Microbiological Testing Results of Cooked Diced Chicken and Pasteurized Egg Products Purchased for United States Federal Nutrition Assistance Programs,2012

Abstract

The Agricultural Marketing Service (AMS) purchases cooked diced chicken, pasteurized liquid whole eggs, and pasteurized dried egg mix for federal nutrition assistance programs. Purchases are made from establishments that have met the financial and technical requirements to become AMS-approved vendors. Cooked diced chicken is tested for the presence of Salmonella and Listeria monocytogenes, and for levels of aerobic plate count (APC) organisms, coliforms, and generic Escherichia coli (GEC). Of 3668 samples collected from October 2012 through September 2018, none were positive for Salmonella, 3 (0.8%) were positive for L. monocytogenes, 8 (0.22%) exceeded the APC critical limit (CL) of 1000 colony-forming units (CFUs)/mL, 15 (0.41%) exceeded the coliform CL of 50 CFU/mL, and 5 (0.14%) exceeded the GEC CL of 10 CFU/mL. Pasteurized liquid whole egg and pasteurized dried egg mix are tested for the presence of Salmonella and for levels of APC and CL. Of 984 pasteurized liquid whole egg samples collected from October 2012 through September 2018, 1 (0.10%) was positive for Salmonella, 29 (2.5%) exceeded the APC CL, and 4 (0.41%) exceeded the coliform CL. Of 380 pasteurized dried egg mix samples collected during this period, none was positive for Salmonella, none exceeded the APC CL, and 3 (0.79%) exceeded the coliform CL. All production lots from which samples found to contain pathogens or to exceed indicator organism CLs were identified and rejected for purchase by AMS. These data suggest that cooked diced chicken and pasteurized egg products produced for federal nutrition assistance programs are done so under effective food safety systems.

Introduction

The Agricultural Marketing Service (AMS) purchases food for federal nutrition assistance programs. Approximately 75% of food is distributed to recipients of Child Nutrition Programs, including the National School Lunch Program and the School Breakfast Program, while ∼25% goes to recipients of food assistance and disaster relief programs, including the Commodity Supplemental Food Program, the Food Distribution Program on Indian Reservations, and the Emergency Food Assistance Program (Agricultural Marketing Service, 2019a).

Cooked diced chicken and pasteurized egg products, due to their nutritional benefits, ease of preparation, and varied uses are popular items among federal nutrition assistance program recipients. These items are generally recognized as safe and have not been implicated in outbreaks among program recipients. However, given the potential vulnerability of some program recipients and that the items may be consumed without an additional lethality step, stakeholders have indicated a low tolerance for risk of microbiological contamination.

During Fiscal Years (October through September) 2013 through 2018, AMS purchased ∼11.0 million pounds of cooked diced chicken and 6.0 million lbs of pasteurized egg products per year (Agricultural Marketing Service, 2019a). Cooked diced chicken is typically used by program recipients as a topping at school salad bars, in burritos and tacos, and as a standalone item, whereas the two pasteurized egg product items procured by AMS—liquid whole eggs and pasteurized dried egg powder—are typically used to prepare scrambled eggs, omelets, and as ingredients in dressings, cakes, and like items.

To be eligible to bid on solicitations issued by AMS, potential vendor establishments must meet a series of financial and technical criteria (Agricultural Marketing Service, 2019b), the latter of which includes onsite audits and, when process deviations occur, developing and implementing corrective actions. Product acceptability and vendor performance is monitored through microbiological testing. Cooked diced chicken is tested for the presence of Salmonella and Listeria monocytogenes, and for levels of aerobic plate count (APC) organisms, coliforms, and generic Escherichia coli (GEC). Pasteurized egg products are tested for the presence of Salmonella and for levels of APC and coliforms. Product found positive for Salmonella or L. monocytogenes, or to exceed predefined critical limits (CLs) for APC, coliforms, or GEC (1000 colony-forming units [CFUs]/mL, 50 CFU/mL, and 10 CFU/mL, respectively) are rejected for purchase and diverted from federal nutrition assistance programs.

This report describes results of microbiological testing of cooked diced chicken, pasteurized liquid whole eggs, and pasteurized dried egg mix purchased by AMS from October 2012 through September 2018 for federal nutrition assistance programs. Potential revisions to AMS purchase specifications are discussed.

Materials and Methods

AMS vendors

For the period of this study—October 2012 through September 2018, three vendors produced cooked diced chicken under contract with AMS, six vendors produced pasteurized liquid whole egg mix under contract with AMS, and one vendor produced pasteurized dried egg mix under contract with AMS. All products were produced in the continental United States and distributed to federal nutrition assistance programs.

Sample collection

Samples of cooked diced chicken were collected by establishment employees under direct observation of AMS Graders. Three 0.50-lb samples from each lot were collected, frozen, and shipped on dry ice overnight to the AMS National Science Laboratory (NSL) in Gastonia, North Carolina or to an AMS-approved commercial laboratory. Samples of pasteurized egg products were collected by Food Safety and Inspection Service (FSIS) officials. For pasteurized liquid eggs and for pasteurized dried egg mix, six-ounce samples, using the lot size criteria described below, were collected and shipped on dry ice overnight to the AMS NSL.

Sampling frequency

From October 2012 through September 2017, a lot of cooked diced chicken was defined as the amount of finished product produced for AMS by one establishment in 1 d, from clean-up to clean-up. The average lot size, based on production worksheets, during this period was 20,938 lbs. Three samples were collected per lot. Beginning in October 2017, based on feedback from vendor establishments indicating that the risk of rejecting an entire lot because of a single positive sample might be disproportionately high in relation to the actual incidence of product contamination, AMS provided vendor establishments the option of defining a lot as either the amount produced for AMS during (i) a day or (ii) an hour (not to exceed 15,000 lbs per hour). Of the two vendor establishments that produced cooked diced chicken for AMS since that time, both opted to define a lot as 1 h's production. The average lot size, based on production worksheets, was 10,764 lbs. Three samples were collected per lot.

For pasteurized liquid whole eggs, a lot was defined as the amount of finished product produced for AMS by one establishment in 1 d. Based on production worksheets, the average lot size from October 2012 through September 2018 was 31,667lbs. The number of samples collected depended on the lot size. AMS procures pasteurized liquid whole eggs in containers that contain 6 5-lb cartons and in containers that contain twelve 2-lb cartons. Regardless of container configuration, 4 samples were taken for lots of 1200 or less containers, 8 samples were taken for lots of 1201–3200 containers, 12 samples were taken for lots of 3201–5200 containers, and 16 samples were taken for lots of 5201 or more containers.

For pasteurized dried egg mix, a lot was also defined as the amount of finished product produced for AMS by one establishment in 1 d. Based on production worksheets, the average lot size from October 2012 through September 2018 was 13,263 lbs. The number of samples collected depended on lot size and production unit. AMS procures pasteurized dried egg mix in 0.38-lb. pouches and in 10-lb. bags. For product procured in pouches, 8 samples were taken for lots of 1200 or fewer pouches, 12 samples were taken for lots of 1201 to 35,000 pouches, and 16 samples were taken for lots of 35,001 or more pouches. For product procured in bags, 8 samples were taken for lots of 45 or fewer bags, 12 samples were taken for lots of 46 to 13,000 bags, and 16 samples were taken from lots of 13,001 or more bags.

Sample processing

Samples were logged upon arrival using a laboratory identification number and tested for proper receipt temperature (0°C or less for cooked diced chicken, 0°C–4°C for pasteurized liquid egg, and 22°C or less for pasteurized dried egg mix). For cooked diced chicken, samples from each lot were thawed, composited, and tested qualitatively for Salmonella and L. monocytogenes and quantitatively for APCs, coliforms, and GEC E. coli. For pasteurized egg products, samples from each lot were thawed, composited, and tested qualitatively for Salmonella and quantitatively for APC and coliforms. Testing for Salmonella and for L. monocytogenes was done as per the FSIS Microbiology Laboratory Guidebook (Food Safety and Inspection Service, 2019). Samples were selectively enriched and screened for the presence of Salmonella and for Listeria using the 3M™ Molecular Detection System. Presumptive positive samples were further selectively enriched for culture confirmation. Culture-confirmed samples were reported as positive. Testing for APC and for GEC was done by plating onto 3M Petrifilm and using AOAC International Official Method (AOAC-OM) 990.12 (AOAC International, 2019a) and AOAC-OM 998.08 (AOAC International, 2019b), respectively.

Data analysis

Data were organized, and statistical analyses were conducted, in Microsoft^® Excel. The difference in proportions between each sequential pair of fiscal years for each pathogen and for each indicator microorganism was examined using the z-test for two independent proportions.

Results

Of 3668 lots of cooked diced chicken produced for federal nutrition assistance programs from October 2012 through September 2018, none was positive for Salmonella, 3 (0.08%) were positive for L. monocytogenes, 8 (0.22%) exceeded the APC CL, 15 (0.41%) exceeded the coliform CL, and 5 (0.14%) exceeded the generic E. coli CL (Table 1). The increase in the number of lots tested from October 2017 through September 2018 resulted from the policy change made at the beginning of the period, in which vendors could choose to define a lot as the amount of product produced for AMS (i) in 1 d or (ii) in 1 h: all lots produced during this period were defined as an hour of production. The two samples found positive for L. monocytogenes in 2014–2015 were collected from the same vendor on the same day in June.

Table 1.

Microbiological Testing Results of Cooked Diced Chicken Produced for Federal Nutrition Assistance Programs from October 2012 Through September 2018

Year (October–September)	No. lots	No. (%) L. monocytogenes-positive	No. (%) APC CL (1000 CFU/mL) exceeded	No. (%) coliform CL (50 CFU/mL) exceeded	No. (%) generic Escherichia coli CL (10 CFU/mL) exceeded
2012–2013	409	0	1 (0.24)	1 (0.24)	0
2013–2014	570	0	0	0	0
2014–2015	436	2 (0.46)^a	0	0	0
2015–2016	609	0^b	0^a	0	1 (0.16)
2016–2017	570	0	4 (0.70)^b	0^a	0
2017–2018	1074	1 (0.09)	3 (0.28)	14 (1.30)^b	4 (0.37)
Total	3668	3 (0.08)	8 (0.22)	15 (0.41)	5 (0.14)

Product found positive for Salmonella or Listeria monocytogenes or with CFU levels exceeding the CL for APC microorganisms, coliforms, or generic Escherichia coli were rejected for purchase.

a,b

Statistically significant changes between two values in the same column are indicated.

CFU, colony-forming unit; CL, critical limit; APC, aerobic plate count.

Of 984 lots of pasteurized liquid whole eggs produced for federal nutrition assistance programs from October 2012 through September 2018, 1 (0.10%) was positive for Salmonella, 29 (2.95%) exceeded the APC CL, and 4 (0.41) exceeded the coliform CL (Table 2). The one Salmonella-positive sample among those analyzed in this study was found in a lot of pasteurized liquid whole eggs produced in December 2015.

Table 2.

Microbiological Testing Results of Pasteurized Liquid Whole Eggs Produced for Federal Nutrition Assistance Programs from October 2012 Through September 2018

Year (October–September)	No. lots	No. (%) Salmonella-positive	No. (%) APC CL (1000 CFU/mL) exceeded	No. (%) coliform CL (50 CFU/mL) exceeded
2012–2013	23	0	2 (8.70)	0
2013–2014	99	0	7 (7.07)	0
2014–2015	88	0	7 (7.95)	1 (1.14)
2015–2016	137	1 (0.73)^a	4 (2.92)	2 (1.46)^a
2016–2017	387	0^b	9 (2.33)	0^b
2017–2018	250	0	0	1 (0.40)
Total	984	1 (0.10)	29 (2.95)	4 (0.41)

Product found positive for Salmonella or with CFU levels exceeding the CL for APC microorganisms or coliforms were rejected for purchase.

a,b

Statistically significant changes between two values in the same column are indicated.

Of 380 lots of pasteurized all-purpose egg mix produced for federal nutrition assistance programs from October 2012 through September 2018, none was positive for Salmonella, none exceeded the APC CL, and 3 (0.79%) exceeded the coliform CL (Table 3).

Table 3.

Microbiological Testing Results of All-Purpose Egg Mix Produced for Federal Nutrition Assistance Programs from October 2012 Through September 2018

Year (October–September)	No. lots	No. (%) coliform CL (50 CFU/mL) exceeded
2012–2013	33	0
2013–2014	39	0
2014–2015	42	0
2015–2016	126	1 (0.79)
2016–2017	73	0
2017–2018	67	2 (2.99)
Total	380	3 (0.79)

Product found positive for Salmonella or with CFU levels exceeding the CL for APC microorganisms or coliforms were rejected for purchase.

No sample found positive for Salmonella was also found positive for L. monocytogenes or for indicator CL exceeds. One of the samples found to exceed the CL for APC also exceeded the CL for coliforms. Four of the samples found to exceed the CL for coliforms also exceeded the CL for GEC.

For cooked diced chicken (Table 1), statistically significant differences were found comparing 2014–2015 to 2015–2016 for detection of L. monocytogenes (p < 0.05), comparing 2015–2016 to 2016–2017 for exceeding APC CLs (p < 0.05), and comparing 2016–2017 to 2017–2018 for exceeding coliform CLs (p < 0.05). For pasteurized liquid whole egg (Table 2), statistically significant results were found comparing 2015–2016 to 2016–2017 for detection of Salmonella (p < 0.05) and for exceeding coliform CLs (p < 0.05). No statistically significant differences were observed for all-purpose egg mix (Table 3) test results.

Discussion

Microbiological testing requirements used by AMS reflect stakeholder views that pathogen contamination of cooked diced chicken and pasteurized products procured for federal nutrition assistance programs is an unacceptable risk, especially considering the vulnerable populations often served through these programs. Although these products are generally regarded as microbiologically safe, there is a risk from postprocess contamination, especially given the likelihood that they will be consumed without an additional lethality step.

Salmonella have been shown to survive at −20°C for up to 16 weeks on precooked chicken products (Dominguez and Schaffner, 2009), and such products have been identified as the vehicle of infection in large outbreaks of salmonellosis, including an outbreak in Spain in 2005 that sickened over 2000 people (the largest known outbreak of foodborne illness in that country) (Lenglet and National Epidemiological Surveillance Network of Spain, 2005) and a 2012 outbreak in Finland in which 53 confirmed cases were attributed to cooked “chicken cubes” produced in China (Huusko et al., 2017). We are unaware of any outbreaks in the United States that have been definitively linked to cooked chicken products. Chicken salad was reported as a potential vehicle of infection in ∼14 confirmed outbreaks of salmonellosis in the United States from 1998 through 2018 (Centers for Disease Control and Prevention, 2019). The one outbreak for which additional detailed information is available suggest contamination was introduced from contaminated eggs and/or ill foodservice workers (Gicquelais et al., 2014).

None of the cooked diced chicken sampled by AMS tested positive for Salmonella. This observation comports with findings from the FSIS ALLRTE (random sampling) and RTE001 (risk-based sampling) regulatory sampling programs. Of ∼19,880 samples of products containing or potentially containing chicken (including 480 samples of ground chicken patties, >10,000 samples of “other chicken” items, and 9400 samples of multicomponent “chicken and/or pork” items) collected from calendar years 2005 to 2012, 12 (0.06%) were positive for Salmonella (Mamber et al., 2018).

In a retail market-basket survey across four United States states from December 2010 to March 2013, 15 (0.25%) of 5917 ready-to-eat meat and poultry samples tested positive for Listeria and Listeria-like organisms (Luchansky et al., 2017). The observation that 3 (0.08%) of the cooked diced chicken samples collected by AMS were positive for L. monocytogenes, although not directly comparable because taken at different points in the continuum, is generally consistent with those from the market-basket survey. Given the severity of infection from L. monocytogenes (Scallan et al., 2011), and because cooked diced chicken may be consumed without an additional lethality step, continued testing of for L. monocytogenes by AMS is warranted. AMS is unaware of any outbreaks of listeriosis attributed to cooked diced chicken, and through our partners at the Food and Nutrition Service AMS will continue to work with school foodservice agencies and other recipients to promote the use of best practices to mitigate the potential for cross-contamination of cooked diced chicken supplied to federal nutrition assistance programs. The observation that <1% of the cooked diced chicken samples collected in the study contained levels of APC, coliforms, or GEC above predefined CLs suggest that the vendor processes used to produce the product were well controlled and that proper sanitation was maintained. Nevertheless, the statistically significant increase in the number of samples exceeding APC CLs from 2015–2016 to 2016–2017 and in the number of samples exceeding coliform CLs from 2016–2017 to 2017–2018 indicates the need for continued close scrutiny of sanitary process control in production of cooked diced chicken for AMS.

Pasteurization is effective for reducing risk of human infection from Salmonella in egg products (Latimer et al., 2008). Nonetheless, some Salmonella, especially heat-resistant strains, have been isolated from pasteurized egg products in the United States (Gurtler et al., 2015). Whether these isolates survived pasteurization or resulted from postpasteurization contamination is unclear. The observation that 1 (0.10%) of the pasteurized liquid whole egg samples and none of the pasteurized egg mix samples tested by AMS were positive for Salmonella is in line with data from FSIS, in which 41 (0.13%) of 30,696 samples of pasteurized liquid, frozen, and dried egg products collected from calendar years 2008 through 2017 were found positive for Salmonella (Food Safety and Inspection Service, 2018).

The proportion of pasteurized liquid whole egg samples tested by AMS and found to contain APC levels exceeding the predefined CL trended downward over the period of this study, from 8.7% in 2012–2013 to 0% in 2017–2018. It is not clear why APCs were present in these pasteurized products at levels exceeding the CLs, but we believe it is likely due to postpasteurization contamination and/or improper storage. The downward trend may be due partially to improved sanitation brought about by revisions to AMS purchase specifications in 2017 requiring vendors to develop and submit cause-and-effect analyses and implement corrective actions in the event of elevated indicator organism levels specifically, and it may be due partially to implementation of various sanitation and food safety practices, including indicator organism testing, in the egg products industry more generally (Viator et al., 2016). This cannot be demonstrated empirically, however.

Conclusions

Data from this study suggest vendor establishments that produce cooked diced chicken, pasteurized liquid whole eggs, and pasteurized dried egg mix for federal nutrition assistance programs do so under effective food safety control systems. However, there are limitations to the data, including: the relatively low frequency of sampling and the likelihood that low levels of pathogen contamination or indicator organisms exceeding CLs were not detected in same samples; that samples were taken at a discreet point and thus not reflective of the overall production system; and that information to characterize isolates, including serotype, was not gathered.

Based on observations from this study, AMS purchase specifications for cooked diced chicken and pasteurized egg products will be updated so that Salmonella and Listeria isolates recovered through testing are further characterized for serotype, antimicrobial susceptibility, and whole-genome sequence; and so that sampling for pathogens and for indicator microorganisms is required immediately before product shipment. These updates will provide additional information to devise mitigations to control pathogen occurrence in and strengthen overall sanitation of cooked diced chicken and pasteurized egg products produced for AMS.

AMS will continue to work closely with vendor establishments, and with our sister agency, FSIS, to help ensure the safety of cooked diced chicken and pasteurized egg products procured for federal nutrition assistance programs.

Footnotes

Acknowledgments

The article was prepared in memory of Chandramohan Chawan, who was instrumental in developing specifications and facilitating purchase of poultry and egg items for federal nutrition assistance programs, and of Jackie Glover, who oversaw most of the microbiological testing described herein. The authors thank staff at the AMS National Science Laboratory, Gastonia, North Carolina, for processing and analyzing microbiological samples.

Disclosure Statement

No competing financial interests exist.

References

Agricultural Marketing Service (AMS). Food Commodity Purchasing. 2019a. Available at: https://ams.usda.gov/reports/food-commodity-purchasing, accessed February 1, 2019 .

Agricultural Marketing Service (AMS). Become a USDA Foods Vendor. 2019b. Available at: https://ams.usda.gov/selling-food/becoming-approved, accessed February 1, 2019 .

AOAC International (AOAC). Aerobic Plate Count in Foods. 2019a. Available at: http://eoma.aoac.org/methods/info.asp?ID=46762, accessed February 1, 2019 .

AOAC International (AOAC). Coliform and Escherichia coli Counts in Foods. 2019b. Available at: http://eoma.aoac.org/methods/info.asp?ID=46949, accessed February 1, 2019 .

Centers for Disease Control and Prevention (CDC). National Outbreak Reporting System (NORS). 2019. Available at: https://wwwn.cdc.gov/norsdashboard, accessed May 24, 2019 .

Dominguez

, Schaffner

. Survival of Salmonella in processed chicken products during frozen storage. J Food Prot, 2009; 72:2088–2092.

Food Safety and Inspection Service (FSIS). FSIS Microbiological Testing Program for Pasteurized Egg Products, Calendar Years 1995–2017. 2018. Available at: https://fsis.usda.gov/wps/portal/fsis/topics/data-collection-and-reports/microbiology/testing-program-for-rte-meat-and-poultry-products/microbiological-testing-program-for-pasteurized-egg-products/pasteurized-egg-products, accessed February 1, 2019 .

Food Safety and Inspection Service (FSIS). Microbiology Laboratory Guidebook. 2019. Available at: https://fsis.usda.gov/wps/portal/fsis/topics/science/laboratories-and-procedures/guidebooks-and-methods/microbiology-laboratory-guidebook/microbiology-laboratory-guidebook, accessed February 1, 2019 .

Gicquelais

, Morris

, Matthews

, Gladden

, Safi

, Grayson

, Slayton

, Newton

, Bordonaro

, Wheeler

, Smith

, Bosch

, Haselow

. Multiple-serotype Salmonella outbreaks in two state prisons—Arkansas, August 2012. MMWR Morb Mortal Wkly Rep, 2014; 63:169–173.

10.

Gurtler

, Hinton

Jr. , Bailey

, Cray

Jr ., Meinersmann

, Ball

, Jin

TZ.

Salmonella isolated from ready-to-eat pasteurized egg products: Thermal resistance, biochemical profile, and fatty acid analysis. Int J Food Microbiol, 2015; 206:109–117.

11.

Huusko

, Pihlajasaari

, Salmenlinna

, Sõgel

, Dontšenko

, De Pinna

, Lundström

, Toikkanen

, Rimhanen-Finne

. Outbreak of Salmonella enteritidis phage type 1B associated with frozen pre-cooked chicken cubes, Finland 2012. Epidemiol Infect, 2017; 145:2727–2734.

12.

Latimer

, Marks

, Coleman

, Schlosser

, Golden

, Ebel

, Kause

, Schroeder

. Evaluating the effectiveness of pasteurization for reducing human illnesses from Salmonella spp. In egg products: Results of a quantitative risk assessment. Foodborne Pathog Dis, 2008; 5:59–68.

13.

Lenglet

, National Epidemiological Surveillance Network of Spain. E-alert 9 August: Over 2000 cases so far in Salmonella Hadar outbreak in Spain associated with consumption of pre-cooked chicken, July–August, 2005. Euro Surveill, 2005; 10:E050811.1.

14.

Luchansky

, Chen

, Porto-Fett

ACS

, Pouillot

, Shoyer

, Johnson-DeRycke

, Eblen

, Hoelzer

, Shaw

Jr ., Van Doren

, Catlin

, Lee

, Tikekar

, Gallagher

, Lindsay

, The Listeria Market Basket Survey Multi-Institutional Team, Dennis

Survey for Listeria monocytogenes in and on ready-to-eat foods from retail establishments in the United States (2010 through 2013): Assessing potential changes of pathogen prevalence and levels in a decade. J Food Prot, 2017; 80:903–921.

15.

Mamber

, Mohr

, Leathers

, Mbandi

, Bronstein

, Barlow

. Occurrence of Salmonella in ready-to-eat meat and poultry product samples from U.S. Department of Agriculture-regulated producing establishments. I. Results from the ALLRTE and RTE001 random and risk-based sampling projects, from 2005 to 2012. J Food Prot, 2018; 81:1729–1736.

16.

Scallan

, Hoekstra

, Angulo

, Tauxe

, Widdowson

, Roy

, Jones

, Griffin

. Foodborne illness acquired in the United States—major pathogens. Emerg Infect Dis, 2011; 17:7–15.

17.

Viator

, Cates

, Karns

, Muth

, Noyes

. Food safety practices in the egg products industry. J Food Prot, 2016; 79:1210–1215.

Microbiological Testing Results of Cooked Diced Chicken and Pasteurized Egg Products Purchased for United States Federal Nutrition Assistance Programs,2012–2018