Illnesses | elmsprogressivemedia

Two types of bacteria, Salmonella and Campylobacter, cause 80% of the illnesses and 75% of the deaths associated with meat and poultry products. They are estimated to cost consumers nearly $3.5 billion each year in the form of lost wages and medical expenses. Consumers also suffer illnesses that range from mild to severe and the loss of loved ones.

The foods most commonly contaminated with Salmonella and Campylobacter are poultry products, especially chicken. Estimates of contamination range from 15% to 62%, depending on the bacteria and the poultry involved.

Despite increasing rates of food poisoning from Salmonella and Campylobacter during the 1980s, and continuing high levels today, the poultry industry has maintained processing practices that actually increase the percent of contaminated products. Instead of minimizing the contamination in processing plants, the poultry industry relies on consumers to cook the problem away.

Government regulations have contributed to the problem as well. The United States Department of Agriculture (USDA), which oversees the production of 7 billion chickens and 277 million turkeys each year, allows several practices that contribute to high contamination rates in poultry:

USDA allows poultry carcasses to be placed in hot water baths, called scalders, to loosen feathers and then chilled in large vats of water, a process known as immersion chilling. This allows contaminated poultry carcasses to infect clean carcasses. Because Salmonella populations can grow from a few bacteria into a health threat, contamination of clean carcasses with even small numbers of harmful bacteria must be prevented.

Based on a policy change in 1978 allowing “reprocessing,” the USDA decided that, instead of condemning contaminated carcasses, the industry can simply wash the contamination off with chlorinated water. This resulted in condemnation rates for poultry dropping dramatically, which means that consumers are eating more of this once-contaminated poultry. While some plants can reprocess poultry and produce a product at least equivalent to one that wasn’t initially contaminated, other plants show large jumps in the percent of reprocessed carcasses that are contaminated with Salmonella.

USDA allows skin to be added to ground poultry products. Since skin harbors bacteria in its pores and folds, it can be the most highly contaminated part of the carcass. By permitting the addition of skin into ground product, USDA is allowing much higher levels of certain harmful bacteria in ground poultry products than in ground beef products, which are not allowed to contain skin.

Ten years after it was first recommended by the National Academy of Science, USDA has proposed to overhaul the meat and poultry inspection system by requiring processing controls called “Hazard Analysis and Critical Control Points” or HACCP. This is a very promising system that focuses on preventing food from becoming contaminated with foodborne hazards.

The Clinton Administration coupled the HACCP concept with a mandatory industry testing program and performance standards for Salmonella. Together, this system represents the single most important development in the regulation of food safety in nearly 40 years, one that is long overdue.

However, the meat and poultry industries may have weakened the rule by persuading USDA to change the requirements from mandatory testing for Salmonella to testing for generic E. coli, which is not associated with food borne illness. This change will reduce the HACCP system’s effectiveness in controlling harmful bacteria on poultry products.

Conclusions and Key Recommendations

Consumers should know that raw poultry coming into their kitchens are contaminated with harmful bacteria, like Salmonella or Campylobacter. Government officials recommend that consumers should treat all raw poultry as if it were contaminated. These experts know that the incidence of illness from Salmonella and Campylobacter is too high, so they urge everyone to take precautions. But consumers should be buying much cleaner products.

Adoption of the following recommendations is essential to clean up the safety of poultry products:

USDA should set performance standards for poultry products to force the poultry industry to reduce or eliminate Salmonella and Campylobacter in its products.

For HACCP systems to be effective for poultry processing plants, USDA should mandate that E. coli testing be combined with ongoing industry testing for Salmonella and Campylobacter.

E. coli testing alone will not ensure control of cross-contamination in poultry plants.

If daily pathogen testing by the meat and poultry industry is not mandated, USDA should require that companies regularly validate their HACCP plans to assure they control both the level and the spread of harmful bacteria and the government should conduct extensive and on-going pathogen monitoring in all plants.

Reprocessing should only be permitted in plants that can demonstrate that they produce products with Salmonella and Campylobacter outcomes better than or equivalent to their ordinary product. Plants that cannot demonstrate an equivalent product should be required to condemn contaminated carcasses.

The poultry industry should utilize alternative chilling technologies, such as dry chilling or blast chilling, that eliminate problems with cross contamination. If shown to produce less contaminated poultry products, new plants should be required to install these technologies.

Where immersion chilling is used, the poultry industry should develop and use better systems to prevent contamination of its products with harmful bacteria. For example, antimicrobial treatments should be mandated for use in conjunction with immersion chill tanks to prevent or counteract the cross-contamination effects of immersion chilling, but only after these treatments have been shown to be the safest and most effective. End-product treatments like steam pasteurization or in-plant cooking could be used as an alternative to antimicrobial treatments. New systems, such as bagging poultry prior to entry into the immersion chiller, should be tested and perfected.

USDA should require the removal of skin prior to grinding poultry products to help reduce the prevalence of pathogens found in ground products.

Introduction

Food safety is a fundamental concern for consumers. In the aftermath of the 1993 E. coli 0157:H7 outbreak from contaminated hamburgers served by the Jack-in-the-Box fast food chain, opinion polls have indicated that consumers fear that rules governing food safety are not tough enough. This report will show that for many raw products, particularly poultry products like chicken, consumers’ concerns are warranted and reforms are urgently needed.

Despite what American consumers have come to expect, contaminated chicken is not inevitable. In fact, Sweden has largely eradicated Salmonella contamination in chicken. Although the federal government oversees and regulates poultry slaughterhouses, for many years it has tolerated widespread contamination in poultry products produced in the United States.

Two types of bacteria are estimated to cause up to 6.5 million illnesses each year. These bacteria are Salmonella and Campylobacter, and if they aren’t household words already, they should be. Salmonella and Campylobacter cause the symptoms that are traditionally associated with mild cases of food poisoning: nausea, vomiting, diarrhea, fever and chills. But these harmful bacteria can occasionally cause more serious conditions, like dehydration, meningitis, kidney or heart disease, certain kinds of arthritis, septicemia, hemolytic-uremic syndrome and even death.

USDA has estimated that these bacteria on meat and poultry may together cause over 4 million illnesses and up to 3,000 deaths each year. That translates into almost 11,000 illnesses and eight deaths each day.

Children and adults over 60 years of age are the groups at greatest risk of becoming seriously ill or dying from eating tainted meat and poultry. For Campylobacter, illnesses also commonly occur among adults between 20 and 30 years of age. People with other health problems, such as AIDS, liver disease, sickle cell anemia, or cancer, are at heightened risk of debilitating illness or death.

Consumers pay a high economic cost for these illnesses in the form of medical expenses and lost wages. The total annual estimated costs just for Salmonella and Campylobacter food poisoning illnesses from meat and poultry products is as high as $3.4 billion, according to USDA (see Table I). Clearly, preventing meat and poultry products from becoming contaminated with harmful bacteria like Salmonella or Campylobacter would result in improved health for consumers, as well as significant economic benefits.

Over the past decade, numerous government studies, reports of the National Academy of Sciences and media exposès have documented that meat and poultry products contaminated with disease-causing bacteria like Salmonella and Campylobacter are stamped “Wholesome” or “Inspected” by the federal government and sold to American consumers. Ten years after the first National Academy of Science’s report called for controls to prevent contamination of meat and poultry with harmful bacteria, the Clinton Administration took the first steps to mandate controls for the meat and poultry industry by proposing a new system called HACCP, for Hazard Analysis and Critical Control Points. HACCP is a very promising system that focuses on preventing food from becoming contaminated with foodborne hazards.

The Clinton Administration proposed to couple the HACCP concept with a mandatory industry testing program and performance standards for Salmonella. Together, this system represents the single most important development in the regulation of food safety in nearly 40 years.

The meat and poultry industries have attempted to stop this rule, and when unsuccessful, to weaken some elements of the proposal. This controversy has both delayed final implementation of the rule, which is expected out this spring, and has put significant pressure on the Department to soften the more controversial and important aspects of the regulation.

How Contaminated is Poultry?

Salmonella contamination is more common in broiler chickens than in any other meat product. In February 1995, the USDA’s Food Safety and Inspection Service published a rule designed to improve the safety of meat and poultry products. At that

Table I. Estimated medical costs and productivity losses for selected human pathogens (1993).

Pathogen	Meat/poultry cases per year	Meat/poultry deaths per year	Total costs to the public ($ in billions)
Campylobacter	1,031,250 – 1,312,500	83-383	0.5 – 0.8
Salmonella	348,000 – 2,880,000	348 – 2,610	0.3 – 2.6
Other pathogens	768,263 – 774,384	964 – 1,198	1.0 – 1.4
Parasites	2,056	41	2.7
Totals	2,149,569 – 4,968,940	1,436-4,232	4.5-7.5

Source: USDA, “Pathogen Reduction; Hazard Analysis and Critical Control Point (HACCP) Systems; Proposed Rule, 9 CFR Part 308, et al.,” Federal Register, Vol. 60, No. 23 (1995), p. 6781.

Table II. USDA national baseline levels for Salmonella.

Commodity	Frequency of occurrence of Salmonella (% positive)
Steers/Heifers	1
Broilers	25
Raw Ground Beef	4
Fresh Pork Sausages	12
Cows/Bulls	1
Hogs	18
Turkeys	15

Source: USDA, “Pathogen Reduction; Hazard Analysis and Critical

Control Point (HACCP) Systems; Proposed Rule, 9 CFR Part 308, et al.,”

Federal Register, Vol. 60, No. 23 (1995), p. 6800.

time, the agency reported that 25% percent of broiler chickens were contaminated with Salmonella. By way of comparison, the frequency of contamination of ground beef was 4% (see Table II).

Many other surveys have documented an even higher incidence of Salmonella and Campylobacter in poultry products. A recent report contained a summary of 55 different studies and reported that the median contamination rate documented by these studies was 30% for Salmonella and 62% for Campylobacter jejuni.

Late in 1995, USDA released a survey of ground products, including beef, turkey and chicken. Although the chicken and turkey results are considered interim due to the modest number of samples, the findings, shown in Table III, are still noteworthy.

Table III. Comparison of plant prevalence for ground beef, ground turkey, and ground chicken in USDA Microbial Survey Results (1995).

	Beef *	Turkey **	Chicken ***
Campylobacter rates	0.2	21.6	60.0
Salmonella rates	4.3	27.0	38.2

* Based on 563 samples

** Based on 165 samples

*** Based on 162 samples

Source: USDA Microbiological Survey Results (1995).

“Just Cook It”

Industry groups, such as the National Broiler Council, claim that the solution to foodborne illness from poultry products rests with consumers. “Just cook it,” they say, to eliminate the risk and the problem. The truth is that for the past few decades the poultry industry, with the tacit approval of the government, has been using processing practices that actually increase the percent of its products contaminated with harmful pathogens, like Salmonella and Campylobacter. The poultry industry relies on consumers to cook the harmful bacteria away.

That strategy has not been effective from a public health standpoint (see Table IV). According to CDC data, cases of Salmonella increased throughout the 1980s and have leveled off at a new and very high level. Campylobacter has maintained consistently high levels since surveillance was started in the 1980s. These two pathogens alone cause 80% of the illness and 75% of the deaths associated with meat and poultry products, according to USDA (see Table I). Until 1994, Salmonella typhimirium, a strain associated with meat and poultry products, was the leading identified cause of illness from Salmonella, although it was recently surpassed by the Salmonella strain associated with egg products.

Table IV. CDC summary of reported cases of Salmonellosis,* 1955-1994.

Excluding typhoid fever.

Source: Center for Disease Control and Prevention (CDC), “Summary of Notifiable Diseases, United States, 1994,” Morbidity and Mortality Weekly Report, Vol. 43, No. 53 ( October 6, 1995), p. 51.

Note: 1985 outbreaks includes one major outbreak caused by contaminated pasteurized milk in Illinois.

Chicken processors take live chickens, some of which may be contaminated with Salmonella and Campylobacter, process them using systems that are likely to spread the contamination from bird-to-bird, and then send them to the market for sale to consumers. Chicken carcasses are processed using hot water scald tanks, defeathering equipment, automatic evisceration machines, and immersion chill tanks. Throughout this process, the carcasses can gain up to 8% additional weight in water, as well as picking up harmful bacteria.

Poultry is frequently frozen by processors to extend its shelf life (although it is almost always marked “fresh” on the label) and then thawed out in supermarkets. The high water content of the poultry products due to immersion chilling combined with rapid chilling at the processor results in high water retention. During thawing, this liquid (which can appear either red or clear) collects in chicken packages. Once in the kitchen, it is easy for bacteria to spread from raw poultry or poultry juices onto kitchen surfaces and other food products (cross-contamination). Even hands can spread contamination.

In the summer of 1994, Poultry Processing magazine reported that the Kroger consumer help line handled more calls on “bloody chicken” over a fourteen-month period than on any other single problem. Leaking packages was another concern cited by consumers to the Kroger company. USDA’s Food Safety Hotline also reports that they get calls from consumers complaining about too much liquid in poultry packages.

How Current Processing Spreads Salmonella and Campylobacter

Numerous studies have documented that the percentage of carcasses contaminated with Salmonella and Campylobacter increases during processing. One study of Campylobacter in four plants found that 20% of birds tested positive when they entered the plant, 52% tested positive following immersion chilling, and 31.6% tested positive at retail outlets.

Other studies have documented that the percent of carcasses contaminated with Salmonella increases during processing. In a series of studies conducted for USDA, both the baseline data and data on improvements to the processing practices documented that the extent of cross- contamination that occurs in poultry processing plants is dramatic, even after improvements are implemented. The studies documented that 58% of the chicken was contaminated with Salmonella prior to having their intestines and other internal organs removed (a process known as “evisceration”). Following evisceration and chilling in tanks of cold water, the number of chickens contaminated with Salmonella jumped to 72%. The study also documents that even when improvements were made in the scalder (common bath using hot water which is used to loosen feathers) and in the immersion chill tank, the percent of carcasses contaminated with Salmonella were still increased (see Table V).

Table V. Cross-contamination during processing: Percent of carcasses positive for Salmonella.

Sampling Point	Baseline	Processing modification (Scalder)	Processing modification (chiller)
Pre-evisceration	58	24	33
Postchill	72	49	46

* Scalder water counterflow and post scald rinse prior to picking.

** Chlorination of chiller water.

Source: William James et al., “Cost Effective Techniques to Control Human Enteropathogens on Fresh Poultry,” Poultry Science, Vol. 72 (1993), pp. 1174-1176. Includes data from three separate studies.

Representatives of the poultry industry claim that the percentage of birds contaminated with Salmonella is not important because the actual number of the harmful bacteria found on the birds is very low and some scientists agree. In remarks given at a USDA-sponsored scientific conference, Amy Waldroup, a researcher at the University of Arkansas, said: “Researchers agree that this organism is likely present on 30 to 50% of all poultry carcasses, but that the level present is extremely low.”

Waldroup reviewed numerous studies and said the numbers of Salmonella on poultry carcasses range from 1 to 30 organisms. Even assuming this were true, Salmonella can grow very easily on poultry carcasses. With temperature abuse, a few bacteria may be able to grow into a public health problem. Waldroup also reported that Campylobacter is present, possibly at very high levels, on raw poultry produced in the U.S.

The dose of Salmonella necessary to make a healthy person ill is estimated at around 10,000 bacteria. However, the dose is dependent on the health of the consumer, and some outbreaks have been reported from food containing less than 100 Salmonella per gram.

The public health data continue to show high rates of illness from Salmonella and Campylobacter, and poultry products are considered the major source. Controlling cross-contamination in poultry plants is clearly a challenge for the poultry industry, but one that is key to addressing the public health problems with the products.

A brief review of the processing procedure documents how cross-contamination occurs in the processing plant (see Figure I). Chickens and turkeys are shipped in cages crowded with other birds and often arrive at the processing plant with feces on the feathers and the skin. At the processing plant, the birds are hung by their feet, and then stunned. From this point on they are essentially brain dead, but other physiological responses continue. The birds often defecate. Hanging birds by their feet assures that when they defecate, the contamination remains on the skin and feathers.

Figure I. The poultry slaughter process.

Live Birds (then) Enter Plant (then) Antemortem Inspection (then) Stun (then) Scald (then) Defeather (then) Evisceration (then) Postmortem inspection (then) Wash (then) Chill (then) Reinspection of sample of carcasses (and finally) Broilers/Fryers

After their blood is drained, the carcasses enter huge tanks for scalding. All the contamination present on the skin and feathers is now put in the scald tank, a warm communal bath also called “fecal soup.” Although the water is hot enough to loosen the feathers, it is often not hot enough to kill the bacteria, and contamination builds up in the tanks as the processing continues. In addition, the carcasses pick up water which helps the bacteria to firmly attach to the skin. According to one scientific review: “The scalding process opens feather follicles to aid feather removal, and the follicles remain open throughout processing until chilling when they close, thereby retaining microorganisms.”

The next stage is defeathering. This is usually done mechanically, with machines that use little rubber fingers to loosen and remove the feathers. This is a major source of contamination. The mechanical fingers are not cleaned between each bird and can actually collect contamination from the dirtiest bird and redistribute it onto each new carcass. Also, the rubber fingers actually beat contamination into the pores, crevices, and folds of the poultry skin.

Increasingly, removal of the intestinal tract and organs, called evisceration, is done by machines. This mechanized system cannot ensure that the intestinal contents are prevented from spilling all over the cavity of the bird. Whether evisceration is done mechanically or by hand, this is a key point where cross-contamination can occur. One study of evisceration using tracer bacteria on one bird showed that the next 42 birds were contaminated with the tracer bacteria and that there was sporadic contamination up to the 150th bird.

Washing is done at various stages of the process to remove contamination. Although this can remove a lot of contamination on the surface of the bird, it does not reach the contamination trapped in pores, crevices, or folds of the poultry skin.

Finally, birds are chilled in large vats of water called immersion chillers, a common bath where the birds bump against each other. Salmonella and Campylobacter get redistributed from one carcass onto others in the tank. Chlorine is often used to minimize cross-contamination, but there are questions both as to its effectiveness and to the possible chemical residues that it leaves.

Spread of harmful bacteria from carcass to carcass clearly can occur at many points in the poultry slaughter process, starting with the live animal and continuing with many pieces of equipment currently in use in poultry plants: the scalder or communal bath, the defeathering equipment, the evisceration equipment, and finally the immersion chillers. It is imperative that new systems be developed and implemented to minimize or eliminate the contamination that occurs at these points.

Do Government Regulations Promote Dirty Birds?

The federal government oversees and inspects the over 7 billion chickens and 277 million turkeys produced each year in the United States. Federal inspection of poultry has been ongoing on a voluntary basis since 1926. The inspection program was developed primarily to promote the sale of poultry products. In 1957, poultry inspection became mandatory with the passage of The Poultry Products Inspection Act of 1957 (PPIA) for interstate processors of poultry products. This Act was passed in large part because the poultry industry saw that mandatory inspection would benefit the industry by allowing processors to ship products to states and localities that required federal certification. According to the National Academy of Sciences’ 1987 report, Poultry Inspection: The Basis for a Risk-Assessment Approach,

The responsibility for implementing PPIA [initially] remained with the USDA’s Agricultural Marketing Service (AMS), which had administered the voluntary poultry inspection program in effect before the Act was passed. AMS was strongly oriented toward facilitating the industry’s ability to market agricultural commodities. (Italics added.)

In 1968, the law was amended to bring nearly all poultry products under the federal inspection mandate.

Under the PPIA, federal inspectors examine birds prior to slaughter; inspect each bird carcass after slaughter and before processing; inspect plant facilities to ensure sanitary conditions; inspect all slaughter and processing operations; verify the truthfulness and accuracy of product labeling; and inspect imported poultry products at the point of entry. Under the PPIA, USDA conducts at least eight public health related-inspection activities, including inspection of live birds (antemortem inspection); inspection after slaughter (postmortem inspection); condemnation and final disposition; sanitary slaughter and dressing; poultry chilling; plant sanitation; carcass reinspection; and residue monitoring.

During a carcass-by-carcass examination, federal inspectors must designate whether birds unfit for human consumption are reprocessed or condemned. The National Academy of Sciences observed that federal inspectors have only two to three seconds to examine each bird and to decide its disposition.

In June 1993, an expert panel of scientists at Research Triangle Institute in North Carolina compared the processing of meat and poultry products and identified a number of significant regulatory differences that have a public health impact. Most significant is the use of water in poultry processing, both to process the birds, like scalding and chilling, and to wash off contamination that gets on the birds during processing, known as “reprocessing.”

A. Reprocessing

Poultry carcasses can be reprocessed by washing areas contaminated with fecal matter or ingesta with chlorinated water. Contaminated beef and pork, on the other hand, must be trimmed to remove contamination. The Research Triangle Institute report says that the basis for the difference is in the Poultry Products Improvement Act, which specifically allows “reprocessing” of poultry products. The definition of reprocessing is something determined by USDA and has changed over the years.

USDA currently defines “reprocessing” as:

Poultry reprocessing includes methods by which adulterated poultry products can be made not adulterated. All visible specks of contamination must be removed. These methods are to be used under the supervision of an inspector. Acceptable methods to remove visible contamination from poultry include various combinations of trimming, vacuuming, and washing. Cut contaminated surfaces will be removed only by trimming. If the inner surfaces are reprocessed other than by trimming, all surfaces of the carcasses are required to be treated with 20 ppm chlorinated water.

Prior to 1978, poultry contaminated with feces in the internal cavity was condemned. In 1978, the USDA allowed poultry processors to wash off internal fecal contamination. According to a series on poultry processing that ran in 1987 in the Des Moines Register:

The Agriculture Department is condemning less than one-half as many chickens and turkeys for fecal and other contamination in the nation’s processing plants as it did in the early 1970s, even though production increased by 57%.

Table VI. Rates of poultry condemned for contamination, 1973-1986 (selected years).

	Poultry Production	Condemned product	Rate of condemnation
1973	3.2 billion	8.3 million	.26%
1978	3.9 million	5.7 million	.15%
1986	5 billion	3.5 million	.07%

* Condemnation based on contamination

Source: George Anthan, “Washing Lets More Poultry Pass Muster: Rejection Rate Plummets Despite Rise in Production,” Des Moines Register, December 5, 1987, p. 1.

The big drop in contamination-related condemnations by USDA officials came after 1978, when the Department began allowing poultry firms to wash fecal contamination off poultry carcasses. Prior to that, the department had required fecally contaminated poultry to be condemned.

The Des Moines Register provided the figures in Table VI on rates of condemnation of poultry based on contamination both before and after the policy change was made. The change lead to a greater than three-fold decrease in the percentage of poultry that was condemned.

Since 1981, the rate of condemnation due to contamination has varied from .06% to .09%, while poultry production has grown to 7.5 billion carcasses per year.

To reprocess poultry, dirty carcasses are pulled off a fast-moving conveyor belt after examination by federal inspectors who evaluate 30 or more birds per minute. A plant employee sprays the carcasses inside and out with water and removes some of the internal organs. Birds are placed in bins and are sometimes reinspected by federal inspectors before they are rehung on the conveyor belt.

By contrast, meat products are required to have all visible fecal, ingesta, or milk contamination removed by trimming away the contaminated areas followed by federal reinspection of every carcass. Washing beef and pork carcasses raised concerns that, although some contamination may be removed by washing, some of it will certainly spread and become trapped in crevices and cracks in the meat. USDA recognized and reaffirmed the importance of trimming to assure removal of all visible contamination on meat products following the 1993 Jack-in-the-Box outbreak from E. coli 0157:H7.

USDA has not reconsidered its position on reprocessing of poultry products since 1978. In support of its position on poultry reprocessing, the USDA relies on two scientific studies, conducted 18 years apart by the same researcher, L.C. Blankenship, who found that there was no significant difference in bacterial levels on uncontaminated and reprocessed poultry products.

At the time the 1975 study was conducted, USDA required that birds contaminated in their inside cavity be condemned and removed from the line. Washing contamination from external intact skin surfaces was allowed, according to the author. The study compared the numbers of total bacteria, and several specific types of bacteria not known to cause illness in humans, and found that washing poultry reduced the levels of these bacteria on condemned product. In some instances, the amount of the reduction on washed (i.e. reprocessed) poultry did not bring the levels to or below the levels found on poultry that passed inspection. The study also looked for Salmonella, which it found at only very low levels.

Primarily on the basis of that study, USDA changed its policy to permit washing of carcasses that were internally contaminated.

In 1993, the same researcher published a study which documented a higher incidence of Salmonella contamination levels for reprocessed poultry carcasses than for conventionally processed carcasses, but says the difference, just over 5% of the 745 birds tested, was not statistically significant. This study concludes that reprocessing is effective in “maintaining the safety and wholesomeness of broiler products.” Table VII provides a breakdown of the incidence of Salmonella documented in this study.

Table VII. Salmonella incidence for reprocessed birds and conventionally processed birds.

Plant

Processing

Number of Birds

% positive for Salmonella

Conventional

Re-processed

150

31.3

53.3

Conventional

Re-processed

150

76.6

69.3

Conventional

Re-processed

150

70.6

74.0

Conventional

Re-processed

145

24.1

26.2

Conventional

Re-processed

150

70.6

76.6

Overall

Conventional

Re-processed

745

54.6

59.8

Source: L. C. Blankenship, “Broiler Carcass Reprocessing, A Further Evaluation,” Journal of Food Protection, Vol. 56, No. 11 (1993), p. 984.

This study shows that among the five plants tested, there was incredible variability in the results. The overall incidence of Salmonella ranged from 24% to 76%. One plant (A) had reprocessed birds with over 20% higher incidence of Salmonella contamination compared with its conventionally processed birds. For that plant, reprocessing was not equivalent to conventional processing and should be discontinued. Yet USDA used this study to justify the continuation of the reprocessing policy without ever addressing the variability that was revealed by the study.

The variability between plants found in the 1993 study was further confirmed by another study done by the University of Arkansas. This study found that one (Plant D) out of five plants included in the study had a significantly higher percentage of carcasses contaminated with Salmonella on reprocessed poultry when compared to inspection passed-carcasses (see Table VIII). Two of the plants (Plants D and E) also showed significant increases in the levels of Salmonella found on reprocessed carcasses when compared to inspection-passed carcasses. The study also documented very high Campylobacter contamination rates and levels in all the plants involved in the study.

Table VIII. Levels and prevalence of various bacteria on conventionally processed and reprocessed commercial broiler carcasses.

Plant

Sample

SAL

%SAL

CAMP

%CAMP

Conv.

Reproc.

100

0.04

0.05

4.17

3.83

Conv.

Reproc.

100

0.23

0.12

3.53

3.10

Conv.

Reproc.

100

0.13

0.14

3.94

3.62

Conv.

Reproc.

100

0.04

0.25

4.38

4.18

Conv.

Reproc.

100

0.31

0.46

3.79

4.08

a SAL = Salmonella (all levels are 0.05).

b CAMP = Campylobacter jejuni/coli (all levels are 0.14).

c-e Bacterial means for each microbial assessment within columns are significantly different (cP .05, dP .01, eP .001, f .0001) unless designated as not significant (ns).

Source: A.L. Waldroup et al., “Effects of Reprocessing on Microbiological Quality of Commercial Prechill Broiler Carcasses,” Journal of Applied Poultry Research, Vol. 2 (1993), pp. 113-114.

These studies show that reprocessing is a highly variable process, one which does not ensure that products are equivalent to those which pass the conventional inspection process. USDA’s uniform policy allowing reprocessing to treat internally contaminated poultry products should be revised. Reprocessing should only be used by plants that can demonstrate they can reprocess poultry using techniques ensuring that the products are better than or equivalent to inspection-passed products, measured against pathogen standards.

B. Immersion Chilling

USDA permits poultry chilling by totally immersing the poultry carcass in cold water, a practice often done using water with chlorine added to it. By contrast, beef and pork can only be chilled using cold air, together with spray chilling. The Research Triangle Institute report says that the basis for this difference is traditional industry practice. However, many European and Canadian producers use air chilling rather than water chilling to produce their poultry.

Immersion chilling has often been associated with cross-contamination of poultry products. One 1990 study documented that the incidence of Salmonella on carcasses increased 15% in one plant and 28% in a second plant from the pre-chill to the post-chill sampling points. The authors concluded: “This increase in incidence indicates that non-contaminated carcasses acquired Salmonella cells either by direct contact with contaminated carcasses in the chill tank or via the water.” They also examined two other microbial measures, total plate count and Enterobacteriaceae, and found that the levels of bacterial contamination generally declined, even while the percent of poultry carcasses contaminated with Salmonella increased (see Table XII on p. 20).

USDA takes the position that although immersion chilling can cause cross- contamination, using elevated levels of chlorine in the chill water minimizes the problem. However, USDA has never mandated the use of chlorine in poultry chiller waters, so the agency allows the industry to operate as it pleases.

The use of chlorine may lower the overall level of bacteria on poultry and extend its shelf life, but chlorine is not effective at eliminating Salmonella attached to the carcasses. Moreover, chlorine is controversial as a food additive. Toxic byproducts are created by chlorine in water, and the food industry is trying to reduce the use of chlorinated water in food processing. The American Public Health Association has a policy statement calling for using alternatives to chlorine whenever possible.

To address these concerns, USDA commissioned a study of chloroform resulting from chlorine treatment of chicken and determined that the public health impact was minimal. However, there are many mutagenic compounds other than chloroform in chlorinated water, and chlorine treatment of tissue may result in numerous compounds not studied by USDA.

With or without chlorine, the poultry processed through the nation’s chill tanks has an unacceptably high percentage of Salmonella and Campylobacter contamination. One study examined the impact of six different poultry processing modifications, including chlorination of the chill water, on both total bacteria and pathogen levels. Even with all six modifications in place, the incidence of Salmonella in the poultry ranged from 9.4 to 47.9% and Campylobacter ranged from 53.1 to 100%. Clearly, even with the best technology in place, the extensive reliance on water in poultry processing still results in a highly contaminated product.

C. Skin in Ground Product

Ground chicken and turkey products are far more contaminated with Salmonella and Campylobacter than ground beef. While ground beef products have Salmonella contamination rates of 4.3%, the corresponding rate for ground chicken is 38.2% and ground turkey is 27%, according to USDA’s recent survey of ground products (see Table III). Campylobacter showed an even greater differences between beef and chicken, 0.2% vs. 60%.

USDA regulations allow skin in and on poultry products, with allowable amounts ranging from 8% to 20%. Skin may be added to processed products, such as ground poultry and poultry burgers, to the same extent that it is allowed in regular poultry. Hamburger, in comparison, must be made of beef of skeletal origin. The Research Triangle Institute report said that the basis for this difference is traditional poultry industry practice.

As the exterior surface of the animal or bird, it is not surprising that the skin harbors more bacteria than the meat does. This is particularly true for poultry products, which are subject to processing practices like scalding, defeathering, and immersion chilling demonstrated to promote cross-contamination.

Thoroughly cooking whole poultry or cuts of meat will kill most bacteria that exist on the surface, including the skin. However, ground products present different risks. Grinding meat or poultry mixes the surface bacteria with the sterile interior meat. The inclusion of skin when raw products are being ground certainly adds more harmful bacteria than if skin were removed prior to grinding.

If ground product, whether meat or poultry, is not fully cooked, it presents a high risk of causing food poisoning. The addition of skin clearly increases that risk for consumers of ground poultry products.

Innovation in the Wings

In the long run, innovation will likely increase the microbiological safety of poultry. For innovation to be effective, however, the poultry industry must shift its research focus from methods to increase production, and instead develop strategies to reduce the numbers of poultry carcasses contaminated with Salmonella, Campylobacter and other human pathogens.

A number of poultry processors are already trying innovative techniques to minimize or eliminate Salmonella and Campylobacter in their products. A key improvement, discussed below, would be the elimination of Salmonella from poultry flocks, as has been done in Sweden. Pathogen-free birds would then arrive at the slaughterhouse. Here are a few methods that look promising:

Air chilling: Air-chilling is being utilized in Europe and Canada, and one company has announced plans to open a plant this year in Indiana to produce air-chilled chickens. Carcasses are cooled through an air-chill chamber, which eliminates most water uptake and cross-contamination. The company claims that the process, combined with an antimicrobial treatment, will allow it to market a product with significantly reduced pathogen incidence and levels. Although the effectiveness of this process is still uncertain, the lack of excess water in the poultry package will likely reduce the risk of cross-contamination in consumers’ kitchens.

Competitive exclusion: Another process being used in Europe is competitive exclusion. Poultry become colonized with Salmonella within the first few days of life, if Salmonella is present in the environment or in the food. Treatment with a competing bacteria shortly after hatching will reduce the likelihood of the chick being colonized with Salmonella. Some products are currently being tested, while others are already available in the U.S. to treat chicks to reduce flock contamination rates.

Steam Pasteurization: Steam pasteurization is a process that exposes the internal and external surface of carcasses to high pressure steam to kill harmful bacteria. The process was approved for beef carcasses in December 1995, following extensive testing of steam pasteurization equipment both in laboratory and in-plant trials. Testing was specifically targeted to assure the system killed pathogens like E. Coli 0157:H7 and Salmonella on beef carcasses. The equipment manufacturer is currently testing steam pasteurization on poultry carcasses for its effectiveness on Salmonella and Campylobacter. If found to be effective, the manufacturer hopes to have USDA approval by the end of 1996.

Trisodium phosphate (TSP): One company using this rinse on turkey products prior to immersion chilling claims that TSP effectively eliminates Salmonella and significantly reduces other pathogens on the product. However, there have been reports that the results have not held up in other processing plants. The company that has had the best success rate uses TSP in conjunction with other strict processing controls designed to minimize contamination.

Vent sealing: One plant, in conjunction with USDA plant personnel, discovered that super-glue could be used to seal the anal vents of poultry to prevent the excretion of fecal material. However, this system has not been approved for use by the Food and Drug Administration. Proponents of this system claim that if this system were used prior to the carcass entering the scalder, it would likely result in less cross- contamination in the scalder. It would also eliminate later excretion of fecal material, which increases the need for reprocessing. However, there are questions about whether the process would result in residues of super-glue on the poultry coming out of the scalder. Other ideas for vent sealing include some type of mechanical plug or staple.

Irradiation: The Food and Drug Administration has approved the use of irradiation on poultry products. This “star wars” technology uses gamma irradiation, electron irradiation, or X-rays to kill harmful bacteria. The technology would not sterilize the food, and surviving bacteria could start to multiply if the conditions are favorable. The technology is not widely used due to consumer acceptance issues and even the administrator of USDA’s Food Safety and Inspection Service, Michael Taylor, recognizes that irradiation may not be the solution to contaminated meat. In 1994, he told the New York Times that irradiation won’t be widely accepted “until the public is convinced everything else has been done.”

Technical innovation will clearly be crucial to minimize the contamination of poultry products. However, it is not the entire solution. Standardizing processing practices should help minimize contamination in poultry processing plants. However, such practices must be appropriately designed to assure that contamination actually is reduced, and verification using microbial testing is critical to assure that the system is working as intended.

The Impact of HACCP

In February 1995, USDA published a preliminary rule to require that all meat and poultry processors institute systems of preventive controls. These systems, known as Hazard Analysis and Critical Control Points (HACCP), should result in safer meat and poultry products. Coupled with microbial testing and performance standards, this rule represents the most important single development in food safety in nearly 40 years. But efforts by the meat and poultry industry are putting this rule at risk. These industries may have already succeeded in weakening some essential elements of the rule.

One key provision of the proposed rule that industry pressure appears to have weakened is a requirement that all processors sample their products for Salmonella on a daily basis. Instead, the Department is likely to propose a sampling plan that requires industry to test only for generic E. coli bacteria, which are not pathogenic.

One major goal of HACCP is to prevent contamination of food with harmful bacteria. Without pathogen testing, there is no way to measure the HACCP systems’ effectiveness, and it becomes little more than an industry honor system. Replacing industry sampling for Salmonella, an actual pathogen, with sampling for generic E. coli, a microbial “indicator” organism, will greatly reduce the ability of USDA to monitor the HACCP systems’ effectiveness for controlling harmful pathogens.

In its original February 1995 proposed rule, USDA proposed that all meat and poultry processors should test daily for Salmonella. The test was not a quantitative test, but measured whether or not the pathogen was present. This simple testing regime, combined with proposed targets for pathogen reductions, would exert pressure on poultry producers to reduce their Salmonella contamination rates. According to USDA:

FSIS believes that it is reasonable and feasible to require . . . that all establishments control their processes so that their Salmonella incidence is no greater than the current national average. . . . If reductions 25 to 50 percent below the national baseline are reasonably available in the near term for a particular species, all companies should work to achieve them.

USDA also noted that some in the poultry industry were achieving a frequency of occurrence of Salmonella contamination as low as 5% or less, which is significantly below the 25% national contamination rate.

The alternative E. coli sampling plan would require industry to test for the level of these bacteria in their products. Although generic E. coli can provide a measure of fecal contamination and general sanitation, it does not tell poultry processors if any pathogens are actually present.

The USDA described the E. coli sampling proposal as its “current thinking” at meetings it held in September 1995, meetings which were held following an effort by the National Broiler Council, the American Meat Institute, the National Meat Association, the National Turkey Federation, and others to kill the USDA HACCP rule by means of a rider to the Agriculture Appropriations Bill being considered in Congress. Although these trade associations were unsuccessful in getting Congress to stop the proposed HACCP rule, USDA agreed to hold meetings on controversial aspects of the proposed regulation, including the proposed sampling regime.

At these meetings, USDA also announced that the government would conduct Salmonella sampling rather than the industry. However, this commits the resources of the government to conduct all pathogen testing for the new HACCP program and even opens the door for the industry to eliminate the government sampling program through more maneuvers in the Congressional appropriations process.

The proposed E. coli sampling regime could give a false sense of security to poultry processors and regulators, because they will think that the HACCP system is working as long as the levels of E.coli are shown to be decreasing. However, studies have documented that the numbers of E. coli on a carcass can decrease, while the percentage of carcasses contaminated with Salmonella is actually increasing. That reflects the fact that cross-contamination of product is occurring even while levels of E. coli are declining.

Three studies, with findings documented in Tables IX, X and XI, show that the percentage of poultry contaminated with Salmonella generally increases even while the levels of generic E. coli on the products declined. For instance, in Table IX, the percent of carcasses contaminated with Salmonella increased from 58% to 72%, while the level of E. coli declined more than ten-fold. In Table X, we see the percent of carcasses contaminated with Salmonella double, from 24% to 49%, at the same time that E. coli levels declined more than ten-fold.

Table IX. Baseline data on incidence of contamination.1

Stage of Processing	Number of	Escherichia coli	Salmonella
Pre-evisceration	160	2.17	58
Prechill	160	1.46	48
Postchill	158	.87	72

a-c Means within columns with no common superscripts differ significantly (P<.05).

1 Taken from James et al. (1992a).

2 Mean log10 colony forming units per carcass.

3 Percentage of carcasses positive for Salmonella.

Source: William O. James et al., “Cost Effective Techniques to Control Human Enteropathogens on Fresh Poultry,” Poultry Science, Vol. 72 (1993), p. 1175.

Table X. Effect of scalder water counter-flow and postscald rinse prior to picking on incidence of contamination.

Stage of Processing	Number	Escherichia coli	Salmonella
Pre-evisceration	99	2.09	24
Prechill	99	1.61	28
Postchill	49	.89	49

a,b Means within a column with no common superscripts differ significantly (P<.05).

1 Taken from James et al. (1992b).

2 Mean log10 colony forming units per carcass.

3 Percentage of carcasses positive for Salmonella.

* Significant difference (P<.05) from baseline value.

Source: William O. James et al., “Cost Effective Techniques to Control Human Enteropathogens on Fresh Poultry,” Poultry Science, Vol. 72 (1993), p. 1176.

Table XI. Effect of chlorination of chiller water on incidence of contamination.1

contamination.

Stage of Processing	Number	Escherichia coli	Salmonella
Pre-evisceration	99	2.72	33
Prechill	50	2.04	43
Postchill	50	1.20	46

a-c Means within a column with no common superscripts differ significantly (P<.05).

1 Taken from James et al. (1992c).

2 Mean log10 colony forming units per carcass.

3 Percentage of carcasses positive for Salmonella.

Source: William O. James et al., “Cost Effective Techniques to Control Human Enteropathogens on Fresh Poultry,” Poultry Science, Vol. 72 (1993), p. 1176.

Other data are available on Enterobacteriaceae, which is the broader family that includes the E. coli bacteria. This family contains the enteric pathogens, like Salmonella and Yersinia, as well as other bacteria. Enterobacteriaceae are widely distributed in nature, including in the intestinal tracts of animals. In one study, the results of which are summarized in Table XII, the percentage of carcasses contaminated with Salmonella doubled and tripled even while the levels of total Enterobacteriaceae were declining ten-fold.

Table XII. Levels of Enterobacteriaceae and Salmonella incidence before and after chilling on fully processed broilers from two commercial processing plants.1

Sampling Point	Enterobacteriaceae		Salmonella Incidence (%) # +/# sampled
	Plant A	Plant B	Plant A	Plant B
Pre – chill	6.01	6.09	5/40 (12.5%)	4/40 (10%)
Post-chill	4.97	4.97	11/40 (27.5%)	15/40 (37.5%)

1 Means based on 40 samples per sampling point. Means (and incidence) in columns which are not significantly different (P>.05) are followed by the same lower case letter.

Source: H.S. Lillard, “The Impact of Commercial Processing Procedures on the Bacterial Contamination and Cross-Contamination of Broiler Carcasses,” Journal of Food Protection, Vol. 53 (1990), p. 203.

Data also show a similar trend when comparing inspection-passed and reprocessed carcasses. The next two studies, Tables XIII and Table XIV, show that in most plants that were studied, the percent of carcasses contaminated with Salmonella can increase on reprocessed carcasses even while generic E. coli levels decreased. As observed earlier, the data on reprocessing show a great deal of variability between plants.

Table XIII. Means a of E. coli and percent prevalence of Salmonella on conventionally processed and reprocessed commercial broiler carcasses.

Plant	Sample	N	EC	%SAL
A	Conventional Reprocessed	150 150	1.61+/- 0.70 1.87 +/- 0.67	31.3 53.3
B	Conventional Reprocessed	150 150	1.52+/-0.61 1.16+/-0.82	76.6 69.3
C	Conventional Reprocessed	150 150
D	Conventional Reprocessed	145 145
E	Conventional Reprocessed	150 150
Overall	Conventional Reprocessed	745 745

a Means SD of three replications.

b EC = E. Coli

c SAL = Salmonella

Source: L.C. Blankenship et al., “Broiler Carcass Reprocessing, A Further Evaluation,” Journal of Food Protection, Vol. 56 (1993), p. 984.

As illustrated in Table XIV, this study shows the same trend with respect to E. coli levels and the incidence of Salmonella, together with data on Campylobacter.

Table XIV. Meansa of E. coli and percent prevalence of Salmonella and Campylobacter on conventionally processed and reprocessed commercial broiler carcasses.

Plant

Sample

Number of Carcasses

EC (b)

%SAL (c)

%CAMP (d)

Conventional

Reprocessed

100

2.02

1.72 (h)

Conventional

Reprocessed

100

2.74

2.26 (I)

Conventional

Reprocessed

100

2.79

2.15 (I)

Conventional

Reprocessed

100

3.06

2.57 (I)

Conventional

Reprocessed

100

2.77

3.02 (g)

aMeansSD of ten replications.

b EC = E.Coli (all figures 0.08).

c SAL = Salmonella.

d CAMP = Campylobacter jejuni/coli.

e IPC = Inspection-passed carcass.

f RPC = Reprocessed carcass.

g-iBacterial means for each microbial assessment within columns are significantly different (gP .05, hP .01, iP .0001).

Source: A.L. Waldroup et al., “Effects of Reprocessing on Microbiological Quality of Commercial Prechill Broiler Carcasses,” Journal of Applied Poultry Research, Vol. 2 (1993), pp. 113-114.

Recommendations

The following recommendations are based upon the findings discussed in this report. Adoption of these recommendations is essential to set the poultry industry on the path to improving the safety of its products:

USDA should set performance standards for poultry products to force the poultry industry to reduce or eliminate Salmonella and Campylobacter in its products. Historically, the industry has focused on reducing the levels of Salmonella, rather than eliminating it altogether. This strategy ignores the fact that if Salmonella is present at all, its levels will increase even under normal handling conditions.

For HACCP systems to be effective for poultry processing plants, USDA should mandate that E. coli testing be combined with ongoing industry testing for Salmonella and Campylobacter. E. coli testing alone will not ensure control of cross-contamination in poultry plants.

The poultry industry should utilize alternative chilling technologies, such as dry chilling or blast chilling, that eliminate problems with cross-contamination. If shown to produce less contaminated poultry products, new plants should be required to install these technologies.

USDA should re-examine the long-term health consequences of chlorine use in poultry processing. The agency should consider the potential risks of toxic byproducts formed with the use of chlorine.

USDA should require the removal of skin prior to grinding poultry products to help reduce the prevalence of pathogens found in ground products.

The poultry industry should direct research money to projects designed to reduce the levels of harmful bacteria in their products.

USDA and FDA should speed approval of technological innovation demonstrated to improve food safety.