition, I thank past and current technicians, post-docs and labmates for their friendship and research support over the years. I am particularly grateful to Sherry for her consistency and quick wit; Maureen for her hard-work; Alphina for her constant willingness to help; and Dan for his moments of underrated awesomeness. I especially thank Dr. Kathryn Boor for her encouragement and calm, yet strong presence she brings to the lab. This work was supported by a USDA Food Safety and Inspection Service, contract 289918 awarded to M. Wiedmann.
vii

TABLE OF CONTENTS

Chapter 1 Introduction

Chapter 2 Listeria monocytogenes and Listeria spp. contamination

patterns in retail delicatessen establishments in three U.S. states

Chapter 3 Expert elicitation used for the identification and classification

of sample collection sites for pathogen environmental monitoring

Chapter 4

programs for Listeria monocytogenes Learning outcomes of a K – 12 food safety mini-course

Chapter 5 Conclusions

Appendix 1 Listeria subtyping results for 30 delis

1 11
41
82 99 102

viii

LIST OF FIGURES

Figure 2.1 Figure 2.2 Figure 2.3
Figure 2.4
Figure 3.1 Figure 3.2
Figure 3.3
Figure 3.4
Figure 4.1

Listeria subtyping results from Phase I and Phase II for Deli 8 Listeria monocytogenes prevalence for 30 delis L. monocytogenes prevalence by deli for samples collected during Phase II Heat map showing L. monocytogenes and Listeria spp. prevalence data for samples collected during Phase II Sites are classified into Zones, 1,2,3, and 4 Rankings of individual sites based on expert reviewers’ opinion of the level of importance from 1–5 Expert reviewers’ classification of sites into niches (Ni) and transfer sites (TS) and zones Expert reviewers’ classification of sites into (VS) and indicator sites (IS) and zones Individuals’ performance on pre- and post-assessments

20 21 27
29
65 68
70
72
89

ix

LIST OF TABLES

Table 2.1
Table 2.2
Table 2.3 Table 2.4 Table 3.1 Table 3.2 Table 3.3
Table 4.1 Table 4.2

Listeria monocytogenes and Listeria spp. Prevalence among 7 sites sampled pre-operationally Listeria species and sigB allelic types isolated from the 30 Retail deli environments sampled PFGE types isolated from 2 or more delis Re-isolation of PFGE types from a given deli Summary of expert reviewers’ profiles New sites recommended by reviewers Classification of sites based on expert reviewers’ recommendations Pre- and post-assessment results Bell ringer results

18
23
25 33 51 53 55
87 92

x

CHAPTER 1 INTRODUCTION Listeria monocytogenes is a gram-positive facultative anaerobic bacterium that has both pathogenic and saprophytic lifestyles in the natural, farm, and food processing environments (26, 34). Presently there are 17 species in the genus Listeria, including 1 pathogenic species known to cause illness in human and animals (L. monocytogenes), 1 species known to infect ruminants and implicated in a few human cases (L. ivanovii), and 15 non-pathogenic species (L. seeligeri, L. welshimeri, L. innocua, L. marthii, L. grayi, L. rocourtiae, L. weihenstephanensis, L. fleischmannii, L. floridensis, L. aquatic, L. cornellensis, L. riparia, L. grandensis, L. booriae, L. newyorkensis (5, 16, 32, 36), including 11 species identified since 2009 (2, 6, 15, 17, 28). L. monocytogenes ability to survive and grow under conditions used to minimize survival and growth of the pathogen (e.g., refrigeration temperatures, osmotic stress) is difficult to control in both the environment and foods and consequently is a public health risk (29). Since the 1980’s, there has been a steady increase in cases of listeriosis (the infection caused by L. monocytogenes) in North America and Europe (3, 24, 42). L. monocytogenes causes on average 1,455 hospitalizations and 255 deaths annually (35), particularly amongst the elderly, immunocompromised, pregnant women, and neonate populations (18). L. monocytogenes has the third highest hospitalization rate among foodborne pathogens and is cited as having the highest burden cost associated with a foodborne pathogen in the U.S. (20). Costs associated with illnesses include hospitalizations, necessary long-term care, loss in wages, recalls, and fines total in the billions of dollars and can have lasting impact on consumers and food industry (20). Researchers compare L. monocytogenes lifestyle to that of Dr. Jekyll and Mr. Hyde and continue to study the mechanisms that aids in L. monocytogenes pathogenic lifestyle in human
1

and animal hosts and its innocuous lifestyle in the natural environment in soil and vegetation and in food facilities (9). A number of studies focused on L. monocytogenes ecology in produce farms, pristine environments, and urban environment settings (4, 34, 39), in addition to a number of studies that have characterized the organism’s transmission in smoked seafood facilities, and full-service retail delicatessen establishments (19, 40). A variety of food associated environments, industry studies suggest that L. monocytogenes from the natural environment is frequently introduced into food facilities that may support growth, persistence, and transmission of the organisms once introduced (23, 41). Additionally, research indicates L. monocytogenes found in finished food products are also often recovered from the environment (1, 21, 22, 33), suggesting environmental contamination sources.
While control measures help to reduce L. monocytogenes prevalence in facilities, in recent years, overall incidence of foodborne listeriosis has not shown a similar decline that would mirror reduced prevalence seen in foods, and in particular RTE meat products (31). A study by USDA Food Safety and Inspection Service (FSIS) and FDA suggests that of all RTE meat products implicated in listeriosis deaths, RTE meats sliced and packaged at retail, accounts for 83% of all deaths per annum (8, 30). Other reports indicate that L. monocytogenes prevalence found in RTE deli meats sliced at retail is 7 times higher than L. monocytogenes prevalence in prepackaged deli meats in the U.S. and Canada (7, 12). In 2002, turkey deli meat contaminated with L. monocytogenes was the source of a multistate outbreak involving 54 cases of listeriosis, 8 deaths, and 3 miscarriages (13). Soon thereafter, the USDA Food Safety and Inspection Service adopted new regulations for a L. monocytogenes testing program for RTE meats and poultry processing facilities to minimize transmission and contamination of final RTE meat products (11, 38).
2

In addition to adhering to the regulations set forth by USDA FSIS, the RTE meat industry has been proactive in tracking and eliminating L. monocytogenes from the environment to prevent final product contamination. Federal government, industry support groups, and academic institutions work together to (i) identify mechanisms and contributing factors by which L. monocytogenes persists in the environment (10, 43), (ii) understand the ecology in facilities (22, 37), (iii) develop science-based strategies to control L. monocytogenes (25), and (iv) guide the development of pathogen environmental programs (PEM) (11, 14). Additionally, studies led by academic researchers use molecular subtyping methods to track contamination sources (27) and identify contamination patterns and trends across facilities and regions in the U.S. and new and emerging strains. The following chapters will describe the: application of subtyping methods to study the ecology of L. monocytogenes in the RTE delicatessen environment, the development of an assessment tool used to elicit expert opinion for environmental monitoring sites for L. monocytogenes programs, and translation of research into a food safety module for k – 12 education.
3

REFERENCES 1. Aarnisalo, K., Autio, T., Sjöberg, A.M., Lundén, J., Korkeala, H. and Suihko,
M.L. 2003. Typing of Listeria monocytogenes isolates originating from the food processing industry with automated ribotyping and pulsed-field gel electrophoresis. Journal of Food Protection 66:249-255. 2. Bertsch, D., Rau, J., Eugster, M.R., Haug, M.C., Lawson, P.A., Lacroix, C. and Meile, L. 2013. Listeria fleischmannii sp. nov., isolated from cheese. International journal of systematic and evolutionary microbiology 63:526-532. 3. Bille, J.1990. Epidemiology of human listeriosis in Europe, with special reference to the Swiss outbreak. Foodborne listeriosis 71-74. 4. Chapin, T.K. 2014. Spatio-temporal distribution and ecology of Listeria, Salmonella and shiga toxin-producing Escherichia coli in New York state (Doctoral dissertation, Cornell University). 5. Cummins, A.J., Fielding, A.K. and McLauchlin, J. 1994. Listeria ivanovii infection in a patient with AIDS. Journal of infection 28:89-91. 6. den Bakker, H.C., Warchocki, S., Wright, E.M., Allred, A.F., Ahlstrom, C., Manuel, C.S., Stasiewicz, M.J., Burrell, A., Roof, S., Strawn, L.K. and Fortes, E. 2014. Listeria floridensis sp. nov., Listeria aquatica sp. nov., Listeria cornellensis sp. nov., Listeria riparia sp. nov. and Listeria grandensis sp. nov., from agricultural and natural environments. International journal of systematic and evolutionary microbiology 64:1882-1889. 7. Draughon, A.F., Ryser, E., Oyarzabal, O., Hajmeer, M. and Cliver, D. 2006. A collaborative analysis/risk assessment of Listeria monocytogenes in ready-to-eat
4

processed meat and poultry collected in four FoodNet states. In International Association for Food Protection 93rd Annual Meeting. 8. Endrikat, S., Gallagher, D., Pouillot, R., Hicks Quesenberry, H., LaBarre, D., Schroeder, C.M. and Kause, J. 2010. A comparative risk assessment for Listeria monocytogenes in prepackaged versus retail-sliced deli meat. Journal of Food Protection 73:612-619. 9. Fenlon, D.R. 1999. Listeria monocytogenes in the natural environment. Food Science and Technology-New York-Marcel Dekker 21-38. 10. Ferreira, V., Wiedmann, M., Teixeira, P. and Stasiewicz, M.J. 2014. Listeria monocytogenes persistence in food-associated environments: epidemiology, strain characteristics, and implications for public health. Journal of Food Protection 77:150170. 11. USDA. 2014. Compliance guideline: controlling Listeria monocytogenes in postlethality exposed ready-to-eat meat and poultry products. Food Safety and Inspection Service. 12. Gombas, D.E., Chen, Y., Clavero, R.S. and Scott, V.N. 2003. Survey of Listeria monocytogenes in ready-to-eat foods. Journal of Food Protection 66:559-569. 13. Gottlieb, S.L., Newbern, E.C., Griffin, P.M., Graves, L.M., Hoekstra, R.M., Baker, N.L., Hunter, S.B., Holt, K.G., Ramsey, F., Head, M. and Levine, P. 2006. Multistate outbreak of listeriosis linked to turkey deli meat and subsequent changes in US regulatory policy. Clinical Infectious Diseases 42:29-36. 14. GMA. 2014. Listeria monocytogenes guidance on environmental monitoring and corrective actions in at-risk foods.
5

15. Graves, L.M., Helsel, L.O., Steigerwalt, A.G., Morey, R.E., Daneshvar, M.I., Roof, S.E., Orsi, R.H., Fortes, E.D., Milillo, S.R., Den Bakker, H.C. and Wiedmann, M. 2010. Listeria marthii sp. nov., isolated from the natural environment, Finger Lakes National Forest. International Journal of Systematic and Evolutionary Microbiology 60:1280-1288.
16. Guillet, C., Join-Lambert, O., Le Monnier, A., Leclercq, A., Mechaï, F., Mamzer-Bruneel, M.F., Bielecka, M.K., Scortti, M., Disson, O., Berche, P. and Vazquez-Boland, J. 2010. Human listeriosis caused by Listeria ivanovii. Emerg Infect Dis 16:136-138.
17. Halter, E.L., Neuhaus, K. and Scherer, S. 2013. Listeria weihenstephanensis sp. nov., isolated from the water plant Lemna trisulca taken from a freshwater pond. International journal of systematic and evolutionary microbiology 63:641-647.
18. Hitchins, A.D., 1996. Assessment of alimentary exposure to Listeria monocytogenes. International journal of food microbiology 30:71-85.
19. Hoelzer, K., Sauders, B.D., Sanchez, M.D., Olsen, P.T., Pickett, M.M., Mangione, K.J., Rice, D.H., Corby, J., Stich, S., Fortes, E.D. and Roof, S.E. 2011. Prevalence, distribution, and diversity of Listeria monocytogenes in retail environments, focusing on small establishments and establishments with a history of failed inspections. Journal of Food Protection 74:1083-1095.
20. Hoffman, S., Batz, M.B., Morris, J. 2012. Annual cost of illness and qualityadjusted life year losses in the United States due to 14 foodborne pathogens. Journal of Food Protection 75:1292–1302.
6

21. Kells, J. and Gilmour, A. 2004. Incidence of Listeria monocytogenes in two milk processing environments, and assessment of Listeria monocytogenes blood agar for isolation. International journal of food microbiology 91:167-174.
22. Lappi, V.R., Thimothe, J., Nightingale, K.K., Gall, K., Scott, V.N. and Wiedmann, M. 2004. Longitudinal studies on Listeria in smoked fish plants: impact of intervention strategies on contamination patterns. Journal of Food Protection 67:2500-2514.
23. Lianou, A. and Sofos, J.N. 2007. A review of the incidence and transmission of Listeria monocytogenes in ready-to-eat products in retail and food service environments. Journal of Food Protection 70:2172-2198.
24. Linnan, M.J., Mascola, L., Lou, X.D., Goulet, V., May, S., Salminen, C., Hird, D.W., Yonekura, M.L., Hayes, P., Weaver, R. and Audurier, A. 1988. Epidemic listeriosis associated with Mexican-style cheese. New England Journal of Medicine 319:823-828.
25. Malley, T.J., Butts, J. and Wiedmann, M. 2015. Seek and destroy process: Listeria monocytogenes process controls in the ready-to-eat meat and poultry industry. Journal of Food Protection 78:436-445.
26. Nightingale, K.K., Schukken, Y.H., Nightingale, C.R., Fortes, E.D., Ho, A.J., Her, Z., Grohn, Y.T., McDonough, P.L. and Wiedmann, M. 2004. Ecology and transmission of Listeria monocytogenes infecting ruminants and in the farm environment. Applied and Environmental Microbiology 70:4458-4467.
27. Norton, D.M., McCamey, M.A., Gall, K.L., Scarlett, J.M., Boor, K.J. and Wiedmann, M. 2001. Molecular studies on the ecology of Listeria monocytogenes in
7

the smoked fish processing industry. Applied and Environmental Microbiology 67:198-205. 28. Orsi, R.H. and Wiedmann, M. 2016. Characteristics and distribution of Listeria spp., including Listeria species newly described since 2009. Applied microbiology and biotechnology 1-15. 29. Painter, J.A., Hoekstra, R.M., Ayers, T., Tauxe, R.V., Braden, C.R., Angulo, F.J. and Griffin, P.M. 2013. Attribution of foodborne illnesses, hospitalizations, and deaths to food commodities by using outbreak data, United States, 1998–2008. Emerg Infect Dis 19:407-415. 30. Pradhan, A.K., Ivanek, R., Gröhn, Y.T., Bukowski, R., Geornaras, I., Sofos, J.N. and Wiedmann, M. 2010. Quantitative risk assessment of listeriosis-associated deaths due to Listeria monocytogenes contamination of deli meats originating from manufacture and retail. Journal of Food Protection 73:620-630. 31. Quesenberry, H.H., Gallagher, D., Endrikat, S., La Barre, D., Ebel, E., Schroeder, C. and Kause, J. 2012. FSIS comparative risk assessment for Listeria monocytogenes in ready to eat meat and poultry deli meats. 32. Rocourt, J. and Grimont, P.A. 1983. Notes: Listeria welshimeri sp. nov. and Listeria seeligeri sp. nov. International Journal of Systematic and Evolutionary Microbiology 33:866-869. 33. Sauders, B.D., Mangione, K., Vincent, C., Schermerhorn, J., Farchione, C.M., Dumas, N.B., Bopp, D., Kornstein, L., Fortes, E.D., Windham, K. and Wiedmann, M. 2004. Distribution of Listeria monocytogenes molecular subtypes among human and food isolates from New York State shows persistence of human
8

disease–associated Listeria monocytogenes strains in retail environments. Journal of Food Protection 67:1417-1428. 34. Sauders, B.D., Overdevest, J., Fortes, E., Windham, K., Schukken, Y., Lembo, A. and Wiedmann, M. 2012. Diversity of Listeria species in urban and natural environments. Applied and environmental microbiology 00282 35. Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R.V., Widdowson, MA., Roy, S.L., Jones, J.L., Griffin, P.M. 2011. Foodborne illness acquired in the United States— major pathogens. Emerging Infectious Diseases. 36. Seeliger, H.P., Rocourt, J., Schrettenbrunner, A., Grimont, P.A. and Jones, D. 1984. Notes: Listeria ivanovii sp. nov. International Journal of Systematic and Evolutionary Microbiology 34:336-337. 37. Simmons, C., Stasiewicz, M.J., Wright, E., Warchocki, S., Roof, S., Kause, J.R., Bauer, N., Ibrahim, S., Wiedmann, M. and Oliver, H.F. 2014. Listeria monocytogenes and Listeria spp. contamination patterns in retail delicatessen establishments in three US states. Journal of Food Protection 77:1929-1939. 38. Stone, W.E. 2014. Regulatory Testing Guidelines and Recommendations. In The Microbiological Safety of Low Water Activity Foods and Spices 347-366. 39. Strawn, L.K., 2014. Ecology And Epidemiology Of Salmonella And Listeria Monocytogenes In New York State Produce Production Environments (Doctoral dissertation, Cornell University). 40. Thimothe, J., Nightingale, K.K., Gall, K., Scott, V.N. and Wiedmann, M. 2004. Tracking of Listeria monocytogenes in smoked fish processing plants. Journal of Food Protection 67:328-341.
9

41. Todd, E.C.D. and Notermans, S. 2011. Surveillance of listeriosis and its causative pathogen, Listeria monocytogenes. Food Control 22:1484-1490.
42. Vázquez-Boland, J.A., Kuhn, M., Berche, P., Chakraborty, T., Domı́nguezBernal, G., Goebel, W., González-Zorn, B., Wehland, J. and Kreft, J. 2001. Listeria pathogenesis and molecular virulence determinants. Clinical microbiology reviews 14:584-640.
43. Wang, J., Ray, A.J., Hammons, S.R. and Oliver, H.F. 2015. Persistent and transient Listeria monocytogenes strains from retail deli environments vary in their ability to adhere and form biofilms and rarely have inlA premature stop codons. Foodborne pathogens and disease 12:151-158.
10

CHAPTER 2 LISTERIA MONOCYTOGENES AND LISTERIA SPP. CONTAMINATION PATTERNS IN
RETAIL DELICATESSEN ESTABLISHMENTS IN THREE U.S. STATES Published in: Journal of Food Protection
Abstract Post-processing contamination in processing plants has historically been a significant
source of L. monocytogenes in Ready-To-Eat deli meats, and therefore a major cause of human listeriosis cases and outbreaks. Recent risk assessments suggest that a majority of human listeriosis cases linked to consumption of contaminated deli meats may be due to L. monocytogenes contamination that occurs at retail. To better understand the ecology and transmission of L. monocytogenes in retail delicatessen, we tested food and non-food contact surfaces for L. monocytogenes in a longitudinal study conducted in 30 retail delis in 3 US states. In Phase I of the study, 7 sponge samples were collected once a month for 3 months in 15 delis (5 delis/state) prior to start of daily operation; in Phase II, 28 food contact and non-food contact sites were sampled in each of 30 delis during daily operation for 6 months. L. monocytogenes were isolated from 7 of the 15 delis sampled during Phase I. Among the 314 sites sampled during Phase I, 6.8% were positive for L. monocytogenes. Among 4503 sites sampled during Phase II, 9.5% were positive for L. monocytogenes; 10 of 30 delis showed low L. monocytogenes prevalence (<2%) for all surfaces. A total of 446 L. monocytogenes isolates were characterized by Pulsed-Field Gel Electrophoresis (PFGE). PFGE showed that for 12 of 30 delis, one or more PFGE types were isolated at least 3 times, providing evidence for persistence of a given L. monocytogenes subtype in the delis. For some delis, PFGE patterns for isolates from non-food contact surface were distinct from patterns for occasional food contact surface isolates,
11

suggesting limited cross-contamination between these sites in some delis. This study provides initial data on L. monocytogenes contamination patterns in retail delis, which should facilitate further development of L. monocytogenes control strategies in retail delicatessens.
12

INTRODUCTION Control of L. monocytogenes represents a particular challenge for the Ready-to-Eat
(RTE) food industry due to the common presence and persistence of L. monocytogenes in virtually all environments along the food continuum (e.g., (6, 12, 16). The 2003 L. monocytogenes risk assessment identified RTE deli meats as the food responsible for most human listeriosis cases (20). Recent data show that the reduction in incidence of listeriosis has not been as great as expected (2) given the substantial reductions in the prevalence of the pathogen in deli meat. A recent FSIS and FDA Interagency Retail Listeria monocytogenes Risk Assessment (21) as well as an independent risk assessment suggests that up to 83% of human listeriosis cases linked to RTE deli meats may be attributable to products contaminated at retail (14), possibly explaining in part why the frequency of human cases has not declined as expected.
Overall, only a few studies have investigated L. monocytogenes transmission, prevalence and persistence in retail and retails delis with many of the larger and more comprehensive studies published to-date conducted by our group. For example, Sauders et al., found in a crosssectional study involving 121 retails establishments that 60% of establishments tested had at least one sample that was positive for L. monocytogenes and that 151 (13.0%) of the 1,161 environmental samples tested positive for L. monocytogenes, including 125 (16.7%) and 26 (6.3%) non–food contact and food contact surface samples, respectively (18). Another crosssectional study (7) of 120 retail operations predicted to be at an increased risk for L. monocytogenes contamination (i.e., small establishments and establishments that largely had a history of failed New York State Department of Agriculture and Markets inspections) identified establishment size, geographic location, and inspection history as significant predictors of L. monocytogenes presence and prevalence. Specifically, the odds of an establishment being L.
13

monocytogenes positive were approximately twice as high for (i) large establishments, (ii) establishments located in New York City, or (iii) establishments with poor inspection history when compared to establishments without these attributes. These studies (7, 18), however, provide limited insight into L. monocytogenes persistence and cross-contamination dynamics in retail delis due to the design of the studies (e.g. cross-sectional or minimal longitudinal evaluation). Two studies have also investigated potential cross-contamination dynamics in mock delis using surrogates for L. monocytogenes (9, 19).
The goal of this study was to investigate the prevalence and persistence of L. monocytogenes and other Listeria spp. in 30 retail delis in three states before and during daily operations. We used Pulsed Field Gel Electrophoresis (PFGE) to (i) determine L. monocytogenes persistence patterns in the deli and (ii) infer the potential direction the pathogen moves throughout the deli. With the data gathered in this study along with existing data, we aimed to identify potential niches and to make recommendations for practices that could potentially reduce L. monocytogenes considerably not only in the deli environment but also in RTE environments other than RTE deli meats and in food processing facilities.
MATERIALS AND METHODS Study Design. A total of 30 retail delis with full service delicatessen establishments (referred to as “delis” for the rest of this manuscript) were enrolled in this study; all of these delis handled and sliced deli meats. Collection of environmental sponge samples in the delis was performed by the authors and university-based collaborators. A third party blinded all samples. For Phase I of the study, samples were collected before initiation of daily deli operations once a month for 3 months (April to June 2010) from 15 delis (5 delis in each of three US states). Sampling times
14

were selected so that sampling occurred after routine cleaning and sanitation and before slicing or other handling commenced. Sites sampled in Phase I represented 6 non-food contact surfaces (NFCS) and one food contact surface (FCS) (Table 2.1); all 7 sites were sampled in each deli store. Sites were selected to represent areas that may be “niches” where L. monocytogenes could survive over time.
For Phase II, samples were collected during deli operations once a month for 6 months from 30 delis (10 deli establishments in each of three US states); these 30 delis included the 15 delis sampled in Phase I as well as 15 additional delis (5 in each state). For delis included in Phase I, Phase II sample collection initiated the month immediately after the last months of phase I sample collection. A total of up to 28 designated sites were sampled at each Phase II sampling; these sites included 10 FCS sites, 15 NFCS sites, and 3 sites representing transfer points (TP) (i.e., deli case handle, slicer knob, and scale touchpads). Samples were collected if the site or piece of equipment was available on the day of sampling. While most sites were available at all or most visits, some sites were consistently absent in some of the delis (e.g., cold room drain). Sampling sites and sampling procedure. Samples from all sites were collected using gamma irradiated Whirl-pak bag sponge kits (Nasco, Modesto, Calif.), which contain sterile gloves, a sterile sponge with bag, and 10 mL of Phosphate buffered saline (PBS) with neutralizing buffer (containing sodium thiosulfate, and aryl sulfonate). Collection of sponge samples involved either (i) sampling of approximately 645 cm2 (100 in2) of a given surfaces (e.g., floor samples); (ii) sampling of different surfaces for a total of approximately 645 cm2 (e.g., sinks), or (iii) sampling of all available surfaces up to an estimated total of 645 cm2 (e.g., squeegee, deli case handles). L. monocytogenes and other Listeria spp. isolation from sponge samples. Each sponge was tested for L. monocytogenes and Listeria spp. using the FDA Bacteriological Analytical Manual
15

(BAM) procedure with minor modifications. Listeria spp. in this study refers to all species excluding L. monocytogenes. Briefly, 90 ml of Buffered Listeria Enrichment Broth (BLEB; Difco, Becton Dickinson, Sparks, MD) was added to each bag with the sample sponge, followed by homogenization in a Stomacher 400 Circulator at 230rpm for 60 seconds. After 4 h of incubation at 30°C, 360 µl of Listeria Selective Enrichment Supplement (LSES; stock solution contains 9 mg/ml nalidixic Acid, 2.25 mg/ml acriflavine; 11.25 mg/ml cycloheximide) was added to each enrichment. After incubation at 30°C for 24 and 48 h, enrichments were streaked to Modified Oxford (MOX; DIFCO, Becton Dickinson) and Listeria monocytogenes Plating Medium (LMPM; R&F Laboratories, Downers Grove, IL)); MOX and LMPM were incubated at 30°C and 35°C, respectively, for 48 h. If blue colonies were present on LMPM, four presumptive L. monocytogenes colonies were sub-streaked onto LMPM. If only white colonies were found on LMPM, two colonies were sub-streaked to LMPM. If LMPM plates did not reveal blue or white Listeria like colonies, up to two Listeria-like colonies from MOX were sub-streaked to LMPM for identification. For each sample, up to four isolates representing putative L. monocytogenes and up to two isolates representing putative Listeria spp. other than L. monocytogenes were grown in BHI overnight then frozen at -80°C in 15% glycerol. For each sample, one putative L. monocytogenes and one putative Listeria spp., if present, were further characterized and confirmed by sigB sequencing; confirmed L. monocytogenes were subtyped with Pulsed-Field Gel Electrophoresis. sigB PCR-based identification Listeria spp. A previously described approach that uses sequencing of the gene sigB was used to (i) confirm isolates as Listeria spp. and to (ii) classify isolates to the species level; this approach has been shown to provide for rapid, economical, and reliable species identification (11). PCR and sequencing were performed as previously described
16

(11). sigB sequences were compared to the Food Safety Laboratory sigB allelic type database, which allows for classification to species and characterization to allelic type. An allelic type was defined as a distinct sequence for the 660 nt sigB fragment sequenced; two isolates were assigned a different allelic type if they differed by at least one nucleotide among the 660 nt. Pulsed-Field Gel Electrophoresis (PFGE). L. monocytogenes PFGE (including plug preparation, restriction digestion with AscI and ApaI, and gel electrophoresis) was performed according to the standardized CDC PulseNet protocol (Available at: http://www.cdc.gov/pulsenet/PDF/listeria-pfge-protocol-508c.pdf) with minor modifications as needed. For example, SDS was omitted for some samples to reduce degradation of large DNA fragments. Patterns were analyzed using BioNumerics 5.10 software (Applied Maths, Inc., Austin, TX) and unique identifiers were assigned to AscI and ApaI patterns. For example, two isolates that share AscI pattern CU-258 and ApaI pattern CU-322 would both assigned the PFGE type CU-258,322 while isolates that share the same AscI pattern CU-258, but have different ApaI patterns (e.g. CU-323 and CU-333) would be assigned PFGE patterns CU-258,323 and CU258,333.
RESULTS AND DISCUSSION Pre-operational sampling reveals between 15 and 20% L. monocytogenes prevalence on deli floor and floor–wall juncture. Overall, 314 pre-operational samples were collected from 15 deli establishments over 3 visits to each deli (Table 2.1). Among these samples, 6.8% were positive for L. monocytogenes. Overall, the floor-wall juncture under the 3-basin showed the highest L. monocytogenes prevalence among pre-operational samples (18%, Table 2.1) followed by the deli floor adjacent to the drain (16%, Table 2.1). There were 8 delis where no L. monocytogenes
17

positive samples were found pre-operation. Three and two delis had 1/21 and 2/21 positive samples, respectively. In deli 7, 8/21 samples were positive and in deli 2, 5/21 samples were positive for L. monocytogenes. In deli 7, the floor-wall juncture under the 3-basin was positive for L. monocytogenes at each of the three monthly samplings, while in deli 2 the floor adjacent to the drain was positive at each of the three monthly samplings. These data indicate that current cleaning and sanitation protocols and execution do not eliminate L. monocytogenes from nonfood contact surfaces in all stores.

Table 2.1 L. monocytogenes and Listeria spp. prevalence among 7 sites sampled preoperationally in 15 stores

Sample sites Pans
Cold room racks
Wheeled carts

Description Aluminum, stainless steel, plastic pans used for deli meats
Rack in cold room used to store meat

No. (%) of samples positive for

L. monocytogenes
1/44 (2%)

Listeria spp.a
0/44 (<2%)

2/45 (4%)

1/45 (2%)

Mobile carts used only in the deli area to transport food.

2/45 (4%)

2/45 (4%)

Deli area floor adjacent to the drain

Floor in high traffic area of deli

7/45 (16%)

Trash can

Large trash can used only in the deli area

1/44 (2%)

Floor-wall juncture under 3-basin The wall/ floor juncture underneath 3-basin sink

7/45 (18%)

Slicer casing

Casing below the blade

0/45 (<2%)

aListeria spp. indicates Listeria spp. other than L. monocytogenes

4/45 (9%) 1/44 (2%) 4/45 (9%) 0/45 (<2%)

18

Longitudinal operational sampling identified deli establishments with low (<1%) and high (>10%). L. monocytogenes prevalence among sites. In Phase II of this study, a total of 4,503 samples were collected and tested from 6 monthly sampling visits to each of 30 delis; these samples were collected during delicatessen operation and between 20 and 28 samples were collected during each visit from each deli. Phase I and II sampling results, and subsequent subtyping, for a single deli are shown in Figure 2.1, with the complete data set for all 30 stores presented in Appendix Figures 1 – 30. Among these 4,503 sites sampled, 9.5% were positive for L. monocytogenes; 9 of 30 delis showed low L. monocytogenes prevalence (<1%); 13 showed between 1 and 10% and 8 showed >10% L. monocytogenes prevalence (Figure 2.2). The overall L. monocytogenes prevalence among environmental sites in deli establishments is comparable to a 13.0% prevalence previously reported among a total of 1,161 environmental samples collected in a cross-sectional study of 121 deli establishments located in New York state (18). Phase II L. monocytogenes prevalence for FCS, NFCS, and TP was 4.5, 14.2, and 3.3%, respectively. L. monocytogenes prevalence by site ranged from 0.6% (re-wrap table) to 34.5% (cold room drain) (Figure 2.2). These L. monocytogenes prevalence data are consistent with prevalence data from other types of environments. For example, a study on urban environments in New York state reported an L. monocytogenes prevalence of 7.5% (based on 898 samples collected in 4 cities) (15), while another study reported an L. monocytogenes prevalence of 14.6 to 35.3% in soil samples collected from farms with and without a history of animal listeriosis (12). These data highlight the likelihood of L. monocytogenes introduction into retail and deli establishments from outside sources.
19

Deli-8
Figure 2.1 Complete environmental sampling and Listeria subtyping results from both Phase I and Phase II for one example deli, Deli 8. The top panel shows the sampling results for each site over each of 9 sampling months, sorted by contact surface type. Both L. monocytogenes PFGE subtype and Listeria species for an L. spp. isolate are shown. Sites that have no samples were
20

41/119

46/180

36/130

31/110

30/164

16/86

36/180

36/179

PrevaPlreevnalceence

23/179

24/179

14/180

14/179

included as they were sampled in other delis in the data set. The bottom panel shows the band patterns from two enzyme PFGE of all L. monocytogenes isolates and the CU pattern number assigned to those patterns (the PFGE subtype). Each cell describes a single unique sample, where ‘–‘ indicates sampling was negative for L. monocytogenes, ‘.’ indicates the site was not sampled, a ‘CU-‘ number indicates the PFGE subtype assigned to the L. monocytogenes isolate, and the ‘in’ after the ‘;’ indicates the L. spp enrichment for that site resulted in an isolated identified as L. innocua. Analogous panels for all 30 delis are found in Appendix Figures 1 – 30.
00..4400
00..3355
00..3300
00..2255
00..2200
00..1155
00..1100
00..0055
00..0000
Site
Figure 2.2 L. monocytogenes prevalence for the different sites that were sampled during processing (i.e., in Phase II). Numbers above the bars indicate the number of positive samples/total number of samples tested for a given site; sites are sorted by decreasing prevalence. Details on these sites as well as L. monocytogenes and Listeria spp. prevalence data for these sites are provided in Appendix Figure 1 – 30.
21

Rewrap1t/a1b7l9e 1/179

Slicer3/k1n8o0b 3/180

3/152

4/180

4/180

4/180

4/179

5/180

5/178

7/180

7/180

7/164

8/180

6/134

9/180

2/32

Cutting 3b/o1a5r2d

Deli case4t/r1a8y0s

Cold room4/1wa8l0l

Cold room4/r1a8c0k

Deli4/c1a7s9e

S5l/i1ce8r0

Co5u/n1t7e8r

S7/c1al8e0

3-basin7/s1i8n0k…

1-basin7/s1i6n4k…

Deli Case H8a/n1d8l0e

6/H1o3s4e

Tras9h/1c8a0n

Deli case 2n/e3a2r…

14/C1ar8t0s

3-basi1n4/s1i7n9k…

Del2i3f/l1o7o9r

Floo2r4//1wa7l9l…

1-basi3n0/s1i6n4k…

Standing 1w6a/t8e6r

Deli36d/r1a8i0n

Cold roo 3m6f/l1o7o9r

Floor adja4ce6/nt18t0o…

Sq3ue6/e1g3e0e

Floo3r1//1wa1l0l…

Cold room41d/r1a1i9n

Overall, 237/4,503 samples (5.3%) were positive for Listeria species other than L. monocytogenes, including sites positive for both L. monocytogenes and other Listeria species. Listeria spp. in this study refers to all species excluding L. monocytogenes. Listeria spp. prevalence by site ranged from 0% for the re-wrap table to 16.8% for the cold room drain (Appendix Figures 1 – 30). For most sites, the prevalence of other Listeria spp. was lower than L. monocytogenes prevalence (Table 2.1). This is consistent with the lower overall prevalence for Listeria spp. (5.4%) as compared to L. monocytogenes (9.5%). These data suggest that testing for L. monocytogenes may be a more appropriate strategy for environmental monitoring of retail environments than testing for general Listeria spp. as an indicator, a conclusion consistent with a study showing Listeria spp. may not be a good indicator for presence of L. monocytogenes on most types of surfaces in seafood processing plants (13). In addition, L. monocytogenes testing results will allow for a focus on interventions that target the presence of the actual pathogen. L. innocua is the most common Listeria spp. isolated aside from L. monocytogenes. Characterization of 245 putative Listeria spp. isolates (12 isolates obtained in Phase I and 237 isolates from Phase II) by sigB PCR and sequence analysis classified these isolates into three species, including L. innocua (184 isolates), L. seeligeri (48 isolates) and L. welshimeri (13 isolates). These findings are consistent with previous studies, which showed that L. innocua is the predominant Listeria spp. in urban environments (17).
sigB sequence data also provided an opportunity for initial subtype discrimination among the Listeria spp. isolates. Overall, 17 sigB allelic types were differentiated among the 249 Listeria spp. isolates characterized, with 8, 5, and 4 allelic types found among the L. innocua, L. seeligeri, and L. welshimeri isolates (Table 2.2). Consistent with previous studies that have shown limited discriminatory capability of sigB allelic typing within a given species, we also
22

identified a number of common sigB allelic types that were found among delis in different states. For example, L. innocua isolates with allelic type AT 11 were found in 13 delis that were located across the three different states represented by the delis included in our study (Table 2.2). Similarly, 65 L. innocua isolates with AT 37 were obtained from 13 delis among three states.

Table 2.2 Listeria species and sigB allelic types isolated from the 30 retail deli environments sampled.

Allelic Type (AT)

No. of Delis where an AT was isolateda

L. innocua AT 11 AT 37

13 13

AT 22

7

AT 56

3

AT 31

2

AT 71 AT 116

1 1

AT 145

1

L. seeligeri AT 3 AT 24

8 3

AT 20

2

AT 35 AT 92

1 1

L. welshimeri AT 27

5

AT 32

2

AT 15 AT 89

1 1

L. monocytogenes lineage I AT 63

22

AT 64 AT 58

8 8

AT 61

9

AT 60

7

L. monocytogenes II AT 57

6

AT 144

1

No. of isolates a
79 65 27 7 2 2 1 1
26 16 2 3 1
7 4 1 1
177 113 91 24 11
10 1

a Data shown include all isolates from Phase I and II, except for 18 L. monocytogenes and 3 L. spp. isolates that were not available for sigB allelic typing

23

Some PFGE types are common and found among multiple delis and states, including one PFGE type isolated from 12 different delis. PFGE analysis of 446 L. monocytogenes isolates obtained in this study (i.e., in both the pre-operational “Phase I” sampling and during operation “Phase II” sampling) identified a total of 24 AscI and 39 ApaI PFGE types. Combined analyses allowed for classification of these isolates into 44 two-enzyme PFGE types; 25 of these PFGE types were isolated at least twice. PFGE types of L. monocytogenes isolates from each deli for each month and sample site are available in Appendix Figures 1 – 30. A number of PFGE types were isolated from multiple deli establishments (Table 2.3), with the most highly dispersed being 113 isolates of PFGE type CU-11,320 found among 11 delis in 3 states. Previous studies have also shown that some L. monocytogenes PFGE types are common and can be isolated from diverse sources. For example, Fugett et al. (4) not only showed that L. monocytogenes isolates from listeriosis outbreaks in Los Angeles (in 1985) and Switzerland (1983 - 1987) represented the same PFGE type, but also found that 7 isolates collected from farms and 3 isolates from environmental sources in New York state matched this same PFGE type. Our findings are also consistent with a number of previous studies that suggested limited discriminatory power of PFGE for some L. monocytogenes strains and particularly strains representing serotype 4b (10, 23). Alternatively, our data may suggest that some of the delis here may share common sources (e.g., suppliers) that lead to re-introduction of these PFGE types. Future characterization of isolates representing these common PFGE types with advanced subtyping methods with superior discriminatory power (e.g. full genome sequencing) may be able to provide further insight into the relatedness of isolates that share the same PFGE type but were obtained from different delis, including delis located in different states.
24

Table 2.3 PFGE types isolated from 2 or more delis.

PFGE type CU-11-320 CU-40-96 CU-258-69 CU-294-321 CU-262-79 CU-8-96 CU-57-267 CU-299-338 CU-258-323 CU-200-227 CU-258-333 CU-258-322 CU-143-316 CU-262-318 CU-11-326 CU-258-67 CU-82-215

No. of delisa 11 9 7 7 6 6 6 4 3 3 3 2 2 2 2 2 2

No. of isolatesb 113 12 74 12 61 28 14 4 14 3 3 13 4 4 3 2 2

No. of statesc 3 2 2 2 2 2 1 2 1 1 2 1 1 2 1 2 1

aNumber of delis in which a given PFGE type was found at least once;
bTotal number of isolates with a given PFGE type (among all 446 isolates from both Phase I and II); cNumber of states that were represented among the delis that were positive for a given PFGE type; delis sampled represented 3 different states.

Based on both L. monocytogenes prevalence data and PFGE, we classified the 30 delis studied into four categories: (i) delis with low prevalence and no evidence for re-isolation of a given PFGE type (12 delis); (ii) delis with low prevalence, but a single sampling event with a high L. monocytogenes prevalence (4 delis); (iii) delis with high prevalence and re-isolation of one or more L. monocytogenes PFGE types over 3 or more sampling visits (12 delis); and (iv) delis with low prevalence and repeated isolation of the same PFGE type on two separate sampling dates (2 delis). Details on these four categories are provided in the sections below. Twelve deli establishments show low L. monocytogenes prevalence and no evidence of
25

repeated isolation of a given PFGE types. The twelve delis that showed low prevalence of L. monocytogenes included 3, 6 and 2 delis with 0, 1, and 2 positive samples, respectively (with 129 to 156 samples collected per deli; Figure 2.3). In addition, one deli with 4 positive samples (deli 26) was classified into this group; three different PFGE types were obtained with both isolates from sampling 4 showing the same PFGE type. Our observation that approx. 40% of deli establishments showed low prevalence is consistent with a previous study that analyzed crosssectional data from 241 delis in NY state (7); this study showed that L. monocytogenes was not detected in 40% of delis (even though only up to 16 samples were tested per establishment in this prior study). Combined, these data show that a considerable proportion of delis in the US show low L. monocytogenes prevalence; these data also provide benchmarks for future studies. Four delis showed low overall L. monocytogenes prevalence, but had a single sampling event with high L. monocytogenes prevalence. Four delis showed high L. monocytogenes prevalence in one single sampling, while having a low L. monocytogenes prevalence in the other 5 operational samplings. In all 4 delis, L. monocytogenes was isolated from NFCS, FCS, and TP during the sampling that represented a single high prevalence event, suggesting widespread cross-contamination and possible introduction of a high L. monocytogenes load. Three of the 4 delis had no positive samples in 5 of the 6 monthly samplings (delis 12, 17, and 27; see Figure 2.4) and one that had a single positive among the other 5 samplings (deli 22; see Figure 2.4). In deli 12, 5/23 sites samples were positive in November, including 1/9 FCS samples and 2/3 TP; isolates from 4 sites shared the same PFGE type (CU-11,320).
26

Prevalence

0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
Store
Figure 2.3 L. monocytogenes prevalence by deli for samples collected during processing (i.e., in Phase II). Numbers above the bars indicate the number of positive samples/total number of samples tested for a given deli; delis are sorted by creasing prevalence.
27

In deli 17, 5/24 samples tested in November were positive for L. monocytogenes including 1/9 FCS samples and 1/3 TP; the isolates from all 5 sites shared the same PFGE type (CU-11,320). In deli 22, 13/24 samples tested in August were positive for L. monocytogenes including 5/9 FCS samples and 3/3 TP; the isolates for 10 of these 13 sites showed the same PFGE type (CU-258,69). In deli 27, 11/23 samples tested in August were positive for L. monocytogenes including 5/9 FCS samples and 2/3 TP; the isolates for 10 of these sites had the same PFGE type (CU-258,69).
While our data do not provide specific evidence for the causation of these high prevalence events, it is intriguing that the two high prevalence events in August were caused by the same specific PFGE type (type CU-258,69) as were the two high prevalence events in November (PFGE type CU-11,320). While these subtype-based observations may indicate a common source for each the two high prevalence events in August and November (e.g., a common supplier), further efforts are needed to explore sources and prevention strategies of these events. The occurrence of sporadic high prevalence events is consistent with a previous study that reported a single high prevalence contamination event with Listeria welshimeri in the environment of a food processing plant; this event was limited to one sampling time on one day in a study that included daily sampling three times a day (8).
28

Store
1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 6 6 6 7 7 7 8 8 8 9 9 9 10 10 10 11 11 11 12
12 12 13 13 13 14 14 14 15 15
15

Site Type JUL.
FCS 0/9 NFC 0/11
TP 0/3 FCS 0/8 NFC 4/14
TP 1/3 FCS 1/9 NFC 0/14
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3 FCS 0/7 NFC 0/12
TP 0/3 FCS 2/9 NFC 10/14
TP 0/3 FCS 0/9 NFC 5/14
TP 0/3 FCS 0/8 NFC 0/10
TP 0/3 FCS 1/8 NFC 7/14
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3 FCS 0/6
NFC 0/11 TP 0/3
FCS 0/9 NFC 0/14
TP 0/3 FCS 0/9 NFC 0/12
TP 0/3 FCS 0/9 NFC 0/12
TP 0/3

AUG.
0/9 0/11 0/3 5/9 7/14 2/3 0/9 0/14 0/3 0/8 0/14 0/3 0/9 0/14 0/3 0/7 0/12 0/3 0/9 8/14 0/3 0/9 4/14 0/3 0/8 0/10 0/3 3/8 8/13 1/3 0/9 0/14 0/3 0/10 0/10 0/3 0/9 0/14 0/3 0/9 0/11 0/3 0/9 0/12 0/3

Panel A - Stores 1-15 L. monocytogenes

SEPT.
0/9 0/11 0/3 0/9 6/14 0/3 0/9 0/14 0/3 0/9 0/14 0/3 0/9 0/14 0/3 0/9 0/14 0/3 0/10 1/15 0/3 0/9 1/14 0/3 0/8 0/10 0/3 1/9 7/14 0/3 0/9 0/14 0/3 0/9 0/11 0/3 1/9 0/14 0/3 0/9 0/12 0/3 0/9 0/13 0/3

OCT.
0/9 0/13 0/3 0/9 5/15 1/3 0/9 0/14 0/3 0/9 2/14 0/3 0/9 0/14 0/3 0/7 1/12 0/3 0/9 7/14 0/3 1/9 0/14 0/3 0/8 0/11 0/3 1/8 3/14 0/3 0/9 0/14 0/3 0/9 0/11 0/3 0/9 2/14 0/3 0/9 0/10 0/3 0/9 0/12 0/3

NOV.
0/9 1/11 0/3 0/10 5/14 0/3 0/9 0/14 0/3 0/10 1/14 0/3 0/9 0/14 0/3 0/10 0/14 0/3 1/9 8/14 0/3 1/9 1/14 0/3 0/8 1/11 0/3 1/9 2/14 0/3 0/9 0/14 0/3 1/9 2/11 2/3 2/9 1/14 0/3 0/9 1/11 0/3 0/9 0/12 0/3

DEC.
0/9 0/11 0/3 1/9 8/15 0/3 0/9 0/14 0/3 1/8 1/14 0/3 0/9 0/14 0/3 0/7 0/12 0/3 0/9 8/14 0/3 0/9 0/14 0/3 0/8 0/11 0/3 1/8 3/14 0/3 1/9 1/14 0/3 0/9 0/11 0/3 0/9 2/14 0/3 0/9 0/11 0/3 0/9 0/12 0/3

TOT.
0/54 1/68 0/18 6/54 35/86 4/18 1/54 0/84 0/18 1/53 4/84 0/18 0/54 0/84 0/18 0/47 1/76 0/18 3/55 42/85 0/18 2/54 11/84 0/18 0/48 1/63 0/18 8/50 30/83 1/18 1/54 1/84 0/18 1/52 2/65 2/18 3/54 5/84 0/18 0/54 1/67 0/18 0/54 0/73 0/18

JUL.
0/9 0/11 0/3 0/8 6/14 0/3 0/9 0/14 0/3 0/9 0/14 0/3 0/9 0/14 0/3 0/7 0/12 0/3 1/9 7/14 0/3 0/9 0/14 0/3 0/8 0/10 0/3 0/8 0/14 0/3 0/9 0/14 0/3 0/6 0/11 0/3 0/9 0/14 0/3 0/9 0/12 0/3 0/9 0/12 0/3

L. spp not L. monocytogenes
AUG. SEPT. OCT. NOV. DEC. TOT.
0/9 0/9 0/9 2/9 0/9 2/54 0/11 0/11 0/13 1/11 0/11 1/68 0/3 0/3 0/3 1/3 0/3 1/18 0/9 0/9 0/9 0/10 1/9 1/54 4/14 2/14 4/15 0/14 2/15 18/86 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/9 0/9 0/9 0/9 0/54 0/14 0/14 0/14 0/14 1/14 1/84 0/3 0/3 0/3 0/3 0/3 0/18 1/8 0/9 0/9 0/10 1/8 2/53 0/14 0/14 2/14 0/14 3/14 5/84 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/9 0/9 0/9 0/9 0/54 0/14 0/14 0/14 0/14 0/14 0/84 0/3 0/3 0/3 0/3 0/3 0/18 0/7 0/9 0/7 0/10 0/7 0/47 0/12 2/14 0/12 0/14 0/12 2/76 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/10 0/9 1/9 0/9 2/55 7/14 5/15 3/14 7/14 8/14 37/85 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/9 0/9 0/9 0/9 0/54 1/14 0/14 0/14 0/14 0/14 1/84 0/3 0/3 0/3 0/3 0/3 0/18 0/8 0/8 0/8 0/8 0/8 0/48 0/10 3/10 1/11 3/11 0/11 7/63 0/3 0/3 0/3 1/3 0/3 1/18 0/8 0/9 0/8 1/9 1/8 2/50 0/13 0/14 0/14 0/14 0/14 0/83 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/9 0/9 0/9 0/9 0/54 0/14 0/14 0/14 0/14 1/14 1/84 0/3 0/3 0/3 0/3 0/3 0/18 0/10 0/9 0/9 1/9 0/9 1/52 0/10 0/11 3/11 4/11 2/11 9/65 0/3 0/3 0/3 2/3 0/3 2/18 0/9 0/9 0/9 0/9 0/9 0/54 0/14 0/14 0/14 0/14 0/14 0/84 0/3 0/3 0/3 0/3 0/3 0/18 0/9 0/9 0/9 0/9 0/9 0/54 0/11 0/12 0/10 1/11 0/11 1/67 0/3 0/3 0/3 0/3 0/3 0/18 1/9 0/9 1/9 1/9 0/9 3/54 1/12 0/13 0/12 0/12 1/12 2/73 0/3 0/3 0/3 0/3 0/3 0/18

29

Store 16 16 16 17 17 17 18 18 18 19 19 19 20 20 20 21 21 21 22 22 22 23 23 23 24 24 24 25 25 25 26 26 26 27 27
27 28 28 28 29 29 29 30 30 30

Site Type AUG. FCS 1/10 NFC 4/14
TP 0/3 FCS 0/9 NFC 0/12
TP 0/3 FCS 1/9 NFC 0/14
TP 0/3 FCS 0/9 NFC 0/11
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3 FCS 1/8 NFC 5/14
TP 0/3 FCS 5/9 NFC 5/12
TP 3/3 FCS 1/10 NFC 9/15
TP 0/3 FCS 2/8 NFC 4/12
TP 0/3 FCS 0/9 NFC 1/12
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3 FCS 5/9 NFC 4/11
TP 2/3 FCS 0/9 NFC 9/15
TP 0/3 FCS 1/9 NFC 3/14
TP 0/3 FCS 0/9 NFC 0/14
TP 0/3

SEPT.
0/9 2/14 0/3 0/9 0/12 0/3 0/9 0/14 0/3 1/9 0/11 0/3 0/9 0/14 0/3 0/8 5/14 0/3 0/9 0/11 0/3 3/10 8/14 1/3 0/8 6/11 0/3 0/9 0/13 0/3 0/9 0/14 0/3 0/9 0/12 0/3 0/9 9/14 0/3 1/9 1/14 0/3 0/9 0/14 0/3

Panel B - Stores 16-30 L. monocytogenes

L. spp not L. monocytogenes

OCT.
1/10 1/14 0/3 0/9 0/13 0/3 1/9 0/14 0/3 0/9 1/11 0/3 0/9 0/14 0/3 1/10 3/13 0/3 1/9 0/11 0/3 1/10 7/15 0/3 0/9 8/13 0/3 0/9 0/13 0/3 0/9 1/14 0/3 0/9 0/12 0/3 0/10 9/14 0/3 0/9 2/14 0/3 0/9 0/14 0/3

NOV. DEC. JAN. TOT. AUG. SEPT. OCT. NOV. DEC. JAN.
1/10 3/10 1/9 7/58 4/10 0/9 1/10 2/10 2/10 0/9 3/15 4/15 0/14 14/86 3/14 1/14 4/14 1/15 4/15 4/14 0/3 1/3 0/3 1/18 0/3 0/3 0/3 0/3 0/3 0/3 1/9 0/9 0/9 1/54 0/9 0/9 0/9 1/9 0/9 0/9 3/12 0/12 0/12 3/73 1/12 0/12 0/13 3/12 0/12 0/12 1/3 0/3 0/3 1/18 0/3 0/3 0/3 1/3 0/3 0/3 0/9 0/9 1/9 3/54 0/9 1/9 0/9 0/9 0/9 0/9 0/14 0/14 2/14 2/84 0/14 0/14 0/14 0/14 2/14 0/14 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 1/9 1/9 3/54 0/9 0/9 0/9 0/9 0/9 0/9 0/12 0/11 0/11 1/67 0/11 0/11 0/11 0/12 0/11 0/11 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 1/9 1/54 0/9 0/9 0/9 0/9 0/9 1/9 1/14 0/14 0/14 1/84 0/14 0/14 0/14 0/14 0/14 0/14 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 2/9 0/10 0/8 4/53 1/8 0/8 0/10 2/9 0/10 0/8 2/14 4/14 4/14 23/83 3/14 0/14 1/13 0/14 4/14 6/14 1/3 0/3 0/3 1/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 0/9 6/54 1/9 0/9 0/9 0/9 0/9 0/9 0/12 0/12 0/12 5/70 2/12 0/11 0/11 0/12 0/12 0/12 0/3 0/3 0/3 3/18 0/3 0/3 0/3 0/3 0/3 0/3 1/10 0/10 1/9 7/59 1/10 1/10 1/10 1/10 1/10 0/9 5/14 4/15 9/15 42/88 3/15 1/14 1/15 1/14 1/15 3/15 0/3 0/3 0/3 1/18 0/3 0/3 0/3 0/3 0/3 0/3 1/8 0/8 0/8 3/49 0/8 0/8 0/9 1/8 0/8 0/8 10/13 7/13 6/12 41/74 0/12 0/11 0/13 6/13 5/13 3/12 0/3 0/3 1/3 1/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 0/9 0/54 0/9 0/9 0/9 0/9 0/9 0/9 0/12 0/11 0/11 1/72 1/12 0/13 0/13 0/12 0/11 0/11 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 0/9 0/54 0/9 0/9 0/9 0/9 0/9 0/9 2/14 1/14 0/14 4/84 0/14 0/14 1/14 0/14 0/14 0/14 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 0/9 5/54 1/9 0/9 0/9 0/9 0/9 0/9 0/12 0/11 0/11 4/69 0/11 0/12 0/12 0/12 0/11 0/11 0/3 0/3 0/3 2/18 1/3 0/3 0/3 0/3 0/3 0/3 0/8 0/9 1/8 1/53 0/9 0/9 0/10 0/8 0/9 0/8 9/15 8/15 9/15 53/88 0/15 0/14 4/14 5/15 8/15 2/15 1/3 0/3 0/3 1/18 0/3 0/3 0/3 0/3 0/3 0/3 1/9 1/9 1/9 5/54 0/9 1/9 0/9 1/9 0/9 1/9 0/14 0/14 2/14 8/84 0/14 2/14 4/14 2/14 3/14 2/14 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3 0/9 0/9 0/9 0/54 0/9 0/9 0/9 0/9 0/9 0/9 0/14 0/14 0/14 0/84 0/14 0/14 5/14 2/14 2/14 1/14 0/3 0/3 0/3 0/18 0/3 0/3 0/3 0/3 0/3 0/3

TOT.
9/58 17/86 0/18 1/54 4/73 1/18 1/54 2/84 0/18 0/54 0/67 0/18 1/54 0/84 0/18 3/53 14/83 0/18 1/54 2/70 0/18 5/59 10/88 0/18 1/49 14/74 0/18 0/54 1/72 0/18 0/54 1/84 0/18 1/54 0/69 1/18 0/53 19/88 0/18 3/54 13/84 0/18 0/54 10/84 0/18

30

Figure 2.4 Heat map showing L. monocytogenes and Listeria spp. prevalence data for samples collected during processing (Phase II) for (A) delis 1 to 15 and (B) delis 16 to 30. Prevalence is shown as the number of positive samples/total samples tested for a given sampling site category and a given month (abbreviated as JUL, AUG, etc.) or the total number of samples collected over 6 month (abbreviated as “TOT”). Samples site categories are abbreviated as FCS (food contact surface sites), NFC (non-food contact surface sites) and TP (transfer point); see Appendix Figures 1 – 30 for detailed data and description of the specific sites in each sample site category)
In twelve deli establishments, one or more PFGE types were isolated over multiple sample visits. We isolated one or more PFGE types from at least three different sampling dates (Table 2.4, for example CU-8,340 in example Deli 8 in Fig 1) in twelve delis. Deli 2 had four different PFGE types isolated on at least three separate sampling dates; delis 10 and 23 had three different PFGE types isolated on at least three separate sampling dates. Delis 21, 24, 28, and 29 had two different PFGE types were each isolated on at least three separate sampling dates. In 3 delis a single PFGE type was isolated in each of the six phase II samplings; in one additional deli (deli 28), two different PFGE types were each isolated in each of the 6 Phase II samplings. Reisolation of the same L. monocytogenes subtype (e.g., PFGE type) is often interpreted as indicating persistence of a given subtype in the environment sampled (3), particularly if sampling occurred in a relatively closed environment (e.g., a processing plant) where repeat reintroduction from an outside source is unlikely and is prevented through extensive barriers and appropriate GMPs. Due to the fact that the deli are “open” environments and not isolated from surrounding areas (e.g., other parts of the retail deli, outside areas such as parking lots), it is not possible to rule out that the PFGE data reported here do not necessarily indicate persistence of specific PFGE types in the 12 delis. While these “re-isolated” PFGE types may persist surrounding environments or even in an up-stream processing environment, it is likely that persistence occurs in the actual deli environment. This is supported by the fact that in 5 of the 6
31

delis showed re-isolation during operation (Phase II) and at least one PFGE type that was reisolated was also isolated in the pre-op sampling (Table 2.4). As presence of a PFGE type after cleaning and before operation is initiated is unlikely to present re-introduction, these data support persistence of at least some of the re-isolated PFGE types in the respective delis. Persistence is also supported by the observation that for three of the delis (delis 2, 7, and 10) with re-isolation of a given PFGE type, this PFGE type was isolated at least once at a preoperational (Phase I) sampling from the floor/wall juncture sample (Appendix Figures 1 – 30), which is well established as possible niche that allows Listeria survival over time. In one of these delis (deli 7), a given PFGE type (CU-258,69) was isolated from the floor wall juncture at each of the three pre-operational samplings as well as at 5 of the 6 Phase II samplings which occurred during processing, providing particularly strong evidence for persistence in the actual deli establishments. Overall, our findings support the importance of developing and implementing strategies to identify, control, and manage L. monocytogenes persistence in deli establishments.
Overall, 15 different PFGE types were found to be isolated over at least three different sampling dates in a given deli. Three PFGE types were found to be re-isolated in multiple deli establishments, including (i) PFGE type CU-11,320 (re-isolated in four delis); (ii) PFGE type CU-262,79 and CU-8,96 (re-isolated in three delis), and (iii) PFGE type CU-57,267 (re-isolated in two delis). Overall, a total of 23 instances of re-isolation of a given PFGE type were observed. While two of the three most common PFGE types identified here (CU-11,320 and CU-262,79) were re-isolated in multiple delis, the second most common PFGE type isolated here (CU258,69; isolated 74 times and in delis from all three states; Table 2.3) was only re-isolated in one deli (deli 7; Table 2.4).
32

Table 2.4 Re-isolation of PFGE types from a given deli

Deli PFGE Typea 2 CU-11,320 CU-55,266 CU-262,79 CU-8,96

No. isolates (FCS , TP)b
17 (1, 1) 9 (4, 2) 14 (0,0) 4 (0,0)

APR. 1 0 0 1

No. of isolates with a given PFGE types found in:

MAY 0 0 1 1

JUNE 0 0 1 0

JULY 2 2 1 0

AUG. 3 6 2 0

SEPT. 4 0 1 1

OCT. 3 1 0 1

NOV. 5 0 0 0

DEC. 0 0 8 0

4 CU-57,267

4 (0,0)

0 0 0 0 0 0 211

7 CU-258,69

48 (1,1)

2 2 4 11 6 0 7 8 8

8 CU-8,340

10 (0,0)

1 0 0 5 2 1 010

10 CU-11,320 CU-57,267 CU-294,321

29 (1,0) 4 (3,0) 3 (3,0)

0 0 1 7 9 6 213 0 0 0 0 0 1 120 0 0 0 1 1 0 001

13 CU-11,282

7 (1,0)

1 0 0 0 0 0 222

16 CU-8,96

15 (1,1) NT a NT NT NT 3 2 1 3 6

21 CU-11,320 CU-262,318

18 (0,0) 3 (0,0)

NT NT NT NT 4 NT NT NT NT 0

4 303 1 011

23 CU-258,322 CU-258,323 CU-259,322

12 (0,1) 11 (0,0) 9 (1,0)

NT NT NT NT 2 NT NT NT NT 2 NT NT NT NT 3

4 211 3 112 2 011

24 CU-11,320 CU-8,96

31 (1,0) 6 (0,0)

NT NT NT NT 5 NT NT NT NT 0

3 666 3 020

28 CU-262,79 CU-262,319

32 (0,1) 20 (1,0)

NT NT NT NT 8 NT NT NT NT 1

7 455 2 443

29 CU-40,96

3 (3,0)

NT NT NT NT 1 0 0 0 1

CU-262,79

6 (0,0)

NT NT NT NT 3 0 2 0 0

a Only PFGE types that were isolated three or more times in a given deli establishment were listed.

b Number of isolates (number of isolates from FCS, number of isolates from TP).

c NT, not tested; this indicates that no samples were collected during that month.

JAN NTc NT NT NT
NT
NT
NT
NT NT NT
NT
0
4 0
2 2 2
5 1
3 6
1 1

33

Our data suggest that a number of different L. monocytogenes subtypes can persist over time in a given source (either the deli establishment itself or another environment that serves as a source for re-introduction), consistent with the hypothesis that persistence of L. monocytogenes does not just represent the ability of specific strains to establish persistence, but likely requires the presence, in a given environment, of growth niches that support persistence and growth (1, 5, 13, 22).
Further analysis of the re-isolation patterns observed here (representing 23 instances of re-isolation of a given PFGE types across 12 delis) also provided an opportunity to further characterize potential patterns and frequency of cross-contamination between non-FCS and FCS sites. Among the 23 instances of re-isolation of a given PFGE type, on 9 instances the re-isolated PFGE type was only obtained from non-FCS and in one instance the re-isolated PFGE type was only obtained from FCS. In 13 instances the re-isolated PFGE type was obtained from both nonFCS as well as FCS and/or transfer points (TP), suggesting at least some cross-contamination. In 7 and 3 instances the PFGE types were only isolated once or twice, respectively, on FCS and/or TPs, despite sometimes frequent isolation from non-FCS, suggesting occasional but not frequent cross contamination in these delis. For example, in deli 7, PFGE type CU-258,69 was represented by 48 isolates from non-FCS, but was only represented by one FCS isolate. In one instance (PFGE type CU-55,266 in deli 2), isolates with this PFGE type were obtained from FCS (3 isolates), non-FCS (4 isolates), and TPs (2 isolates), suggesting frequent cross contamination. In another instance (PFGE type CU-57,267 in deli 10), three isolates with PFGE type CU-57,267 were obtained from FCS sites, while one isolate with this PFGE type was obtained from a nonFCS sites; this suggests either L. monocytogenes persistence in a FCS site or repeated reintroduction to an FCS site.
34

Two delis had low overall L. monocytogenes prevalence with re-isolation of the same PFGE type on two separate sampling dates. In two deli establishments (delis 18 and 19) that showed low L. monocytogenes prevalence (5/156 and 4/139 samples positive for L. monocytogenes, respectively), a given PFGE type was isolated on two different sampling dates. In deli 18, PFGE type CU-40,96 was isolated twice from the cold room rack (August and October) (Appendix Figures 1 – 30). In deli 19, PFGE type CU-182,173 was isolated twice (September and January) from the slicer. As discussed above, we cannot determine whether these instances represent persistence or re-introduction of these PFGE types. Nevertheless, our data indicate that even in delis with lower L. monocytogenes prevalence, common source events can be responsible for contamination of food contact surfaces, which represent a considerable risk for product contamination. Identifying the sources of these low frequency contamination events is likely to be challenging however.
Overall, our data indicate that some retail deli establishments are able to control L. monocytogenes in their environments, including some delis where (i) L. monocytogenes is rarely isolated from environmental sites and (ii) where L. monocytogenes is found in NFCS sites with no evidence for cross-contamination to FCS and TP. In some delis where L. monocytogenes is regularly found in NFCS sites, PFGE data support cross-contamination between NFSCS and FCS, even though the frequency of cross-contamination typically seems to be low. Our data suggest that L. monocytogenes persistence occurs in a number of deli establishments therefore adaptation of strategies that have been used by Ready-To-Eat food processors to control L. monocytogenes persistence and to eliminate growth niches may also be beneficial. This study also underscores that continued efforts are needed to control and reduce L. monocytogenes contamination in deli environments. A wide range of strategies, including reformulation of deli
35

meats with growth inhibitors, enhanced sanitation, and rigorous temperature control, may be effective, as detailed in a recent risk assessment (21).
36

REFERENCES 1. Autio, T., J. Lunden, M. Fredriksson-Ahomaa, J. Bjorkroth, A. M. Sjoberg, and H.
Korkeala. 2002. Similar Listeria monocytogenes pulsotypes detected in several foods originating from different sources. Int J Food Microbiol 77:83-90. 2. CDC. 2010. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food 10 states - 2009. Morb. Mortal. Wkly Rep 59:418422. 3. Ferreira, V., M. Wiedmann, P. Teixeira, and M. J. Stasiewicz. 2014. Listeria monocytogenes persistence in food-associated environments: epidemiology, strain characteristics, and implications for public health. J Food Prot 77:150-70. 4. Fugett, E. B., D. Schoonmaker-Bopp, N. B. Dumas, J. Corby, and M. Wiedmann. 2007. Pulsed-field gel electrophoresis (PFGE) analysis of temporally matched Listeria monocytogenes isolates from human clinical cases, foods, ruminant farms, and urban and natural environments reveals source-associated as well as widely distributed PFGE types. J Clin Microbiol 45:865-73. 5. Graves, L. M., S. B. Hunter, A. R. Ong, D. Schoonmaker-Bopp, K. Hise, L. Kornstein, W. E. DeWitt, P. S. Hayes, E. Dunne, P. Mead, and B. Swaminathan. 2005. Microbiological aspects of the investigation that traced the 1998 outbreak of listeriosis in the United States to contaminated hot dogs and establishment of molecular subtyping-based surveillance for Listeria monocytogenes in the PulseNet network. J Clin Microbiol 43:2350-5. 6. Gray, M. J., R. N. Zadoks, E. D. Fortes, B. Dogan, S. Cai, Y. Chen, V. N. Scott, D. E. Gombas, K. J. Boor, and M. Wiedmann. 2004. Listeria monocytogenes isolates from
37

foods and humans form distinct but overlapping populations. Appl Environ Microbiol 70:5833-41. 7. Hoelzer, K., B. D. Sauders, M. D. Sanchez, P. T. Olsen, M. M. Pickett, K. J. Mangione, D. H. Rice, J. Corby, S. Stich, E. D. Fortes, S. E. Roof, Y. T. Grohn, M. Wiedmann, and H. F. Oliver. 2011. Prevalence, distribution, and diversity of Listeria monocytogenes in retail environments, focusing on small establishments and establishments with a history of failed inspections. J Food Prot 74:1083-95. 8. Hu, Y., K. Gall, A. Ho, R. Ivanek, Y. T. Grohn, and M. Wiedmann. 2006. Daily variability of Listeria contamination patterns in a cold-smoked salmon processing operation. J Food Prot 69:2123-33. 9. Maitland, J., R. Boyer, D. Gallagher, S. Duncan, N. Bauer, J. Kause, and J. Eifert. 2013. Tracking cross-contamination transfer dynamics at a mock retail deli market using GloGerm. J Food Prot 76:272-82. 10. Miya, S., B. Kimura, M. Sato, H. Takahashi, T. Ishikawa, T. Suda, C. Takakura, T. Fujii, and M. Wiedmann. 2008. Development of a multilocus variable-number of tandem repeat typing method for Listeria monocytogenes serotype 4b strains. Int J Food Microbiol 124:239-49. 11. Nightingale, K., L. Bovell, A. Grajczyk, and M. Wiedmann. 2007. Combined sigB allelic typing and multiplex PCR provide improved discriminatory power and reliability for Listeria monocytogenes molecular serotyping. J Microbiol Methods 68:52-9. 12. Nightingale, K. K., Y. H. Schukken, C. R. Nightingale, E. D. Fortes, A. J. Ho, Z. Her, Y. T. Grohn, P. L. McDonough, and M. Wiedmann. 2004. Ecology and transmission of Listeria monocytogenes infecting ruminants and in the farm environment.
38

Appl Environ Microbiol 70:4458-67. 13. Norton, D. M., M. A. McCamey, K. L. Gall, J. M. Scarlett, K. J. Boor, and M.
Wiedmann. 2001. Molecular studies on the ecology of Listeria monocytogenes in the smoked fish processing industry. Appl Environ Microbiol 67:198-205. 14. Pradhan, A. K., R. Ivanek, Y. T. Grohn, R. Bukowski, I. Geornaras, J. N. Sofos, and M. Wiedmann. 2010. Quantitative risk assessment of listeriosis-associated deaths due to Listeria monocytogenes contamination of deli meats originating from manufacture and retail. J Food Prot 73:620-30. 15. Sauders, B. D., M. Z. Durak, E. Fortes, K. Windham, Y. Schukken, A. J. Lembo, Jr., B. Akey, K. K. Nightingale, and M. Wiedmann. 2006. Molecular characterization of Listeria monocytogenes from natural and urban environments. J Food Prot 69:93-105. 16. Sauders, B. D., K. Mangione, C. Vincent, J. Schermerhorn, C. M. Farchione, N. B. Dumas, D. Bopp, L. Kornstein, E. D. Fortes, K. Windham, and M. Wiedmann. 2004. Distribution of Listeria monocytogenes molecular subtypes among human and food isolates from New York State shows persistence of human disease--associated Listeria monocytogenes strains in retail environments. J Food Prot 67:1417-28. 17. Sauders, B. D., J. Overdevest, E. Fortes, K. Windham, Y. Schukken, A. Lembo, and M. Wiedmann. 2012. Diversity of Listeria species in urban and natural environments. Appl Environ Microbiol 78:4420-33. 18. Sauders, B. D., M. D. Sanchez, D. H. Rice, J. Corby, S. Stich, E. D. Fortes, S. E. Roof, and M. Wiedmann. 2009. Prevalence and molecular diversity of Listeria monocytogenes in retail establishments. J Food Prot 72:2337-49. 19. Sirsat, S. A., K. Kim, K. E. Gibson, P. G. Crandall, S. C. Ricke, and J. A. Neal. 2014.
39

Tracking microbial contamination in retail environments using fluorescent powder - a retail delicatessen environment example. J Vis Exp 20. U.S. Food and Drug Administration, F. S. a. I. S. 2003. Quantitative assessment of the relative risk to public health from foodborne Listeria monocytogenes among selected categories of ready-to-eat foods, Washington, D.C. Available at: http://www.fda.gov/downloads/food/scienceresearch/researchareas/riskassessmentsafetya ssessment/ucm197330.pdf. 21. USDA-FSIS/FDA. 2010. FSIS risk assessment for Listeria monocytogenes in Ready-toEat meat and poultry deli meats. Available at: http://www.fsis.usda.gov/PDF/Comparative_RA_Lm_Exec_Summ_May2010.pdf 22. Williams, S. K., S. Roof, E. A. Boyle, D. Burson, H. Thippareddi, I. Geornaras, J. N. Sofos, M. Wiedmann, and K. Nightingale. 2011. Molecular ecology of Listeria monocytogenes and other Listeria species in small and very small ready-to-eat meat processing plants. J Food Prot 74:63-77. 23. Zhang, W., B. M. Jayarao, and S. J. Knabel. 2004. Multi-virulence-locus sequence typing of Listeria monocytogenes. Appl Environ Microbiol. 70:913-20.
40

CHAPTER 3 EXPERT ELICITATION USED FOR THE IDENTIFICATION AND CLASSIFICATION OF
SAMPLE COLLECTION SITES FOR PATHOGEN ENVIRONMENTAL MONITORING PROGRAMS FOR LISTERIA MONOCYTOGENES
ABSTRACT Pathogen Environmental Monitoring (PEM) programs for Listeria are important in the
food industry to reduce the contamination risk for exposed Ready-To-Eat (RTE) food products with the pathogen L. monocytogenes. Specific guidance to identify appropriate sampling sites in individual facilities, including equipment and other sites, will facilitate effective L. monocytogenes control and PEM programs. Key goals of Listeria PEM programs are to (i) identify and subsequently eliminate niches that allow for Listeria growth and survival over time and (ii) verify and validate preventive controls such as sanitation programs and SSOPs, sanitary equipment design, and sanitary facility design. An initial list of sampling sites covering zones 1– 4 was assembled based on current literature and guidance documents from different organizations with initial classification of sites into (i) Zones 1, 2, 3, and 4; (ii) likely niches or transfer sites, and (iii) verification or indicator sites. A group of 36 experts were asked to (i) recommend additional sampling sites not included in the initial set, (ii) review the classification of sites, and (iii) rank sites on level of importance from 1–5 . The final sample site list includes 76 sampling sites classified by zone, etc. This document thus provides an initial set of sites that can be used by industry to help in the development of Listeria PEM programs.
41

INTRODUCTION Listeria monocytogenes is a facultative anaerobic foodborne pathogen that is commonly
found in natural and food processing environments (1, 8). L. monocytogenes has been estimated to cause 1,455 hospitalizations and 255 deaths annually in the U.S. causing billions of dollars in productivity loss and medical costs every year (10, 11). In addition to foodborne disease cases and outbreaks, food recalls due to L. monocytogenes contamination, even in the absence of associated human cases, also incur significant costs to industry. Sources of L. monocytogenes contamination of food can be either raw material or food associated environments. As L. monocytogenes is effectively inactivated by heat treatment, including typical commercial pasteurization treatment, contamination of finished Ready-to-eat (RTE) products typically occurs through contamination from the environment after heat treatment (or other critical control points). While raw material may play a role as a contamination source for RTE foods that do not undergo heat treatment (e.g., fresh produce, cold smoked seafood, raw milk cheeses), L. monocytogenes contamination from environmental sources in food facilities has shown to play an important role in finished product contamination, even for those products that do not include listericidal process step (16). The importance of identifying L. monocytogenes contamination sources in the processing plant environment is also supported by a number of listeriosis outbreaks linked to RTE foods that were contaminated after heat treatment. For example, in a listeriosis outbreak in Finland (25 cases, 6 deaths), which was linked to contaminated butter, the outbreak strain was isolated from the processing plant environment and butter-producing equipment and post-pasteurization contamination from screw conveyor in the butter wagon was suggested to be the likely contamination source. Similarly, in a multi-province outbreak in Canada linked to RTE deli meats in 2008 (57 cases, 24 deaths) the slicer was identified as a
42

likely source of finished product contamination (2, 21). In a 2011 listeriosis outbreak in 28 U.S. states linked to contaminated cantaloupe (146 cases, 30 deaths), molecular subtyping also identified the environment of the packing house as the likely source of all five outbreak related subtypes (7). These outbreaks, in addition to several recent and on-going outbreaks, underscore the importance of environment contamination sources in processing plants and packing houses as a root cause of human listeriosis outbreaks.
Due to the well-recognized importance of environmental L. monocytogenes contamination sources as detailed above, industry as well as regulatory agencies increasingly emphasize the importance of strategies to control L. monocytogenes in processing plant and retail environments. Key control strategies include sanitary equipment and facility design as well as implementation of Sanitation Standard Operating Procedures (SSOPs) and master sanitation schedules for facilities and equipment. Effective PEM programs help to monitor the effectiveness of Listeria control programs and strategies and to identify whether the facility design and infrastructure supports the safe production and handling of products. In addition to providing overall monitoring, effective PEM programs helps to identify specific problem sites in facilities. Importantly, regulatory agencies across the world are also increasingly requiring PEM program (including specific programs targeting Listeria). For example, in the US, the USDA requires monitoring of food contact surfaces in RTE meat processing plants, while new FSMA rules also require PEM programs in FDA-regulated processing facilities where RTE products are exposed to the environment.
Sites used to collect PEM samples can be classified using different approaches. Commonly, environmental sampling sites are classified into four zones, including (i) Zone 1 (food contact surfaces); (ii) Zone 2 (non-food contact surfaces immediately adjacent to product
43

contact surfaces; (iii) Zone 3 (non-food contact surfaces that are within the processed product area but are not near food or food contact surfaces and, (iv) Zone 4 (non-food contact sites outside of the area where processed product is exposed (United Fresh, 2013; Stone, 2014). In addition to zones, sampling sites can also be classified as growth niches or transfer points. Growth niches, often used interchangeably with the term “harborage sites”, represents areas where various bacteria (e.g., indicator organisms and pathogens) can grow and survive sanitation. Harborage sites are growth niches that are inhabited by a specific pathogen or indicator organism of interest (e.g., a specific subtype of L. monocytogenes) (14). A classical growth niches may be a hollow roller or hollow legs of equipment. Potential growth niches that represent good sampling sites are usually difficult to reach and clean areas where moisture and food can accumulate over a period of time. A transfer point (sometimes also called “transfer site”) is a site with a likely potential of transferring an organism from one location to another (e.g., a gloved hand or a door handle) (14).
In addition to classifying environmental sampling sites into zones, niches and transfer points, sites can further be classified into verification sites and indicator sites. Verification sites are typically Zone 1 food contact surfaces but can also include Zone 2 sites and are sampled to verify the effectiveness of control programs for a targeted pathogen (i.e., L. monocytogenes for the context of this paper) (14). In some cases, facilities collect and test samples from environmental verification sites in addition to finished product samples to validate their food safety system. If a verification site tests positive for L. monocytogenes or Listeria spp., than that suggests a food safety system failure. Indicator sites are typically Zone 3 and Zone 4 sites that often are transfer points in proximity to possible growth niches. Transfer points that test positive for L. monocytogenes provide an early indication of the presence of potential food safety
44

hazards; however they does not indicate a pathogen control system failure (14). A positive indicator site warrants a “not for-cause” investigation of potential pathogen sources or entry points that are not in proximity to food contact surfaces.
Routine environmental monitoring typically focuses on detection of the target pathogen or an index organism using classical or molecular based detection systems. In some cases, subtype characterization of pathogen isolates obtained as part of PEM programs can provide additional relevant information. Commonly used subtyping methods for L. monocytogenes and other Listeria species include ribotyping and Pulsed-Field Gel Electrophoresis (PFGE), as well as, more recently, whole genome sequencing (WGS). Subtyping of pathogen isolates can be used by industry as part of their routine PEM programs (with every pathogen isolate characterized by subtyping) or can be used as a specialized tool when necessary (for example, to identify the specific source responsible for finished product contamination). Molecular subtyping also has been used regularly by researchers to better understand the ecology and transmission of L. monocytogenes and other Listeria species in food processing plants (5,6). Importantly, molecular subtyping studies on L. monocytogenes and other Listeria species have provided convincing evidence that one (or sometimes multiple) specific subtypes can survive (“persist”) in processing facilities over time. For example, one study that used both ribotyping and whole genome sequences found evidence that a specific L. monocytogenes subtype persisted in a food processing facility for at least 12 years (17). A number of other studies have reported evidence for persistence for time periods of 1 to 10 years in different types of processing facilities, including RTE dairy, meat, seafood, and produce processing facilities (3). For example, two studies reported L. monocytogenes persistence in smoked-seafood plants for several months to two years (8,12). More recently, molecular subtyping also provided evidence that L.
45

monocytogenes can persist in retail environments for 9 months and longer (18,19). Importantly, subtyping data have also helped identify actual sites (“niches”) where L. monocytogenes or Listeria spp. can survive over time in food associated environments. For example, one study reported apparent persistence of L. monocytogenes in floor mats for at least 2 years (8), as supported by the fact that the persistent L. monocytogenes was eliminated from the facility when the apparently uncleanable floor mats were removed. Subtype characterization of pathogen isolates obtained from processing plant environments can thus be valuable not only for individual facilities, but these type of data can be used to (i) design effective environmental control and monitoring programs and to (ii) identify appropriate sampling sites to be included in PEM programs (22).
Based on the importance of PEM programs for control of L. monocytogenes a number of industry and government guidance documents (9,20) and review articles (3,14) on this topic have been published. In addition, training programs on development and implementation of Listeria PEM programs are offered by a number of groups. Despite the availability of these resources, selection and classification of sampling sites used in PEM programs still represents a challenge for many segments of the food industry, in particular small and medium sized enterprises. Few resources, if any, detail appropriate sampling sites across industries. To address this need and gap, we compiled a list of potential sampling sites appropriate for Listeria PEM programs, followed by an expert elicitation to (i) suggest additional sampling sites not included in the initial set and (ii) review classification of sites. While the scope of this effort focused on industries where L. monocytogenes is a particular concern (i.e., RTE seafood, dairy, meat, produce and retail), the information provided here should also be useful for other industries.
46

MATERIALS AND METHODS Initial identification of sample site list. An initial list of sampling sites (Appendix A1 was developed based on (i) existing industry and government guidance documents and (ii) peerreviewed manuscripts identified using searches with appropriate key words (e.g., sample site, smoked seafood plant, dairy plant equipment, meat, RTE deli meat, Listeria persistence); the specific references that supported a given sampling site are listed in Supp. Material 1. The initial list assembled included 77 sites, including food contact (FCS) and non-food contact (NFC) sites that exist across food facilities and sites that are exclusive to specific industries. The initial listing of sites included (i) a general description of the site; (ii) a description of specific areas that represent good sampling sites (e.g., underside of mats, cracks that harbor moisture, condensation on pipes, etc.); (iii) industry where a site was applicable (i.e., meat, seafood, dairy, produce, retail); and (iv) references supporting selection of a specific site. All sites were initially classified into (i) Zones (1, 2, 3, or 4; sites could be classified to possibly represent more than 1 zone), (ii) transfer point or niches, and (iii) indicator or verification sites. The initial sample site list compiled by the authors was subsequently refined by two industry experts that (i) provided clarification on sample site description and classification and (ii) recommended additional sampling sites (6 additional sites). Development of assessment tool. The sample site list described above was used to develop an expert assessment tool that would allow expert reviewers to (i) rank sites in terms of importance on a scale of 1 – 5 with 5 being the most important, (ii) classify sites into zones, (iii) classify sites as either transfer point or niche, and (iv) classify sites as either indicator or verification site. Reviewers were allowed to assign multiple zones to a site. Similarly, reviewers were asked to classify sites as either a niche or transfer point, but if deemed appropriate could designate a given
47

site as niche or transfer point (for example, crates that hold finished products could be a niche or a transfer point). For classification into verification site (VS) or indicator site (IS), one category (VS or IS) had to be selected for each site. The assessment tool also allowed expert reviewers to provide additional sample sites to be added to the initial list provided to them. Selection of expert reviewers and assessment tool administration. Expert reviewers were selected to have background and experience with food safety and environmental monitoring programs in the seafood, dairy, meat, produce and retail industries. The assessment tool described above was emailed to 36 expert reviewers representing US food industry (n=14), industry support groups (n=6), US academic institutions (n=4), US federal and state government (n=7), as well non-US experts (n=5). In addition to the assessment tool, a questionnaire requesting relevant professional information (e.g., years of experience in food safety, years of experience with pathogen environmental monitoring programs) was sent to expert reviewers. At least 3 reminders to respond were sent to each expert. If they were unavailable, expert reviewers were given the opportunity to select an appropriately qualified designee to complete the assessment or to recommend another qualified expert to participate. Analysis of completed assessment tools for site classification. Expert reviewers were asked to classify sites for importance on a scale of 1 to 5 with 1 being least important and 5 being most important. If reviewers used intermediate values, the values were averaged before the analyses; for example if a reviewer assigned an importance value of “1 or 2”, the response was recorded as 1.5. Reviewers classified sites into either single (e.g., Zone 1) or multiple zones (e.g., Zones 1 or 2; Zones 1 or 2 or 3). If reviewers reported multiple zones for a given site, equal proportions of 1 were assigned to each site in the analyses; for example if a reviewer’s response was “Zones 1 or 2 or 3”, a response was recorded as a value of 0.33 for each of the 3 zones during the analyses.
48

Reponses for classification of sites into niche (Ni) and transfer points (TP) were recorded as “Ni”, “TS”. Reponses for classification of sites into either indicator site (IS) or verification site (VS), were as recorded as “IS”, “VS” or “IS or VS”. If no response was given it was reported as “NR” (no response). Similar to zone analysis, equal proportions of 1 was assigned if a reviewer indicated more than one classification; for example for each response of “Ni or TS,” 0.5 was recorded. Compilation of new sites reported in the completed assessment tools. Reviewers were asked to provide suggestions for additional sites to add to the initial list. If reviewers provided suggestions for additional sites, site suggestions were evaluated and used to either (i) refine an existing site description or (ii) create an additional new site description. For example, the initial list included a description for “floor” that read “under equipment, areas that have reels and posts bolted down.” The description was modified to include “under equipment, areas that have expansion joints, reels and posts bolted down“. “Gloves” for example, was not included in the initial list of sites, but after reviewers’ suggestions, was added (Table 3.2).
RESULTS Initial sample site list. The initial sample site list used to create the assessment tool included 77 sites; this list included a description of sites as well as, where appropriate, a more detailed description of the specific areas of the sites, equipment, or infrastructure that should be sampled. The initial list included 34 sites in Zone 1, 22 sites in Zone 2, 19 in Zone 3, and 2 sites in Zone 4; initial classification into zones was based on guidance documents and the expertise of the authors and two initial industry reviewers. Among the initial 77 sites, 10 and 12 were classified as niches or transfer points, respectively; 55 sites were classified as either a niche or a transfer
49

point. Furthermore, the initial 77 sites were classified as 38 verification sites or 39 indicator sites verification or indicator sites. Response rate and expert reviewers’ profiles. Overall, 16 of the 38 individuals returned the assessment tool (42% response rate). Reponses rates by categories were 7/14 (50%) for US food industry, 2/6 (33%) for US academic institutions, 1/6 (17%) for US federal and state government, as well 2/5 (40%) for non-US experts. Three reviewers declined participation because of work responsibilities, family obligations, and health complications. The remaining 19 reviewers did not respond or could not complete the assessment tool within the allotted time frame. All 16 reviewers had a minimum of 10 years of experience working in food safety, including 10 reviewers with more than 20 years of food safety experience in academia, outreach and extension, consulting, regulatory agencies, seminars, and continuing education classes (Figure 3.1). The majority of reviewers (n= 13) gained at least some of their experience with PEMs while working in food industry (Table 3.1), the other three reviewers gained their expertise while working in extension, academia, and consulting (n=1), regulatory (n=1), and regulatory and academia (n=1) (Table 3.1).
50

Table 3.1 Summary of expert reviewers' profiles

No experience indicated

Experience 1 – 5

<1 year

years

5 – 10 years

10 – 20 >20

Total

years

yearsa Responses

Years of experience in food safety

0 0 0 3 3 10 16

Years of experience working with PEM programs

0

0 3 4 6 3 16

Years of experience in each sector

Meat

5 1 1 1 2 6 16

Seafood

9 2 2 2 0 1 16

Dairy

5 1 4 2 1 3 16

Produce

5 1 1 5 2 2 16

Retail

6 0 5 1 3 1 16

a one expert reviewer indicated that he has broad experience working in each industry. Based on peer-reviewed publications, it is determined that this individual has greater than 20 years of experience in each industry.

51

Expert reviewer identification of new sites. Overall, expert reviewers initially provided 31 suggestions for new sites; the descriptions of these suggested new sites were used to (i) refine the description of 6 existing sites and to (ii) define 6 new sites. Briefly, a number of the suggested new sites represented sampling areas that could be part of different equipment or sites that had been included in the initial site list (for example, asset tags, tape, electrical cords); these suggestions were used to refine the “Description and Comment” column of Table 3.3. For example, for Site 1.8 this column was changed from “special attention given to hard to clean areas such as hollow rollers, cloth backed belting, and metal joints” to “special attention given to hard to clean areas such as hollow rollers, cloth backed belting, metal joints, Asset Tags, stickers, etc.” The 6 new sites identified based on the reviewer comments included (i) chemical drums, (ii) gloves, (iii) sole scrubber, (iv) hand air dryer, (v) display cases, and (vi) light fixture guards in production area (Table 3.2).
52

Table 3.2 New sites recommended by reviewers

Site Name Chemical Drums

Description and comments

Type Zone of surface

Special attention given to drums or jugs that are used across zones. NFCS 3 or 4 Focus on handles and the underside of drums and jugs

Verification site (VS) or Indicator site (IS)
IS

Niche (N) Transfer point (TP)
TP

Type of plants where most applicable
Meat, seafood

Gloves Sole Scrubber

Special attention given to heat resistant gloves employees wear to remove products from ovens

NFCS NFCS

2 4

IS IS

N Meat, seafood N or TP Meat, seafood

Hand air dryers

Special attention given to handles

Display cases

Special attention given to hard to clean areas and handles

NFCS 4

Is

NFCS 2

IS

TP Meat, seafood, retail, dairy, produce
N Retail

Light fixture guards in production area

Special attention given to areas with moisture

NFCS 3

IS

N Meat, seafood, dairy, retail, produce

53

Zone classification analysis. Expert reviewer classifications of sites into zones were analyzed as detailed in the “Materials and Methods”; the zone that was assigned to a given site by the highest proportion of responses is provided in Table 3.3 additional zones that were assigned by reviewers to a given site are listed in parenthesis as long as 2 or more reviewers assigned a given site to that zone. Based on expert reviewers’ responses, the zones assigned to eight sites were reclassified (Table 2). All 34 sites initially identified as Zone 1 remained classified as Zone 1; 18 of the Zone 1 sites were classified as Zone 1 by all reviewers that provided a zone classification for a given site (Figure 3.1). The remaining 16 Zone 1 sites were classified as Zone 1 by the majority of reviewers, while some reviewers also classified these sites into other zones, typically Zone 2 (Figure 3.1). For example, one of the Zone 1 sites with most reviewer classifications into zones other than Zone 1 was site 1.32 (“vacuum sealer”); 11, 2, and 1 reviewers classified this site as Zone 1 only, Zone 2 only, and Zone 3 only, respectively. The remaining reviewers classified “vacuum sealers” as possibly representing multiple zones; one reviewer did not provide zone classification for this site. In the final sample site list 34 sites were classified as Zone 1, 14 sites were classified as Zone 2, 26 sites as Zone 3, and 3 sites as Zone 4. Ten sites were classified as “Zone 2 or 3”. (Table 3.3).
54

Table 3.3 Classification of sites based on expert reviewers' recommendations

ID Site Name (Importance Description and comments Rank)

Type of plants where most applicable

References

Zonea

1.1 Bowlcutters (2)

Special attention given to interior of castings bowl, blade spanner, and other difficult to clean food contact areas

Meat, seafood, produce

Rodriguez-Marval et al. 1 2010

1.2 Cutting boards used for Special attention given to damaged and/or Meat, seafood, dairy,

finished product (2)

difficult to clean areas

retail

Cunningham et al. 2011; Vongkamjan et al. 2013

1

1.3 Grinders (2)

Special attention given to blades and other Meat, seafood difficult to clean food contact areas

1.4 Other Equipment (e.g., vats, tanks, tables) that comes in contact with product after CCP (2)

Meat, seafood, dairy, produce

United Fresh Produce 1 Association 2013; FSIS 2012
Kushwaha and Muriana 1 2009; GMA 2014

1.5 Brushes and other

Examples include brushes used to brush Dairy

equipment that touch

cheeses with brine (during aging)

finished product (3)

1.6 Slicer/peelers/choppers Special attention given to food contact (4) areas (e.g., blades, conveyor belt) and areas that show sanitary design flaws.

Meat, seafood, dairy, retail, produce

1.7 Bumper guards (e.g., on sorting tables, along conveyances) that contact finished product (4)

Produce

Expert Reviewer

1

Tompkin et al. 1999; Scott et al. 2005; Kornacki 2012

1

United Fresh Produce Association 2013

1(2)

Verification
site (VS) or
Indicator site (IS)b

Niche (Ni)
or
Transfer point (TP)c

Rank

VS Ni 4.00

VS

TP(Ni)

4.00

VS Ni 4.00 VS TP 4.00

VS

Ni (TP)

3.94

VS Ni 3.93

VS(IS)

Ni or TP

3.93

55

1.8 Conveyor systems, belt Run system for at least 15 min. prior to Meat, seafood, produce Scott et al. 2005;

1

and other food contact sampling; special attention given to hard

Sauders et al. 2009;

sites (4)

to clean areas such as hollow rollers, cloth

Malley et al. 2015

backed belting, metal joints, and asset tags

and stickers

1.9 Mincers (4)

Special attention given to trays, splash

Retail

pans, and difficult to clean areas that come

in contact with product

Ryser and Marth 2007 1 Barmpalia-Davis 2008

1.10 Packaging machine (4) Special attention given to sites that come Meat, seafood, dairy,

in direct contact with exposed product

produce,

1.11 Interior of pipes and tubes that transfer finished product (5)

Interior food contact surfaces with special attention given to joints, gaskets, dead ends, areas with milkstone buildup or obvious biofilm formation, and difficult to clean areas

Dairy

1.12 Holding vats used to store finished product (6)

Seafood

Scott et al. 2005 Kabuki et al. 2004
Rørvik et al. 1997;

1(2) 1
1

1.13 Buckets/bins that come Special attention given to inside of bucket Seafood, produce in contact with finished used to soak/rinse product product (7)

Vongkamjan et al. 2015

1

1.14 Fillers (7)
1.15 Scraper blade (8) 1.16 Skinning machine (8)

Special attention given to nozzle, lid seals, other rubber parts, and difficult to clean food contact areas

Meat, dairy

Seafood, meat

Special attention given to blade attached to machine and other sites that come in direct contact with product areas where trapped product may reside

Seafood, meat

Dalton et al. 1997

1

FSIS 2012 Scott et al. 2005

1 1

VS

Ni (TP)

3.93

VS(IS)

Ni

3.93

VS

TP(Ni)

3.93

VS(IS)

Ni

3.88

VS(IS)

Ni(TP)

3.87

VS(IS)

TP(Ni)

3.81

VS

Ni(TP)

3.81

VS TP 3.80 VS Ni 3.80

56

1.17 Sorting table (surfaces appropriate sample site if smoke sticks are Produce

that contact finished

not heat treated after each use and are

product) (8)

difficult to clean

1.18 Flume Wash (9) 1.19 Mixers (10)

Special attention given to difficult to reach Produce areas (e.g., cover, filter, pump motor, shakers, and controls)

Special attention given to bowls, rotors and other difficult to clean food contact areas

Meat, seafood, dairy

1.20 Re-wrap counter/table used for finished product (10)

Meat, seafood, dairy, retail

1.21 Chutes (11)

Special attention given to conveyors, crevices and hard to clean areas

Seafood, meat, dairy, produce

United Fresh Produce Association 2013

1

United Fresh Produce Association 2013

1

United Fresh Produce Association 2013

1

Cunningham et al. 2011; Simmons et al. 2014

1

FSIS 2012

1

1.22 Ice maker used for ice that comes in contact with exposed finished product (11)
1.23 Hopper Surface (12)

Sample interior area of machine

Meat, seafood

Special attention given to rotating sweep, paddle agitators and other difficult to clean food contact areas

Meat, seafood, dairy, produce

1.24 Scale (surface that comes in contact with finished product) (13)

Meat, seafood, retail, produce

1.25 Valves (13)
1.26 Brine chiller chamber/tunnel (14)

Special attention given to difficult to clean areas and areas where trapped product may reside
Special attention given to sponge/rubber seals around edge and other difficult to clean areas. Sampling could also include brine itself.

Dairy Meat, seafood

Tompkin et al. 1999

1

FSIS 2012

1

Kushwaha and Muriana 1 2009; Simmons et al. 2014

Kornacki 2012

1

GMA 2014

1(2)

VS TP 3.80

VS

Ni(TP)

3.79

VS(IS)

Ni

3.75

VS TP 3.75

VS(IS)

TP (Ni)

3.73

VS(IS)

TP(Ni)

3.73

VS

TP(Ni)

3.71

VS TP 3.69

VS(IS) VS

Ni(TP)

3.69

Ni 3.67

57

1.27 Band Saws (14)

Special attention given to carriage track, wheel door, door panels, guide blocks/bars portion, and other difficult to clean food contact areas

Meat, seafood

FSIS 2012

1 VS Ni 3.64

1.28 Cases/crates that hold finished products (16)

Special attention given to cases/crates that Dairy, retail, Produce are damaged and/or have sanitary design problems

1.29 Food preparation sink

Good sampling areas include difficult to Seafood, retail

used for finished product clean areas such as unsanitary and poor

(17) welds

1.30 Baggers (19)

Meat, seafood, dairy, produce

1.31 Smoke sticks (19)

Appropriate sample site if smoke sticks are not heat treated after each use and are difficult to clean

Meat

Expert Reviewer

1(2)

Sauders et al. 2009

1

United Fresh Produce 1(2,3) Association 2013; FSIS 2012

FSIS 2012

1

VS(IS) VS VS(IS) VS

TP(Ni)

3.60

TP 3.57

TP(Ni)

3.54

Ni(TP)

3.54

1.32 Vacuum sealers (27)

Special attention given to chamber unit and nozzle sealer

Meat, seafood,

1.33 Thermometers, Thermocouples, etc. that contact finished products (31)

Meat, seafood, dairy, retail, produce

1.34 Aprons that contact finished product (44)

Focus on re-usable aprons; special attention given to front, strings, worn and heavily soiled aprons, and aprons that hang low and may come in contact with the outer guides of the conveyor

Meat, Seafood, dairy, retail, produce

FSIS 2012 FSIS 2012 FDA 2008

1(2) VS 1 VS

TP(Ni)

3.33

TP 3.25

1(2)

VS(IS)

TP

2.75

58

2.1 Conveyor systems (8) Sample areas that do not come in contact Meat, seafood with product (e.g., housing, rollers, sprockets, idlers, greased bearings). If samples are collected pre-op, run conveyor for 15 min before sampling. Spectial attention given to hollow legs and other hollow structures and areas with poor welding

2.2 Framework and non-

Sample framework and other non-FCS

Meat, seafood, dairy,

food contact areas of

surfaces of equipment (e.g., slicers,

produce

food contact equipment peelers, packaging equipment, baggers,

(10) vats, tanks, tables, electrical cords, etc.);

special attention given to hollow areas

(e.g., hollow legs),undersides, areas with

poor welds, rust, and areas with moisture.

Samples could also be fluid that exudes

from hollow areas

Tompkin et al. 1999

2(1)

FDA 2008; Scott et al. 2005; Kushwaha and Muriana 2009; GMA 2014; Malley et al. 2015

2(3)

2.3 Racks that are used for Special attention given to hollow parts, Meat, seafood, dairy, exposed finished product wheel covers, and other difficult to clean retail (10) areas. Parts that contact exposed finished products would be considered Zone 1.

2.4 Maintenance tools (15) Sample maintenance tools used in RTE Meat, seafood, dairy,

areas

retail

Expert Reviewer

2(1)

Tompkin et al. 1999

2(3)

2.5 Spiral freezers (20)

Special attention given to interior walls and difficult to reach areas

Meat, Seafood

2.6 Control switch/HMI

Special attention given to switches, screen Meat, seafood

screens on equipment

surfaces, and other difficult to clean areas

close to FCS (22)

FDA 2008 Expert Reviewer

2(1,3) 2(3)

2.7 Air blower (24)

Sample side channel, air inlet, impeller chamber

Seafood, dairy, produce FSIS 2012

2(1,3)

IS(VS)

Ni(TP)

3.80

IS(VS)

Ni(TP)

3.75

VS(IS)

Ni or TP

3.75

IS(VS)

TP

3.63

VS or IS

Ni or TP

3.53

IS(VS)

TP

3.47

VS(IS)

Ni(TP)

3.40

59

2.8

Pipelines and exposed

Sample exterior surface with special

Meat, seafood, dairy

overhead piping in area attention given to areas with condensation,

with exposed product

moisture and drips and difficult to clean

(25) areas (e.g., clamps, cracked insulation). If

pipes are in areas with exposed products,

drips have the potential to come in contact

with exposed product, piplines would be

considered Zone 1. Pipelines outside of

the finished product areas would be Zone

3.

2.9 Railings in finished product areas (29)

Special attention given to joints

Meat, seafood, dairy, produce

2.10 Scale (32) 2.11 Drip pans (33)

Sample areas that do not come in contact with product (i.e., keypads, under scale pan)

Meat, seafood, retail

Special attention given to pans that are not Meat, seafood, dairy,

clean and used often.

retail

Kabuki et al. 2004; Tompkin et al. 1999

2(3)

Expert Reviewer

2(3)

Kushwaha and Muriana 2(3) 2009; Simmons et al. 2014

FDA 2008

2(3)

2.12 Flaps or strip curtains used in areas where exposed finished product is present (36)

2.13 Boots/footwear worn by Special attention given to welts of

personnel who handle

footwear

exposed product (37)

Meat, seafood, produce FSIS 2012

Meat, seafood, dairy, produce

FSIS 2012

2(3) 2(3)

2.14 Cooler doors and other Special attention given to difficult to clean Meat, seafood, retail doors used for transport and sanitize areas such as rubber door of open finished product seals, cracks behind kick plates, worn and (40) damaged areas, and areas with sanitary design issues

Malley et al. 2013; FDA 2008

2(3)

VS or IS

Ni(TP)

3.38

IS(VS)

TP(Ni)

3.30

IS(VS)

TP(Ni)

3.22

IS(VS)

Ni(TP)

3.13

IS(VS)

TP(Ni)

3.07

IS(VS)

TP

3.06

IS(VS)

Ni(TP)

2.93

60

3.1 Drains (1)

Special attention given to covers, baskets, cracked areas, and deeper areas of the drain. In situations where drains are managed and flow to the drain is minimal (controlled conditions), the top of the drain should represent what is coming to the drain, while sampling down in the drain then represents the conditions inside the drain system. This distinction is not posisble in conditions where drainage issues allow for standing water in the drain area.

Meat, seafood, dairy, retail, produce

3.2 Floor mats (18)

Sample underside with special attention given to any existing crack or crevices. Pressure may have to be applied to mats to sample moisture in the interior of the mat

Meat, seafood, dairy

3.3* Floor (21) 3.4 Carts (21)

Special attention given to standing water, cracks, and difficult to clean areas (e.g. under equipment, areas that have expansion joints, reels and posts bolted down). Floors near FCS would be classified as Zone 2 ( e.g., shelves or conveyors with exposed product that are close to floor); most floor sampling sites in other areas will be Zone 3 or 4

Meat, seafood, dairy, retail, produce

Special attention given to racks, sides, wheels, hollow areas, and other difficult to clean areas. Sample the same cart every sampling visit and ensure carts are labeled accordingly to allow for easy identification

Meat, seafood, dairy, retail

3.5* Floor/wall junction (23) Floor-walls junctions near FCS could be Meat, seafood, dairy,

classified as Zone 2 (e.g., shelves or

retail, produce

conveyors with exposed product that are

close to floor/wall junction); most

floor/wall junctions in other areas will be

Zone 3 or 4. Samples may also include

moisture that exudes from floor/wall

junction

Kabuki et.al 2004; Ho et al. 2007; Scott et al. 2005; Kornacki 2012; Vongkamjan et al. 2013; Hoffman et al. 2005

3

Scott et al. 2005; Malley et al. 2015

3(2)

Sauders et al. 2009; Simmons et al. 2014

3(2)

Malley et al. 2013

3

Simmons et al. 2014

3

IS(VS)

Ni(TP)

4.06

IS(VS)

Ni

3.56

IS(VS)

TP

3.50

IS(VS)

TP(Ni)

3.50

IS(VS)

Ni(TP)

3.44

61

3.6* In-floor weighing equipment and floor scales (24)
3.7 Hoist chain bags (26)

Equipement and scales located close to finished products. Special attention given to platform and around crevices. If located in areas away from exposed finished product, these sites may be a different zone ( Zone 3).

Meat, seafood, produce

Sample area includes chain sprocket, load Produce hooks, etc.

FDA 2008

3(2)

United Fresh Produce Association 2013

3

IS(VS)

Ni(TP)

3.40

IS(VS)

Ni(TP)

3.35

3.8 Footbath (28)

Sample area includes underside of footbath, collect with special attention given to any existing cracks or crevices

Meat, seafood, dairy

3.9 Cleaning equipment (30) Special attention given to equipment that Meat, Seafood, dairy, is not easily disassembled and equipment retail, produce that is used to clean floors (e.g., squeegees, sponges, mops)

3.10 Air return cover (31)

Sample covers and deeper inside vents with special attention given to poorly maintained vents and filters. Focus on areas with moisture and potential for condensation

Meat, seafood, dairy, retail, produce

3.11* Trash cans (31)

Special attention given to the underside of trash cans, cracked areas, and other difficult to clean areas. Trash cans nearest exposed finished product may be Zone 2 or 3

Meat, seafood, dairy, retail, produce

3.12 Cooler doors and other Special attention given to difficult to clean Meat, seafood, retail

doors not used for

and sanitize areas such as rubber door

transport of open

seals, cracks behind kick plates

finished product (33)

Sauders et al. 2009

3(4)

Tompkin et al. 1999; Kornacki 2002; Simmons et al. 2014

3

FDA 2008

3

Kushwaha and Muriana 3 2009; Malley et. al 2013

Malley et al. 2013; FDA 2008; Tompkin et al 1999

3

IS(VS)

Ni or TP

3.31

IS(VS)

Ni(TP)

3.28

IS(VS)

Ni(TP)

3.25

IS(VS)

Ni(TP)

3.25

IS(VS)

TP(Ni)

3.13

3.13* Handwash sink (34)

Sample interior and exterior, with special Meat, seafood, dairy, attention given to difficult to clean areas retail, produce

Simmons et al. 2014

3(2)

IS(VS)

Ni(TP)

3.13

62

3.14*

Trolleys, lifters, forklifts and pallet jacks (34)

Special attention given to racks, handles, wheels, axle area, forks, and underside of forks, and difficult to clean areas

Meat, seafood, dairy, retail, produce

Tompkin et al. 1999; Kornacki 2012

3(2)

3.15* 3.16 3.17*

Walls (34)
Hoses (and hose holders) (35)

Special attention given to cracks, moist areas, and exposed wet insulations; walls near FCS maybe classified as Zone 2 (e.g., wall penetration with conveyor that carries exposed product). Most wall sampling sites will be Zone 3 or 4
Sample area includes cracked hoses and nozzles and other difficult to clean areas. Special attention given to buildup and visible biofilm

Meat, seafood, dairy, retail, produce
Meat, seafood, dairy, retail

Foot pedals (36)

Special attention given to foot pedals on food contact equipment (e.g. bagger); foot pedals on non-FCS equipment may represent Zone 3

Seafood, meat, produce

3.18 Ceiling in production area (38)

Special attention given to areas with condensation, cracks, and other difficult to clean areas. If drips have the potential to come in contact with exposed product, ceiling should be considered Zone 1.

Meat, seafood, dairy, retail, produce

Tompkin et al. 1999; Kabuki et.al 2004; Kushwaha and Muriana 2009

3

Tompkin et al.1999;

3(2)

Kushwaha and Muriana

2009

FSIS 2012; United Fresh Produce Association 2013

3(2)

Expert Reviewer

3

3.19 Fans (39)

Special attention given to blades, covers, motors, and difficult to clean areas

Meat, seafood, dairy, produce

3.20 Portable steps/stool/ladder (39)
3.21 Exterior of ice maker (40)

Special attention given to rungs contacted by both hands and shoes, cracks, tight gaps between metal pieces, under steps, end caps, and other difficult to clean areas

Meat, seafood, dairy, retail, produce

Sample exterior area of machine surrounding the lid

Meat, seafood

FSIS 2012 Expert Reviewer

3 3

Tompkin et al. 1999

3(4,2)

3.22 Racks that are not used Sample areas include hallow parts, wheel Meat, seafood, dairy,

for exposed finished

covers, and other difficult to clean areas. retail

product (42)

Kushwaha and Muriana 3(4) 2009

IS(VS) IS(VS)
IS(VS) IS(VS) IS(VS)
IS(VS) IS(VS) IS(VS) IS(VS)

TP(Ni)

3.13

Ni(TP)

3.13

Ni(TP)

3.09

TP(Ni)

3.07

Ni(TP)

3.00

Ni(TP)

2.94

Ni(TP)

2.94

Ni(TP)

2.93

Ni(TP)

2.88

63

3.23 Boots/footwear worn by Special attention given to welts of

personnel who do not

footwear

handle exposed product

(44)

Meat, seafood, diary, produce

3.24 Conduits, electrical

Special attention given to moist and

Meat, seafood, dairy,

boxes, conduit-in-casing cracked areas and inside electrical boxes produce

(44)

3.25 Electrical outlet covers Sample areas around and inside electrical Meat, seafood, dairy, (46) outlet covers in production areas; could be retail, produce Zone 4 (if located in a Zone 4 area)

FSIS 2012 Expert Reviewer

3(4) 3 3 or 4

3.26 Soap dispensers (47)

Sample areas include knobs on dispensers Meat, seafood, dairy,

and difficult to clean areas

retail, produce

Rivera-Betancourt et al. 3(4) 2004

4.1 Break room and locker Special attention given to standing water Meat, seafood, dairy,

room area floor (41)

and cracked tiles. Break and locker rooms retail, produce

with direct connection to production area

with exposed product would be considered

Zone 3

Expert Reviewer

4(3)

4.2 Loading Dock (43)

Special attention given to dock bumpers, dockboards, strip doors

Meat, seafood, dairy, retail, produce

United Fresh Produce Association 2013

4(3)

4.3 Windows (45)

Special attention given to areas with cracked caulk, condensation due to temperature extremes, and difficult to clean areas

Meat, seafood, dairy, retail, produce

Kushwaha and Muriana 4(3) 2009

* Sites were re-classified into different zones to reflect reviewers' classification.
a1(2) indicates that the majority of reviewers classified a site as a zone (e.,g., Zone 1) but >1 reviewer selected another zone (e.g., Zone 2) bVS(IS) or IS(VS) indicates the majority of reviewers selected VS (or IS) for a site but >1 reviewer classified the site as an IS (or VS). cVS(IS) or IS(VS) indicates the majority of reviewers selected VS (or IS) for a site but >1 reviewer classified the site as an IS (or VS).

IS(VS) IS(VS) IS(VS) IS(VS) IS(VS)
IS(VS) IS(VS)

TP(Ni)

2.75

Ni(TP)

2.75

Ni(TP)

2.53

TP(Ni)

2.44

Ni(TP)

2.91

TP(Ni)

2.84

Ni(TP)

2.594

64

Zone 1 Zone 2 Zone 3 Zone 4

Sites
65

Brushes and other equipment that touch finished product Chutes
Cutting boards used for finished product Fillers
Holding vats used to store finished product Interior of pipes and tubes that transfer finished product
Mixers Scale (surface that comes in contact with finished product)
Band Saws Conveyor systems, belt and other food contact sites
Food preparation sink used for finished product Ice maker used for ice that comes in contact with exposed finished…
Mincers Skinning machine Slicer/peelers/choppers
Bowlcutters Flume Wash Hopper Surface Other Equipment (e.g., vats, tanks, tables) that comes in contact with… Buckets/bins that come in contact with finished product
Grinders Thermometers. Thermocouples, etc. that contact finished products
Re-wrap counter/table used for finished product Sorting table (surfaces that contact finished product)
Valves Bumper guards (e.g., on sorting tables, along conveyances) that…
Scraper blade Cases/crates that hold finished products
Brine chiller chamber/tunnel Baggers
Aprons that contact finished product Smoke sticks
Packaging machine Vacuum sealers
Racks that are used for exposed finished product Spiral freezers Air blower
Conveyor systems Scale
Ceiling in production area Pipelines and exposed overhead piping in area with exposed product
In-floor weighing equipment and floor scales Cleaning equipment Drains Walls
Railings in finished product areas Control switch/HMI screens on equipment close to FCS
Maintenance tools Framework and non food contact areas of food contact equipment
Boots/footwear worn by personnel who handle exposed product Drip pans
Cooler doors and other doors used for transport of open finished product Flaps or strip curtains used in areas where exposed finished product is… Handwash sink Foot pedals Floor Trolleys, lifters, forklifts and pallet jacks Floor mats Hoses (and hose holders) Fans Ice maker for ice not used on finished product Trash cans Floor/wall junction Carts Cooler doors and other doors not used for transport of open finished… Footbath Boots/footwear worn by personnel who do not handle exposed product Portable steps/stool/ladder Hoist chain bags Conduits, electrical boxes, conduit-in-casing Air return cover Soap dispensers Racks that are not used for exposed finished product Electrical outlet covers Windows Break room and locker room area floor Loading Dock

18 16 14 12 10 8 6 4 2 0

No. of responses

Figure 3.1 Sites are classified into Zones, 1,2,3, and 4 and ordered from sites that come into contact with finished sites to those that are least likely to be exposed to finished product. Some sites are classified into multiple zones.
Overall, zone classification of 8 sites was revised based on reviewer feedback. For example, floor, initially classified as a Zone 2 site by the authors, was identified by the majority of reviewers as Zone 3. Similarly, seven other sites that were initially classified as Zone 2 (sites 3.5, 3.6, 3.11, 3.13, 3.14, 3.15 and 3.17 in the final sample site list; see Table 3.3) were reclassified to Zone 3. Also, the majority of responses classified “pipelines and exposed overhead piping in area with exposed overhead piping in area with exposed product” as Zone 2 site (9/16 responses), while 6/16 responses classified this site as Zone 3; we thus removed our initial statement that read “If pipes are in areas with exposed products, drips have the potential to come in contact with exposed product, pipelines would be considered Zone 1” from the description for this site. Based on these responses, it appears reasonable to classify this site as Zone 2 if the specific location is in close proximity to exposed product, but classify the site as Zone 3 if the specific location is further away from exposed product. Expert ranking of sites by importance. As described in the methods section, expert reviewers ranked the importance of individual sites on a scale from 1–5 with 5 being the most important site that should be included in a PEM. Based on the average of reviewer ranking, “drains” (a Zone 3 site) represented the most important site with an average importance score of 4.06. After drains, “bowlcutters”, “cutting boards”, and “grinders” were classified as three next most important sites with average importance scores of 4.00 for each of these three sites. The sites
66

ranked as most important in Zones 1, 2, 3, and 4, respectively, were “bowlcutters” (score = 4.00), “conveyor systems” (score =3.80); drains (score=4.06) and “locker room floor” (score = 2.91). On the other hand site 3.26 (“soap dispensers”) was classified as the least important site with an average importance score of 2.44 (Figure 3.2). Overall, the 25% sites ranked as most important all represent in Zone 1 sites, except for drains. Overall, 83% (n=64) of sites (Figure 4) had an average importance scores of 3.0 and above, including 33/34 Zone 1 sites, 13/14 Zone 2 sites; and 18/26 Zone 3 sites.
67

4.5 4
3.5 3
2.5 2
1.5 1
0.5 0
11111111111111111111111111111111112222222222222233333333333333333333333333444
Sites
68

Bowlcutters Cutting boards used for finished product
Grinders Other Equipment (e.g., vats, tanks, tables) that comes in contact with product…
Brushes and other equipment that touch finished product Bumper guards (e.g., on sorting tables, along conveyances) that contact finished…
Conveyor systems, belt and other food contact sites Mincers
Packaging machine Slicer/peelers/choppers Interior of pipes and tubes that transfer finished product Holding vats used to store finished product Buckets/bins that come in contact with finished product
Fillers Scraper blade Skinning machine Sorting table (surfaces that contact finished product)
Flume Wash Mixers
Re-wrap counter/table used for finished product Chutes
Ice maker used for ice that comes in contact with exposed finished product Hopper Surface
Scale (surface that comes in contact with finished product) Valves
Brine chiller chamber/tunnel Band Saws
Cases/crates that hold finished products Food preparation sink used for finished product
Baggers Smoke sticks Vacuum sealers Thermometers. Thermocouples, etc. that contact finished products Aprons that contact finished product Conveyor systems Framework and non food contact areas of food contact equipment Racks that are used for exposed finished product Maintenance tools Spiral freezers Control switch/HMI screens on equipment close to FCS
Air blower Pipelines and exposed overhead piping in area with exposed product
Railings in finished product areas Scale
Drip pans Flaps or strip curtains used in areas where exposed finished product is present
Boots/footwear worn by personnel who handle exposed product Cooler doors and other doors used for transport of open finished product
Drains Floor mats
Carts Floor Floor/wall junction In-floor weighing equipment and floor scales Hoist chain bags Footbath Cleaning equipment Air return cover Trash cans Cooler doors and other doors not used for transport of open finished product Handwash sink Trolleys, lifters, forklifts and pallet jacks Walls Hoses (and hose holders) Foot pedals Ceiling in production area Fans Portable steps/stool/ladder Exterior of ice maker Racks that are not used for exposed finished product Boots/footwear worn by personnel who do not handle exposed product Conduits, electrical boxes, conduit-in-casing Electrical outlet covers Soap dispensers Break room and locker room area floor Loading Dock Windows

Classification

Figure 3.2 Rankings of individual sites based on expert reviewers’ opinion of the level of importance from 1–5 with 1 being the least important and 5 being the most important. An average of the expert reviewers’ classification is represented in the figure.
Expert classification of sites into niches and transfer points. Overall, 55% of sites (42/77) were classified as niches, while 40%of sites (31/77) were classified as transfer points by the majority of responses (Figure 3.3). In addition, four sites (“footbath, “racks that are used for exposed finished product”, “bumper guards”, and “spiral freezers”) were classified as niches or transfer points by the same number of responses (Figure 3.3). Among the 42 sites classified by the majority of responses as niche, the 5 sites with the highest response as representing a niche were “bowl cutters”, “conduits, electrical boxes, conduit-in-casing”, “pipelines and exposed overhead piping in area with exposed product”, “interior of pipes and tubes that transfer finished product”, and “band saws” with these sites being classified by 73%, 81%, 75%,63%, and 79% of responses as a niche. Among the 31 sites classified by the majority of reviewers as transfer points, the 5 sites with the highest consensus as representing a transfer point were “Aprons that contact finished product”, “Boots/footwear worn by personnel who handle exposed product”, “Thermometers, thermocouples, etc. that contact finished products”, “Food preparation sink used for finished product” and “Scale” with these sites being classified by 100%, 81%,69%, 79%, and 69% of responses as a transfer point (Figure 3.3). Sites that were classified as niches are found across zones but interestingly 12 of the 19 sites that represented the 25% most important sites were classified as niches.
69

Zone 2
Sites
70

Ni TP
Zone 4

Zone 3

Zone 1

Bowlcutters Interior of pipes and tubes that transfer finished product
Mixers Grinders Mincers Band Saws
Valves Brine chiller chamber/tunnel
Fillers Skinning machine Slicer/peelers/choppers Brushes and other equipment that touch finished product Holding vats used to store finished product Conveyor systems, belt and other food contact sites
Flume Wash Bumper guards (e.g., on sorting tables, along conveyances) that contact finished product
Ice maker used for ice that comes in contact with exposed finished product Vacuum sealers Smoke sticks
Other Equipment (e.g., vats, tanks, tables) that comes in contact with product after CCP Cases/crates that hold finished products Packaging machine Baggers Cutting boards used for finished product Chutes
Re-wrap counter/table used for finished product Hopper Surface
Buckets/bins that come in contact with finished product Scale (surface that comes in contact with finished product)
Scraper blade Thermometers. Thermocouples, etc. that contact finished products
Sorting table (surfaces that contact finished product) Food preparation sink used for finished product Aprons that contact finished product
Pipelines and exposed overhead piping in area with exposed product Air blower
Framework and non food contact areas of food contact equipment Drip pans
Conveyor systems Cooler doors and other doors used for transport of open finished product
Racks that are used for exposed finished product Spiral freezers
Railings in finished product areas Flaps or strip curtains used in areas where exposed finished product is present
Maintenance tools Scale
Control switch/HMI screens on equipment close to FCS Boots/footwear worn by personnel who handle exposed product
Conduits, electrical boxes, conduit-in-casing Floor/wall junction
Ceiling in production area Cleaning equipment
Electrical outlet covers Racks that are not used for exposed finished product
Air return cover Walls Drains Fans
Floor mats Trash cans Portable steps/stool/ladder In-floor weighing equipment and floor scales Handwash sink Hoses (and hose holders) Exterior of ice maker Cooler doors and other doors not used for transport of open finished product
Footbath Trolleys, lifters, forklifts and pallet jacks
Hoist chain bags Carts Floor
Soap dispensers Boots/footwear worn by personnel who do not handle exposed product
Foot pedals Windows
Break room and locker room area floor Loading Dock

18 16 14 12 10 8 6 4 2 0

No. of reponses

Figure 3.3 Expert reviewers’ classification of sites into niches (Ni) and transfer sites (TS) and zones. The total number of reviewers’ responses is represented by the total number of reviewers that classified a given site.
Verification sites versus indicator sites. Among the 77 sites, 38 and 39 were classified by the majority of experts as verification and indictor sites, respectively (Figure 3.4). While all 34 Zone 1 sites were classified by the majority of experts as verification sites, only 5 of the Zone1 sites were classified by all reviewers (that provided a responses) as verification sites (Figure 3.4). The 14 Zones 2 sites were classified by 47 to 75% of reviewers as indictor sites. Zone 3 sites were classified as indicator sites by 63 to 81 and individual Zone 4 sites were classified as indicator sites by 81% of reviewers (Figure 3.4). Interestingly, there was not a single site that was classified by all reviewers as an indicator site.
71

Zone 2
72

Zone 4

Zone 3

Zone 1

Other Equipment (e.g., vats, tanks, tables) that comes in contact with product… Brushes and other equipment that touch finished product Cutting boards used for finished product Fillers Re-wrap counter/table used for finished product
Scale (surface that comes in contact with finished product) Thermometers. Thermocouples, etc. that contact finished products
Brine chiller chamber/tunnel Conveyor systems, belt and other food contact sites
Scraper blade Buckets/bins that come in contact with finished product Interior of pipes and tubes that transfer finished product
Mixers Valves Bowlcutters Grinders Hopper Surface Packaging machine Skinning machine Slicer/peelers/choppers Sorting table (surfaces that contact finished product) Vacuum sealers Food preparation sink used for finished product Cases/crates that hold finished products Bumper guards (e.g., on sorting tables, along conveyances) that contact… Chutes Holding vats used to store finished product Mincers Band Saws Flume Wash Aprons that contact finished product Ice maker used for ice that comes in contact with exposed finished product Baggers Smoke sticks Racks that are used for exposed finished product Pipelines and exposed overhead piping in area with exposed product Air blower Spiral freezers Conveyor systems Framework and non food contact areas of food contact equipment Maintenance tools Control switch/HMI screens on equipment close to FCS Railings in finished product areas
Scale Flaps or strip curtains used in areas where exposed finished product is present
Cooler doors and other doors used for transport of open finished product Drip pans
Boots/footwear worn by personnel who handle exposed product In-floor weighing equipment and floor scales Ceiling in production area Drains Floor/wall junction Handwash sink Floor Footbath Hoses (and hose holders) Portable steps/stool/ladder Trolleys, lifters, forklifts and pallet jacks Floor mats Trash cans Walls Exterior of ice maker Air return cover Carts Conduits, electrical boxes, conduit-in-casing Electrical outlet covers Fans Soap dispensers Foot pedals Cleaning equipment
Cooler doors and other doors not used for transport of open finished product Boots/footwear worn by personnel who do not handle exposed product Racks that are not used for exposed finished product Hoist chain bags Break room and locker room area floor Loading Dock Windows

18 Verification Site Indicator Site
16 14 12 10 8 6 4 2 0

No. of responses

Figure 3.4 Expert reviewers’ classification of sites into (VS) and indicator sites (IS) and zones. The total number of reviewers’ responses is represented by the total number of reviewers that classified a given site.
DISCUSSION This study provides an initial expert assessment of sampling sites that should be included
in Listeria PEM programs, including information on classification of different sites into zones, niches or transfer points and verification or indictor sites. The availability of a peer-reviewed document like the one presented here provides an important resource for industry, as the data shown here can be used to provide scientific support for design and implementation of Listeria PEM programs, including selection and classification of environmental sampling sites used. The data presented here (and in particular the sample site list shown in Table 3.3) will be particularly valuable as regulatory agencies as well as 3rd parties increasingly require scientifically supported food safety plans with validated controls and verification procedures. Zones Classification of PEM sampling sites. While zone classification of PEM sampling sites is very commonly used in both industry and government guidance documents (Stone, 2014; FSIS, 2014), classification of a given site to a zone is not always unambiguous and often depends on the specific spatial context of a facility. Zone classification of sampling sites may also differ depending on regulations and regulatory guidance documents. Hence the zone classifications provided here simply provide a guidance for which zone a given site may represent; final assignment of a sampling site that is part of a PEM plan must be determined based on the individual context of a given facility. The challenge in assigning zones to specific sites also is illustrated by the fact that for a number of sites in our final list, more than one possible zone is
73

indicated. For example, sites 3.5 (floor-wall junction) and 3.3 (floors) (see Table 3.3) may be identified as Zone 2 if the specific site is in production area (and particularly if it is in close proximity to equipment with exposed RTE product), while floor-wall junctions and floor further away from production areas (for example, in breakrooms and locker rooms) would be classified as Zone 3 or 4.
In addition to zone classification data, this study also provides valuable data on the relative importance of different sampling sites. These data can help industry and others to prioritize sampling sites, which could include more frequent collection of sampling sites that have been identified as high priority. Similarly to what was discussed about zone classification above, importance of different PEM sites can and may differ by facility characteristics and other factors though. Interestingly, while the vast majority of sites that achieved the highest importance ranking represent Zone 1 sites, drains were classified as the most important sites; drains were identified by 44 and 13% of reviewers as niches and transfer points. This result is consistent with a change in the approaches to Listeria control as in the early stages of development and implementation of environmental Listeria control program it was not uncommon to assume that drains would almost inevitable yield Listeria positive results and hence may not provide a particular value in Listeria PEM programs. Our data here clearly support in emerging industry consensus on the importance of including drains in Listeria PEM programs. Consistent with this, previous studies also identified drains as an indicator for overall facility contamination and proposed that drains may serve as a good predictor for L. monocytogenes presences in other sites (i.e., overhanging drip pans) (Rørvik et al., 1997; Hoffman et al., 2003). Identification of sites that likely represent niches. Expert classification of sampling site into
74

transfer points and niches also provides valuable information for the development of PEM programs, particularly since many of the PEM sites that ranked high for importance also were sites that represent potential niches. Our data on sites that represent likely Listeria niches also provides valuable information that will help industry and others identify sites that require particular attention with regard to sanitary equipment or facility design or implementation of specific and targeted sanitation practices. Examples of sites that were classified as niches by a large proportion of expert reviewers included drains, floor-wall junctures, and pipes with condensation (Table 3.3). Many of the sites that were classified by a large proportion of expert reviewers as niches had previously been proposed by other to represent potential niches. Sites that are identified as niches and that test positive for L. monocytogenes can be considered harborage sites. Zone 1. Previous studies by other groups have linked harborage sites to finished product contamination and outbreaks. For example, floor-wall junctions were identified as sites of persistent Listeria PFGE types in a study of retail delis (Simmons et al., 2014). Also, in a study of two fish plants, drains were identified as likely sites of L. monocytogenes persistence (Malley et al., 2013). Importantly, some industry guidance documents also include very comprehensive lists of areas known to harbor Listeria. For example, a guidance document from the Innovation Center for US Dairy lists “drains, cracked floors, condensation on walls/ceilings/pipes, damp pipe insulation, hoist chains, unsealed electrical conduits, wrapped/bundled cords, and electrical/hydraulic junction boxes” as well as “cooling units, drip pans, difficult-to-access surfaces, difficult-to-clean pieces of equipment such as conveyors, motor housings, bearings, undersides of equipment, pallet jacks, forklifts, and seasonal/limited-use equipment” as areas that are known by industry to harbor Listeria (Innovation Center for U.S. Dairy, 2015). Combined, these different documents that detail niches in processing
75

environments provide important resources for industry for both development of PEM programs as well as for development of sanitation programs. Expert classification of sites into verification and indictor sites. A key goal of a Listeria PEM program will be the verification of a food safety system. Increasingly, verification of the ability of a food safety system to control a target pathogen (e.g., L. monocytogenes) involves testing of both finished product samples as well as environmental samples collected at appropriate sites. Under this scheme, any pathogen positive verification sample would indicate a food safety system failure that would need to be followed up by appropriate corrective actions. Often, facilities will collect two sets of environmental samples, including (i) samples from verification sites and (ii) samples from indicators sites (typically Zone 3 and 4 sites); in this scenario, Listeria positive result on samples collected from indictor sites would not necessarily indicate a food safety system failure but would rather provide information that measures the effectiveness of early hurdles (e.g., door foamers) and facilitates the development of new interventions. Our data collected here show an emerging consensus that Zone 1 sites would typically represent verification sites, while Zone 3 and 4 sites typically would represent indictor sites. The fact that most Zone 2 sites were classified by the majority of experts as verification sites was somewhat surprising, since a number of recent workshops propose that Zone 2 sites may typically also be verification sites. The overall lack of a clear consensus on classification of Zone 2 sites and some Zone 3 and 4 sites (which would have expected to mostly be classified as verification sites) may be due to the fact that the concept of verification and validation sites still is somewhat recent.
CONCLUSIONS Based on an expert elicitation, this paper provides key information that can be used by 76

the food industry for the development and implementation of scientifically supported Listeria PEM programs. As PEM programs and appropriate sites will differ considerably in different facilities, we did not design our study to yield an expert consensus, but rather acquired and summarized expert opinions, which will inform implementation of PEM programs in individual facilities. Importantly, we do not mean to imply that the sampling sites provided here represent a complete list of all sites that should be included in a Listeria PEM program. Rather, the sites listed here provide starting point for site selection and individual facilities should expect to identify and add further sites that may be unique to a facility, process, or type of product produced. We also appreciate the PEM programs targeting organisms other than Listeria (e.g., Salmonella, Cronobacter) are essential in many facilities; the study reported here may provide starting point for similar expert elicitation efforts for these other pathogens.
77

REFERENCES
1. Autio, T., Lundén, J., Fredriksson-Ahomaa, M., Björkroth, J., Sjöberg, AM., Korkeala, H. 2002. Similar Listeria monocytogenes pulsotypes detected in several foods originating from different sources. International Journal of Food Microbiology 77:83–90.
2. Currie, A., Farber, J.M., Nadon, C., Sharma, D., Whitfield, Y., Gaulin, C., Galanis, Bekal, E., Flint, J., Tschetter, L., Pagotto, F., Lee, B., Jamieson, F., Badiani, T., MacDonald, D., the National Outbreak Investigation Team, Ellis, A., May-Hadford, J., McCormick, R., Savelli, C., Middleton, D., Allen, V., Tremblay, FW., MacDougall, L., Hoang, L., Shyng,S., Everett,D., Chui, L., Louie, M., Bangura, H., Levett, P.N., Wilkinson, K., Wylie, J., Reid, J., Major,B., Engel, D., Douey,D., Huszczynski, G., Lecci, JD., Strazds, J., Rousseau, J., Ma,K., Isaac, L.,Sierpinska, U. 2015. Multi-province listeriosis outbreak linked to contaminated deli meat consumed primarily in institutional settings, Canada, 2008. Foodborne pathogens and disease 12:645-652.
3. Ferreira, V., Wiedmann, M., Teixeira, P. and Stasiewicz, M.J. 2014. Listeria monocytogenes persistence in food-associated environments: epidemiology, strain characteristics, and implications for public health. Journal of Food Protection 77:150170.
4. United Fresh Produce. Guidance on environmental monitoring and control of Listeria for the fresh produce industry. 2013.
78

5. Ho, A.J., Lappi, V.R. and Wiedmann, M. 2007. Longitudinal monitoring of Listeria monocytogenes contamination patterns in a farmstead dairy processing facility. Journal of dairy science 90:2517-2524.
6. Kabuki, D.Y., Kuaye, A.Y., Wiedmann, M. and Boor, K.J. 2004. Molecular subtyping and tracking of Listeria monocytogenes in Latin-style fresh-cheese processing plants. Journal of dairy science 9:2803-2812.
7. Laksanalamai, P., Joseph, L.A., Silk, B.J., Burall, L.S., Tarr, C.L., Gerner-Smidt, P. and Datta, A.R. 2012. Genomic characterization of Listeria monocytogenes strains involved in a multistate listeriosis outbreak associated with cantaloupe in US. PLoS One 42448.
8. Lappi, V.R., Thimothe, J., Nightingale, K.K., Gall, K., Scott, V.N., Wiedmann, M. 2004. Longitudinal studies on Listeria in smoked fish plants: impact of intervention strategies on contamination patterns. Journal of Food Protection 67:2500–2514.
9. The Grocery Manufacturers Association. Listeria monocytogenes guidance on environmental monitoring and corrective actions in at-risk foods. 2014.
10. Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R.V., Widdowson, MA., Roy, S.L., Jones, J.L., Griffin, P.M. 2011. Foodborne illness acquired in the United States— major pathogens. Emerging Infectious Diseases 17.
11. Hoffman, S., Batz, M.B., Morris, J. 2012. Annual cost of illness and quality-adjusted life year losses in the United States due to 14 foodborne pathogens. Journal of Food Protection 75:1292–1302.
79

12. Hoffman, A.D., Gall, K.L., Norton, D.M., Wiedmann, M. 2003. Listeria monocytogenes contamination patterns for the smoked fish processing environment and for raw fish. Journal of Food Protection 66:52–60.
13. Malley, T.J., Stasiewicz, M.J., Gröhn, Y.T., Roof, S., Warchocki, S., Nightingale, K. and Wiedmann, M. 2013. Implementation of statistical tools to support identification and management of persistent Listeria monocytogenes contamination in smoked fish processing plants. Journal of Food Protection 76:796-811.
14. Malley, T.J., Butts, J. and Wiedmann, M. 2015. Seek and destroy process: Listeria monocytogenes process controls in the ready-to-eat meat and poultry industry. Journal of Food Protection 78:436-445.
15. McIntyre, L., Wilcott, L., Naus, M. 2014. Listeriosis outbreaks in British Columbia, Canada caused by soft ripened cheese contaminated from environmental sources. BioMed Research International.
16. Norton, D.M., McCamey, M.A., Gall, K.L., Scarlett, J.M., Boor, K.J., Wiedmann, M. 2001. Molecular studies on the ecology of L. monocytogenes in the smoked fish processing industry. Applied and environmental microbiology 67:198–205.
17. Orsi, R.H., Borowsky, M.L., Lauer, P., Young, S.K., Nusbaum, C., Galagan, J.E., Birren, B.W., Ivy, R.A., Sun, Q., Graves, L.M. and Swaminathan, B. 2008. Shortterm genome evolution of Listeria monocytogenes in a non-controlled environment. BMC genomics 9:1.
18. Sauders, B.D., Mangione, K., Vincent, C., Schermerhorn, J., Farchione, C.M., Dumas, N.B., Bopp, D., Kornstein, L., Fortes, E.D., Windham, K. and Wiedmann, M. 2004. Distribution of Listeria monocytogenes molecular subtypes among human and 80

food isolates from New York State shows persistence of human disease–associated Listeria monocytogenes strains in retail environments. Journal of Food Protection, 67:1417-1428. 19. Simmons C, Stasiewicz M.J., Wright E, Warchocki S, Roof S, Kause J.R., Bauer N, Ibrahim S, Wiedmann M, Oliver H.F. 2014. Listeria monocytogenes and Listeria spp. contamination patterns in retail delicatessen establishments in three US states. Journal of Food Protection 77:1929-1939. 20. Stone, W.E., 2014. Regulatory Testing Guidelines and Recommendations. In The Microbiological Safety of Low Water Activity Foods and Spices 347-366. 21. Thomas, M.K., Vriezen, R., Farber, J.M., Currie, A., Schlech, W. and Fazil, A. 2015. Economic cost of a Listeria monocytogenes outbreak in Canada, 2008. Foodborne Pathogens and Disease 12:966-971. 22. Vongkamjan, K., Roof, S., Stasiewicz, M.J. and Wiedmann, M. 2013. Persistent Listeria monocytogenes subtypes isolated from a smoked fish processing facility included both phage susceptible and resistant isolates. Food microbiology 35:38-48. 23. Wilkin, E., Brouillette, R., Carver, J., Cords, B., Dhillon, A., Hall, S., Hedge, T., Kedzierski, D., Ledenbach, L., Lewandowski, V., Smith, M., Stout, J., Stubbs, T. 2015. Control of Listeria monocytogenes: guidance for the U.S. dairy industry. Innovation Center for U.S. Dairy.
81

CHAPTER 4 LEARNING OUTCOMES OF A K – 12 FOOD SAFETY MINI-COURSE
ABSTRACT A graduate student instructor led a 5 day food safety mini-course in 4 middle school life
science classes. The purpose of the mini-course was to introduce local middle school-age students to the fundamentals of foodborne illness outbreak investigation. Through a series of 5 lessons, students learned about foodborne pathogens and basic epidemiological principles by role playing and acting out a fictitious foodborne illness outbreak investigation. Instructors employed a variety of educational approaches to encourage classroom participation and help with retention (i.e., jigsaw learning and scaffolding). Students’ knowledge increased from 41% to 81% on preand post-assessment evaluation content related to handling raw and cooked foods, handwashing, and identifying common foodborne pathogens in the U.S. Students’ performance on assessments and classwork engagement increased as their familiarity with content progressed. These findings, augmented by previous studies, suggest that further translation of food safety content into age appropriate middle school science curriculum to promote science literacy and boost general understanding of food safety principles that could directly influence behavior and mold safe, lifelong habits.
INTRODUCTION Each year in the United States there are 9 million cases of foodborne illnesses due to
known foodborne pathogens that lead to 55, 961 hospitalizations and 1,351 deaths (19). Poor food handling and health and hand hygiene practices are commonly implicated in the
82

transmission of foodborne illness at home (1, 8), institutional settings (e.g., schools, day care, hospitals, schools, etc.) (17), restaurants (11, 12) and in public settings (5, 12, 24). Local, state, and federal government agencies and local non-profit groups outreach initiatives aim to educate consumers and school-age children on the importance of basic food safety practices to reduce the occurrence of foodborne illnesses and prevent the spread of foodborne illness via foods and person-to-person transmission (4, 18). There are a number of educational campaigns targeted towards the general public to teach foodborne illness prevention, such as “Fightbac” and ”Be Food Safe: Clean, Separate, Cook, Chill” national campaigns (7, 25).
In addition to campaigns, university cooperative extensions and local government and non-government agencies partner with primary and secondary schools to (i) develop curricula and instructional materials (2, 3, 6, 22), (ii) promote a culture of food safety among young children that will lead to the development of lifelong behaviors (21, 22), and (iii) equip students with the necessary skill set required for the food service industry-- the most concentrated sector of teenage workers (15, 20). In addition to teaching food safety principles, institutions invest in opportunities to introduce school-age children to agricultural and life science disciplines (e.g., food science and food safety) through a number of outreach efforts such as 4H, summer programs, and graduate student outreach programs (6). In a food science or food safety context, the purpose of the programs is to stimulate interest in the discipline and foster pre-existing interest by engaging students through hands-on activities and other enrichment opportunities.
In the following paper, we describe techniques employed to teach 4 groups of 7th grade life science students basic food safety concepts. In addition, we describe the use of evaluations to measure retention and overall scientific literacy as outlined by the American Association for the Advancement of Science (AAAS). The purpose of the paper is to describe the design and
83

outcomes of a 5-day outbreak investigation mini-course.
METHODS Mini-course overview. A five day food safety mini-course was developed for middle school-age students to introduce basic food safety fundamentals and outbreak investigation principles. A graduate student modified a lesson format developed by the University of Delaware and with the help of a teacher, taught the mini-course in 4 life science classes at a local middle school. Instruction was tailored daily to accommodate individual class sizes.
A pre-assessment evaluation that consisted of 10 general food safety-related questions and multiple choice and true/false answer choices was administered to students. Students completed the pre-assessment at the beginning of Day 1 of the mini-course. Day 1 served as an introduction to the concept of foodborne outbreak illness investigation where instructors provided students with a very general foundation of food safety. Instructors introduced students to the topic of food safety by discussing key concepts such as good hand hygiene practices and acceptable food preparation techniques. Classes were divided into 5 groups, each group was given a term (i.e., cross-contamination, investigation, outbreak, foodborne pathogen, temperature danger zone) and students were instructed to discuss and reach a group consensus to define the word. In each group, instructors randomly designated 1 student to serve as the “expert” of the group and after 3 – 5 minutes of free discussion, all students, except for the “expert” were instructed to rotate to a different group. Experts were encouraged to facilitate the discussion and help new groupmates define or create a description for the assigned term. Another expert was chosen and the students rotated to a new group. Students repeated discussion and rotations until
84

all 5 terms were defined. Instructors began Day 2 of the mini-course with a recap of the previous day and student
volunteers defined the 5 terms. Instructors introduced the outbreak and provided a foundation for a foodborne illness outbreak investigation via a memorandum from a fictitious health department. Instructors distributed materials and divided the class into groups and designated approximately 1– 2 “epidemiologists” and 5 “victims” per group. “Victims” read provided scripts or scenarios and “epidemiologists” asked questions to gain a better context of the victims’ symptoms, food they consumed days prior to feeling ill, etc. Students were instructed to organize data into tables and identify notable patterns or relationships.
Day 3 began with a Bell Ringer, a 3 minute assessment of Day 2 and a review of students’ data tables. Lesson 3 began with a memorandum from the New York State Department of Millennials updating the “epidemiologists” on the status of the foodborne illness outbreak investigation. Students continued to role play and each received a script which served as their victim profile. Student victims came to the front of the classroom and read their script while student epidemiologists interviewed and organized information into tables.
Day 4 included a Bell Ringer and a re-cap of the previous day’s activity. Instructors asked students if they could identify patterns in the data that they collected during interviews. Instructors introduced students to odds ratio and students used the data collected during the previous day’s lesson to calculate the odds ratio for individual foods.
On Day 5, instructors gave a recap of Day 4 and asked students to describe the outcome of the odds ratio. Students discussed their odd ratio calculations and identified the product most likely responsible for illnesses, the salad. Students identified different ingredients in a traditional chef salad and concluded that additional information was required to determine the implicated
85

ingredient. After identifying reasonably likely ingredients, instructors introduced students to the concept of DNA fingerprinting to further identify the source of the outbreak. Instructors provided the students with a picture of an electrophoresis gel with a reference lane and what was described as Listeria monocytogenes (LM) DNA fingerprint isolated from student and teacher victims and various foods in a salad-- lettuce, tomatoes, onions, carrots, and cheese. Students were able to identify that LM DNA isolated from onions closely matched DNA from human isolates which marked the conclusion of the mini-course.
RESULTS AND DISCUSSION We designed a fictitious foodborne illness outbreak scenario to include a foodborne incident at a fictitious New York year-round school. In groups, students learned the roles of epidemiologists during outbreak investigation by asking questions based on the provided victims’ profiles. The objective of Lesson 2 was to (i) define a scientific problem by introducing a foodborne illness outbreak at a fictitious school (ii) introduce epidemiologists to students via role playing and (iii) organize data into tables by identifying patterns and relationships as described in Next Generation Science Standards. Role playing was a successful component of the mini-course as it provided age and culture appropriate content in addition to educational instruction. Because students were comfortable with the delivery of the content, they remained engaged and completed every activity. Bell Ringers were designed to serve as a review of the previous day’s ontent and the re-caps were found to be particularly important to remind students of what they did and provide a foundation for students that may have been absent from class the previous day. Pre-Assessment: All questions were food safety related and designed to assess the students’ knowledge of food preparation, hand hygiene practices, and most common foodborne pathogens.
86

Sixty-seven students were present on Day 1 of the mini-course and completed a pre-assessment evaluation. Student scored an average of 42% with the highest of 70% (7/10) and the lowest of 20% (2/10) (Table 4.1). Eighty-five percent of students (n=57) answered Question 8 correctly, followed behind Question 7 where 56 students (84%) responded correctly. Students performed the lowest on Questions 4 and 3 with 5 (7%) and 10 (15%) responding correctly respectively. For the remaining 6 questions, 39% – 47% of the class responded correctly. Post-Assessment: The Post-assessment was the same as the Pre-assessment. Seventy-one students completed the Post-assessment evaluation after the Day 5 of the mini-course. Students scored an average of 81% with the highest of 100% (n=13) and the lowest of 40%. All seventyone students answered Question 7 correctly, followed behind Question 5 where 65 (92%) of students responded correctly. Students missed Questions 4 and 6 the most with a response rate of 54% (n=38) and 58% (n=41) respectively. The correct response rate for the remaining questions ranged from 72% (n=51) to 93% (n=66). Comparing pre and post-assessment evaluations, overall students’ performance doubled and every student that completed a post-assessment, scored above a 50% with the exception of one student. Of the students that completed both pre and post-assessments, 85% (n=67) scored at or above 70% with the greatest improvement of 8 points (80%) between assessments (Figure 4.1). Overall, students performed significantly better on post-assessments evaluations than pre-assessment evaluations (Figure 4.2).
87

Table 4.1 Results of pre and post-assessments indicated that students comprehension of basic food safety concepts improved significantly

Question
1 At what temperature should you keep your refrigerator? 2 What is the minimum length of time should you wash your hands? 3 Always wash your chicken. True or False? 4 Which bacterium is responsible for the greatest number of diarrhea-related illnesses each year? 5 Which is not a safe way to defrost meat? 6 The majority of foodborne illnesses are a result of which of the following? 7 What are the symptoms of foodborne illness (food poisoning)? 8 What is the best way to prevent foodborne illness (food poisoning)? 9 Where should raw meat be stored in the refrigerator? 10 How can you tell if food has enough bacteria to cause foodborne illness (food poisoning)? * Answers report 67 students that completed pre- and post-assessments

Correct answers (%)*
Pre Post 34 81 36 72 15 96 7 55 37 97 33 57 84 100 85 84 48 91 39 88

Percent improvement
47 36 81 48 60 24 16 -1 43 49

88

Correct no. of responses

12 Series2
10 Series3 8 6 4 2 0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67
Individual Students
Figure 4.1 Individual students’ performance on pre and post-assessments. Correct number of responses correlates to the total number of questions on the assessments that students answered correctly. Students that were present for both assessments (n=16) scores are reflected.
89

The pre and post-assessment not only served as an opportunity to measure students learning and retention of food safety-relate concepts throughout the mini-course but it also prompted discussions that were relevant in a science class and educational setting but not specific to food safety. For example, one of the choices included, “the common cold.” During the review of questions and answers, instructors took a moment to remind students that the common cold is caused by a virus, not a bacterium. Similar examples provided opportunities to re-educate and highlight scientific principles that influence recommendations.
Bell Ringers: Students completed “Bell Ringers” as a review of the previous day’s lesson the first few minutes of class on Days 3 and 4. On Day 3, students completed a Bell Ringer that consisted of 3 separate questions where each question was worth 1 point. Questions 2 and 3 required 3 individual responses, therefore, each correct individual response was counted as a proportion of “1” and the highest score a student could receive was 3. For example, if a student provided 2 correct responses for Question 2, 0.67 of a possible “1” point was given to that student. Sixty-four students averaged an 85% with 70% (n=45) of students scoring 78% or greater and the lowest score awarded was 22%. Question 2 had the highest response rate (93%) with followed by question 3 (89%) and question 1 (71%).
On Day 3, students completed another Bell Ringer (Bell Ringer 2) at the beginning of class. Similar to the previous day, Bell Ringer 2 consisted of 3 questions taken directly from the previous day’s discussion and activity. Each correct response was awarded “1” point and partial credit was not awarded. Sixty-seven students were present and completed Bell Ringer 2. The students earned an overall class score of 94% on Bell Ringer 2 with 88% (n=59) receiving full credit for responses. Among the remaining students scored, 7.5% (n=5) and 3% (n=2) received 67% and 33% respectively. One student attempted all of the questions; however, none of the
90

responses were correct. Correct responses for Bell Ringer 2 fell within 11 categories. For Bell Ringer 2, students
were instructed to list 3 questions an epidemiologist would ask during an interview (Table 4.2). Almost every student (n=55) asked a variation of “what have you eaten recently” or “what kinds of food do you eat?” Only 1 student asked “where did you eat” and “have you shared food” and these questions were placed into separate categories since in a real-life practical setting, an epidemiologist could conceivably ask a variation of these questions, especially in a school setting.
91

Table 4.2 Bell Ringer No. 2 Questions and Categories
No. Student Response
1 What have you eaten recently? 2 What were/are your symptoms? 3 When did you first experience symptoms/how long? 4 Do you have any pre-existing health conditions? 5 Current health status 6 Where did you eat? 7 Have you shared any food? 8 How old are you? 9 Gender 10 Have you been out of the country recently/ Where have you
traveled? 11 Have you been around others with similar symptoms? INC Incorrect response

1st Response 29 10 6 2 2 0 0 15 0 0

2nd Response 16 22 12 1 0 0 1 7 1 2

3rd Response 10 9 14 11 4 1 0 8 1 2

Total Responses 55 41 32 14 6 1 1 30 2 4

0033 3 5 4 12

92

The development of Lesson 3 materials included designing a case profile of the total cases reported in the foodborne outbreak at Lions Central School. The case profiles included, a description of the victim’s symptoms, the on-set date of symptoms, age, gender, pre-existing health conditions, and current status. Each student present role-played and shared the information with the rest of the class who served as the epidemiologists. Based on the student-case information, student epidemiologist collected the information presented by the student-case and filled in into a table. After the completion, every student present role-played and shared the information with the rest of the class that served as the epidemiologists. Based on the student “case” information, student epidemiologist collected the information presented by the student “case” and filled information into a table. By the completion of the exercise, students completed a table with case and case control information.
Any information offered was supported by published peer-reviewed research and federal government recommendations (Kosa et al, 2015; Henley, 2013). Students, however, were skeptical and resistant to some of the information that they received and generally agreed in a few instances that they probably would not modify certain food safety practices in the future. For example, the USDA does not recommend “washing” or “rinsing” chicken (USDA 2013) before cooking. During an open discussion, students mentioned that their families regularly rinse chicken and that they never got sick. Students were, however more receptive to USDA recommendation of not washing chicken after instructors shared a publicly available Youtube clip highlighting cross-contamination of “germs” from chicken to various kitchen surfaces and other foods. The clip traced germ (Glo-Germ) transmission inside the sink, around sink handles, on countertops, and into an open bowl of leafy green salad as the preparer rinsed chicken. Consistent with public health statistics, when asked, the majority of students admitted that when
93

they had an incident of “food poisoning,” they did not seek medical care. Interestingly, students initially did not identify and acknowledge a relationship between in-home food handling practices (i.e., rinsing chicken) and food poisoning. To mitigate skepticism and resistance, in addition to showing the Youtube clip, instructors shared a real-life example with students. In the U.S., around the time that the mini-course was executed, bagged lettuce contaminated with Listeria monocytogenes was implicated in an outbreak that hospitalized 19 people. The minicourse concluded with students connecting the relevance of food safety and the mini-course to a real-life context.
Time did not permit a hands-on DNA fingerprinting activity but this mini-course and similar mini-courses may be designed to include an applied component to teach molecular biology and fingerprint typing techniques. For example, in life science classes, instructors could offer the mini-course during their genetics unit and teach DNA replication using legos. In higher grade levels or where appropriate, legos may be used to teach general molecular typing concepts and methods in tandem when teaching Polymerase Chain Reaction (PCR).
94

REFERENCES
1. Ali, M.M., Verrill, L. and Zhang, Y. 2014. Self-reported hand washing behaviors and foodborne illness: a propensity score matching approach. Journal of Food Protection 77:352-358.
2. Burke, A. and Dworkin, M.S. 2016. High School Students as the Target of an Integrated Food Safety Educational Intervention: Successful Results of a Pilot Study. Food Protection Trends 36:206-220.
3. Byrd‐Bredbenner, C., Abbot, J.M. and Quick, V. 2010. Food safety knowledge and beliefs of middle school children: implications for food safety educators. Journal of Food Science Education 9:19-30.
4. Cates, S.C., Carter-Young, H.L., Conley, S. and O'Brien, B. 2004. Pregnant women and listeriosis: preferred educational messages and delivery mechanisms. Journal of nutrition education and behavior 36:121-127.
5. Centers for Disease Control and Prevention. 2013. Handwashing: clean hands save lives.
6. Chapin, T.K., Pfuntner, R.C., Stasiewicz, M.J., Wiedmann, M. and Orta‐Ramirez, A. 2015. Development and Evaluation of Food Safety Modules for K‐12 Science Education. Journal of Food Science Education 14:48-53.
7. Dharod, J.M., Pérez-Escamilla, R., Bermúdez-Millán, A., Segura-Pérez, S. and Damio, G. 2004. Influence of the Fight BAC! food safety campaign on an urban Latino population in Connecticut. Journal of Nutrition Education and Behavior 36:128-134.
95

8. Fischer, A.R., De Jong, A.E., Van Asselt, E.D., De Jonge, R., Frewer, L.J. and Nauta, M.J. 2007. Food safety in the domestic environment: an interdisciplinary investigation of microbial hazards during food preparation Risk Analysis 27:1065-1082.
9. Food Safety and Inspection Service. 2013. Washing food: does it promote food safety? USDA.
10. Freeman, S., Eddy, S.L., McDonough, M., Smith, M.K., Okoroafor, N., Jordt, H. and Wenderoth, M.P. 2014. Active learning increases student performance in science, engineering, and mathematics. Proceedings of the National Academy of Sciences 111:8410-8415.
11. Green, L.R., Selman, C.A., Radke, V., Ripley, D., Mack, J.C., Reimann, D.W., Stigger, T., Motsinger, M. and Bushnell, L. 2006. Food worker hand washing practices: an observation study. Journal of Food Protection 69:2417-2423.
12. Hall, A.J., Eisenbart, V.G., Etingue, A.L., Gould, L.H., Lopman, B.A. and Parashar, U.D. 2012. Epidemiology of foodborne norovirus outbreaks, United States, 2001– 2008. Emerg Infect Dis 18:1566-1573.
13. Hall, A.J., Wikswo, M.E., Pringle, K., Gould, L.H. and Parashar, U.D. 2014. Vital signs: foodborne norovirus outbreaks—United States, 2009–2012.MMWR Morb Mortal Wkly Rep 63:491-5.
14. Henley, S.C. 2013. “Don’t Wash Your Chicken!” Results of an Interdisciplinary Approach to Reduce Incidence of Infectious Foodborne Diseases.
15. Hirschman, C. and Voloshin, I. 2007. The structure of teenage employment: Social background and the jobs held by high school seniors. Research in social stratification and mobility 25:189-203. 96

16. Kosa, K.M., Cates, S.C., Bradley, S., Chambers, I.V. and Godwin, S. 2015. Consumer-reported handling of raw poultry products at home: results from a national survey. Journal of Food Protection 78:180-186.
17. Kwon, Junehee. 2012. Produce Handlers’ Handwashing Behaviors in Secondary School Food Service Facilities. IAFP Presentation
18. Medeiros, L. 2015. The Healthy Baby, Healthy Me Food Safety Education Program for Pregnant Women: Results of an Education Intervention. 2015 IAFP Annual Meeting Presentation.
19. Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R.V., Widdowson, M.A., Roy, S.L., Jones, J.L. and Griffin, P.M. 2011. Foodborne illness acquired in the United States— major pathogens. Emerg Infect Dis 17.
20. PEW Research Center. 2015. Where teens are finding summer jobs: more food service, less retail. Data extrapolated from the U.S. Bureau of Labor Statistics from July 200 through July 2004.
21. Powell, D.A., Jacob, C.J. and Chapman, B.J. 2011. Enhancing food safety culture to reduce rates of foodborne illness. Food Control 22:817-822.
22. Richards, J., Skolits, G., Burney, J., Pedigo, A. and Draughon, F. 2008. Validation of an interdisciplinary food safety curriculum targeted at middle school students and correlated to state educational standards. Journal of Food Science Education 7:54-61.
23. Stahl, S.A. and Fairbanks, M.M. 1986. The effects of vocabulary instruction: A modelbased meta-analysis. Review of educational research 56:72-110.
24. Stedman-Smith, M., DuBois, C.L., Grey, S.F., Kingsbury, D.M., Shakya, S., Scofield, J. and Slenkovich, K. 2015. Outcomes of a pilot hand hygiene randomized 97

cluster trial to reduce communicable infections among US office-based employees. Journal of Occupational and Environmental Medicine 57:374. 25. Trepka, M.J., Newman, F.L., Davila, E.P., Matthew, K.J., Dixon, Z. and Huffman, F.G. 2008. Randomized controlled trial to determine the effectiveness of an interactive multimedia food safety education program for clients of the special supplemental nutrition program for women, infants, and children. Journal of the American Dietetic Association 108:978-984.
98

CHAPTER 5 CONCLUSIONS
Since L. monocytogenes is very common in the environment and is an ongoing concern food industry, specifically industries that handle ready-to-eat (RTE) products, specifically RTE deli meats, federal government agencies alongside universities continue to work together to identify opportunities to minimize risk of contamination with the goal of preventing illnesses. As researchers and government officials study opportunities to minimize L. monocytogenes contamination in RTE deli products post heat-treatment, they report findings that may have broader implications for other industries as well. For example, since 2015, the U.S. have experienced several food recalls and outbreaks due to L. monocytogenes contamination. Several of the foods involved were RTE products such as ice-cream and lettuce that impacted consumers in different regions of the U.S. Because of its ability to survive a range of temperatures, L. monocytogenes is very common in the environment; therefore impacts a range of food commodities in addition to RTE foods. Our research, supplemented with data from previous investigations, provides the necessary foundation to help us to: understand L. monocytogenes ecology and transmission in one particular setting--retail delis, and interpret and use findings to identify opportunities to minimize contamination in non-retail deli facilities as well.
Our first objective was to measure L. monocytogenes and other L. spp prevalence in the retail delicatessen environment. Our findings indicates that there is a 9.5% (427/4503) and 5.3% (237/4503) L. monocytogenes and L. spp prevalence rate in retail delis, respectively. Moreover, our results are consistent with data gathered from other groups that identified that L. monocytogenes prevalence ranging from 13 – 35% in the retail deli environment. We also
99

identify that the concentration of L. monocytogenes isolated across non-food contact surfaces (NFCS) does not differ significantly (not reported). These results provided the preliminary data needed to study L. monocytogenes transmission and contamination patterns across food contact surfaces (FCS) and NFCS.
Our second objective was to identify contamination patterns using molecular-based allelic and subtyping approaches. sigB gene PCR and sequencing and Pulsed-field Gel Electrophoresis (PFGE) results allowed us to go beyond presence and absence L. monocytogenes and other L. spp. testing to differentiating between persistence and sporadic events in individual delis. While we observe persistence and contamination in 40% (12/30) of delis, we conclude that there is minimal contamination across NFCS to FCS since L. monocytogenes isolated from FCS were rarely isolated from NFCS during the same sampling visit or subsequent visits. These data allow individual delis to design intervention strategies with the goal of minimizing prevalence and persistence by targeting potential harborage sites.
Inspired by our retail deli ecology study, we continued to focus on mitigating L. monocytogenes from the food processing and retail environments. We understand that Listeria is wide-spread, our data indicates there is greater Listeria persistence across specific sites/surfaces (i.e., squeegees and drains) than others, and based on our data and data from other groups, past and present, we know that the meat, seafood, dairy, and produce industries, to name a few, struggle with Listeria prevalence and persistence. For our third objective, we developed a tool synthesizing experts’ opinions regarding environmental monitoring sites. The initial list that we proposed was not all inclusive and we relied on experts in the field to help us accurately classify sites into zones and distinguish between transfer points and niches and verification sites and indicator sites. With their help, we identified 83 sites (77 initial sites and 6 additional sites
100

recommended by experts) in total to include in a Pathogen Environmental Monitoring Program (PEM) for Listeria. Our assessment tool can be used for a range of food industry audiences as it is the initial step in developing a PEM. Facilities with existing PEMs can review our tool to identify additional sites to sample for routine and validation environmental sampling and testing.
Further inspired by research findings, we developed content and taught a food safety mini-course. Our objectives were to teach food safety principles and stimulate interest in food safety and food science by using specific educational techniques. We found that it was important to target middle school-age students because (i) habits are impressed at this age, (ii) many will work in the food service sector as teenagers, (iii) children become more involved with meal preparation around this age, and (iv) they are the next generation of scientists and require exposure to different disciplines at a young age to gain literacy thereby increasing interest.
We have identified approaches to prevent foodborne illness from environmental sources at retail and in food processing facilities to middle-grade education. While a broad food industry audience can appreciate our work, we acknowledge that every food facility does not have the resources to make pathogen environmental monitoring a priority for their operation. Based on our work and others in the field, we recommend that food facilities organize enhanced cleaning and sanitation efforts around sites that traditionally test highest (or routinely) for L. monocytogenes. Furthermore, we are aware that L. monocytogenes is one of many foodborne organisms to test for in the environment. Our work is reproducible, given the appropriate protocols, and can serve as a framework for similar studies and efforts across industries targeting other foodborne organisms.
101

APPENDIX FIGURE LISTERIA SUBTYPING RESULTS FOR 30 DELIS Complete environmental sampling and subtyping results from both Phase I and Phase II for Delis 1 – 30. Each figure shows the sampling results for each site over 9 sampling months for Delis 1 – 15 and 6 sampling months for Delis 16 – 30, sorted by contact surface type. Sites that have no samples were included as they were sampled in other delis in the data set. “NT” indicates that no sample was collected during that given month. “-“ indicates that the sites was sampled but tested negative for L. monocytogenes. “LM” indicates that a sample tested positive for L. monocytogenes; however, the PFGE assignment is not available. Using an internal laboratory nomenclature, PFGE subtype(s) are represented in alpha-numeric starting with “CU”.
Appendix Figure 1 Deli 1, 1 sample tested positive for L. monocytogenes. 102

Appendix Figure 2 Deli 2, 49 samples tested positive for L. monocytogenes. 103

Appendix Figure 3 Deli 3,1 sample tested positive for L. monocytogenes. 104

Appendix Figure 4 Deli 4, 6 samples tested positive for L. monocytogenes. 105

Appendix Figure 5 Deli 5, No sample tested positive for L. monocytogenes. 106

Appendix Figure 6 Deli 6,1 sample tested positive for L. monocytogenes. 107

Appendix Figure 7 Deli 7, 53 samples tested positive for L. monocytogenes. 108

Appendix Figure 8 Deli 8, 15 samples tested positive for L. monocytogenes. 109

Appendix Figure 9 Deli 9, 1 sample tested positive for L. monocytogenes. 110

Appendix Figure 10 Deli 10, 41 samples tested positive for L. monocytogenes. 111

Appendix Figure 11 Deli 11, 3 samples tested positive for L. monocytogenes. 112

Appendix Figure 12 Deli 12, 5 samples tested positive for L. monocytogenes. 113

Appendix Figure 13 Deli 13, 9 samples tested positive for L. monocytogenes. 114

Appendix Figure 14 Deli 14, 1 sample tested positive for L. monocytogenes. 115

Appendix Figure 15 Deli 15, no samples tested positive for L. monocytogenes. 116

Appendix Figure 16 Deli 16, 22 samples tested positive for L. monocytogenes. 117

Appendix Figure 17 Deli 17, 5 samples tested positive for L. monocytogenes. 118

Appendix Figure 18 Deli 18, 5 samples tested positive for L. monocytogenes. 119

Appendix Figure 19 Deli 19, 4 samples tested positive for L. monocytogenes. 120

Appendix Figure 20 Deli 20, 2 samples tested positive for L. monocytogenes. 121

Appendix Figure 21 Deli 21, 28 samples tested positive for L. monocytogenes. 122

Appendix Figure 22 Deli 22, 13 samples tested positive for L. monocytogenes. 123

Appendix Figure 23 Deli 23, 50 samples tested positive for L. monocytogenes. 124

Appendix Figure 24 Deli 24, 44 samples tested positive for L. monocytogenes. 125

Appendix Figure 25 Deli 25, 1 sample tested positive for L. monocytogenes. 126

Appendix Figure 26 Deli 26, 4 samples tested positive for L. monocytogenes. 127

Appendix Figure 27 Deli 27, 11 samples tested positive for L. monocytogenes. 128

Appendix Figure 28 Deli 28, 55 samples tested positive for L. monocytogenes. 129

Appendix Figure 29 Deli 29, 13 samples tested positive for L. monocytogenes. 130

Appendix Figure 30 Deli 30, no samples tested positive for L. monocytogenes. 131