Advances and Challenges in the Control of Salmonella in Poultry
By An Interview With J. Stan Bailey, Ph.D.
J. Stan Bailey, Ph.D., is a research microbiologist with the Russell Research Center, Agricultural Research Service, U.S. Department of Agriculture, in Athens, GA, where is responsible for research directed toward controlling and reducing contamination of poultry meat products by foodborne pathogens such as Salmonella and Listeria. Bailey has authored or coauthored more than 400 scientific publications in the area of food microbiology, concentrating on controlling Salmonella in poultry production and processing, Salmonella methodology, Listeria methodology, and rapid methods of identification.
During his 30 years of experience, he has received many honors, including 14 USDA Certificates of Merit for outstanding achievements in research; the 1992 Continental Grain Poultry Products Research Award; and the 1994 National Broiler Research Award. He is an active member of several professional societies, including the International Association for Food Protection, American Society for Microbiology, and the Poultry Science Society. He was appointed Fellow of the American Academy of Microbiology in 1994, served as an expert consultant to the FAQ in 1997, and was appointed scientific advisor to the International Life Sciences Institute in 1997. Bailey can be reached at firstname.lastname@example.org.
Food Safety Magazine: Salmonella has long been linked with poultry products. Where are we today in terms of contamination rates?
Bailey: In the 1970s, we were probably looking a Salmonella contamination rates on processed chickens in the neighborhood of about 50%. This has fluctuated somewhat since then, but has come down steadily through the years. When the U.S. Department of Agriculture (USDA) Food Safety Inspection Service (FSIS) conducted its baseline study in 1996, it reported the level of Salmonella on poultry at about 22% of carcasses. The agency used these rates as the basis for establishing the Hazard Analysis & Critical Control Points (HACCP) performance standards for Salmonella that are in place today. In the last three years, FSIS HACCP samples have shown levels of around 10-11%. The reduction in Salmonella levels on poultry from close to 50% to around 10-11% is very good. This is probably due to several factors, including changes in how birds are grown and processed.
While Salmonella has been known to us for a long time — and therefore many researchers today are moving their efforts to pathogens like E. coli, Listeria and Campylobacter — this pathogen will always receive a great deal of attention from the industry because it is still one of the leading causes of foodborne illness and the regulatory performance standards are primarily written around it. Years of research experience has taught us much about how Salmonella survives and persists in the poultry environment, how it colonizes the intestinal tract of the bird, how it can be spread through the fecal material, how it can be spread in the processing plant, and what methods are effective for its detection. Based on years of research findings, many prevention and control approaches have been tested and implemented, but much remains to be done if we are to continue to reduce the levels of Salmonella found on the chickens that we eat.
Salmonella often is associated with chickens, and deservedly so to a point, but the U.S. Centers for Disease Control and Prevention (CDC) has not reported in a number of years the full gamut of outbreak sources. Chicken is only one of many, many potential sources. If you look at some of the food- borne illness outbreaks associated with Salmonella we’ve seen over recent years, many have been associated with orange juice, cantaloupe, melons, beef and pork. Chicken, clearly, is an important part of that, but unlike Campylobacter where 50%-70% of cases are epidemiologically linked to chicken, poultry is associated with only 10%-20% of foodborne salmonellosis cases. It should be noted that Salmonella enteritidis is almost exclusively associated with eggs in the U.S. compared to the United Kingdom where S. enteritidis is associated with both eggs and broiler meat.
Food Safety Magazine: What are some of the significant changes you’ve seen in the industry during the last 30 years that have aided in the reduction of Salmonella levels on poultry products?
Bailey: My personal career reflects many of these changes in the poultry industry. In the first 12 years of my career, I worked in the processing plant side of the industry. We worked with different approaches, trying to find a terminal pasteurization treatment for chickens similar to what we had established with milk in the dairy industry. We wanted a treatment with which we could treat the birds at end of processing to eliminate any bacteria on them. That was in the 1 970s, and we still haven’t been able to find any such terminal pasteurization treatment that will do this without actually cooking the chicken. Despite this, Salmonella levels have been reduced in poultry processing plants. Why? The most significant changes involve the types of equipment and treatments that were developed in the last 30 years and now are widely used in processing plants. The industry now uses counterflow chillers and counterflow, multichamber scalders, for example, and the processing involves more wash steps and the use of antimicrobial chemical spray treatments, all of which have helped to reduce Salmonella levels by reducing the amount of bacteria on carcasses or by reducing the occurrence of cross-contamination.
Probably, the biggest single change in the industry occurred 10 years ago as a result of National Broiler Council (now National Chicken Council)-sponsored research conducted by Amy Waldroup (then at the University of Arkansas) and others. This research led to the inclusion of high levels (30-40 ppm) of chlorine in the chill tank for most domestically consumed broilers, which helped to dramatically pull down the levels of Salmonella.
Around 1985, scientists in our Research Unit at the USDA, Agricultural Research Service (ARS) made the decision that we would be able to greater influence reductions in the incidence of Salmonella in poultry by concentrating on controlling what was going on at the farm as opposed to what was going on in the processing plant. We spent two or three years working in the feed arena, looking at types of pelleting, using different lengths of conditioning time, and studying different aspects of feed to see whether it was a big contributor to Salmonella in chickens. Some of the research showed increases in conditioning times helped reduce levels, for example, and has since been picked up by industry. However, we came to realize that while feed can play a role in the transmission of Salmonella to chickens, it probably was not as much a factor compared with many other sources, so we moved our research back to the grow-out level. Thus, for the last 15 years or more, we have concentrated on how to grow chickens in a manner such that they will have reduced levels of Salmonella when they arrive at the processing plant.
A significant part of the research we’ve done has centered on the development of an effective competitive exclusion product. This is a fairly natural, simple process in which you take the normal intestinal bacteria from a healthy adult chicken and inoculate a newly hatched chick with it as soon as possible after the chick is hatched. By doing so, the chicken is far less susceptible to becoming colonized with Salmonella, even if they become exposed to Salmonella during grow-out. This method works on a natural principle that while a newly hatched chick is readily colonized by just a few cells of Salmonella, by the time the chicken is two or three weeks old it is far more difficult to colonize, even with high numbers of Salmonella, primarily because of the competitive microflora in the intestinal tract. Essentially, the competitive exclusion method uses this natural principle. We capture those normal bacteria from the intestinal tract of a healthy adult chicken, grow them up, give them back to the chicks, and within a matter of an hour or two, the chicks will be as resistant to colonization by Salmonella as a healthy adult chicken. We’ve spent many years researching and refining a product and the delivery systems for this process. In many parts of the world, the poultry industry is using competitive exclusion treatments, including the ones we’ve developed in our lab. We currently are working to get U.S. Food and Drug Administration (FDA) approval for the competitive exclusion product here.
In recent years, we’ve also spent a lot of time working on the epidemiology and ecology of Salmonella in broiler production, in which we’re tracking all the sources of Salmonella and following it through the process of growing, transporting and processing the chickens. What we know is that sources of Salmonella might involve anything from the grow-out environment from the previous house, to birds coming from the breeder’s flock in the hatchery, to flies, rodents, wild birds and insects. Other sources could be feed, the farmers themselves, and even other domesticated farm animals in the environment that harbor Salmonella. By trying to identify where the Salmonella on the final processed chicken is coming from, we can better refine research efforts and the industry can better refine its resources in trying to prevent and control Salmonella.
At the present time, we see that if the processor does everything right, including using elevated levels of chlorine and operating with good, properly set up equipment, he can actually reduce the level of Salmonella in the processing plant. However, you can only reduce it by a limited amount, and that is going to be greatly influenced by what comes into the processing plant. You can overwhelm the control system if you have high levels of Salmonella on the birds as they come in. If you have that level down fairly low before the birds enter the plant, and if you have done everything right in the processing plant, you can go a long way in helping the situation.
For many of the companies, paying close attention to what’s going on at the farm level has led to closer attention being paid to what’s going on in the hatcheries. Many companies now use disinfectant sprays during the hatch period, and do a better overall job of hatchery sanitization to try to reduce any levels of Salmonella coming out of the hatchery. Many companies also are paying more attention at the breeder farm level, trying to provide more control a step farther back in the process. Certainly, increases in biosecurity measures have been a byproduct of this, particularly in terms of controlling movement into farms and between houses. For example, many companies use sanitizing footbaths and dips not previously used, and are trying to control moisture in the chicken houses to prevent leakage into the litter that effects a greater chance that Salmonella will grow.
Food Safety Magazine: What are some of the challenges the industry continues to face?
Bailey: Again, in the last three years FSIS HACCP samples have shown Salmonella levels in poultry at around 10-11%, but we still see situations where there are high-level pockets of Salmonella that arise from time to time. Although we know a lot about Salmonella — how it moves around during grow-out, transportation and processing — it is difficult to get a handle on why this is the case, particularly when you consider the numerous potential sources of Salmonella. Even within any particular company that has established similar prevention and control policies throughout their grow-out operations, we’ll still see some areas that have higher Salmonella levels than others. We’re working on identifying what would cause one area to get a sudden, sporadic high level of Salmonella and find out what we can do to help eliminate that.
Another area that continues to be a challenge is transportation, and although there’s a good bit of work that has been done, there is not a lot of effective solutions. We generally see a significant increase in Salmonella any time when transporting chickens, particularly from the field to the processing plant. During that time, from a period of a few hours to 10 or 12 hours, we see a considerable increase of Salmonella in and around the birds. You have to get the birds to the plant and it is difficult to do this in a way other than placing the birds in a closed, confined environment. It’s somewhat stressful on the birds, and just the nature of the animals being close together causes the problem.
On the topic of stressed birds, it should be noted that older birds are more resistant to Salmonella than young chicks even when stressed. However, if they are under stress and their intestinal microflora is disrupted it is possible that Salmonella will be able to colonize the bird. For example, if you’ve had to treat the chickens with therapeutic levels of antibiotics (which changes the gut microflora) and then you take them off of the antibiotics, the birds are more susceptible to Salmonella at that time.
On a related note, another challenge may arise if the use of some sub-therapeutic antibiotics are disallowed in the future by regulation. The growth promoter advantages that these antibiotics provide, as well as just the general chick health that they help to maintain, would be removed. This will pose bigger challenges to the industry in controlling Salmonella and other bacteria. We’re currently looking for alternatives, which is one of the research areas on the forefront.
Food Safety Magazine: Are there any developments in Salmonella detection technologies that have proved useful to the poultry industry?
Bailey: When I started working, the only detection methods we used were conventional culture methods, and while many microbiologists still do use conventional methods, there has been a move toward the use of rapid and automated methods as they’ve become available. The most widely used rapid pathogen tests, including those for Salmonella, are either enzyme-linked immunosorbent assay (ELISA), gene probe or polymerase chain reaction (PCR). The ELISA is done by standard conventional procedures or by using automated systems such as VIDAS (bioMeneux). Another highly automated procedure is the BAX (DuPont Qualicon) PCR systems. Most of the ELISAs, gene probes or PCRs meet minimal performance standards of less than 5% false negatives/false positives. Each has some advantages and disadvantages in terms of up-front equipment or individual test costs, and throughput capability.
We have really good methods for Salmonella compared with other pathogens, such as Campylobacter and Listeria. We can detect down to just one or two cells of Salmonella (before enrichment procedures) by either cultural or molecular methods. The biggest change in pathogen detection methodology is not in sensitivity, but rather, in the time it takes to complete. Standard culture procedures will take about four days, whereas a PCR procedure (even after an overnight enrichment) can be done in about 24-26 hours. For further processed products that are ready-to-eat and where there is a zero tolerance of Salmonella, these rapid methods have definitely helped the industry in hold-and-release times.
In raw products, most of the testing is done by the industry either for historical baseline purposes or to see what to expect on the FSIS HACCP performance standards samples. In this case, the time is not as important as the ability to use the rapid methods equally and with built-in quality control standards at all locations around the country. If you’ve got plants in California, Arkansas, Maryland and Georgia, for example, and you would like to use more or less the same procedures with the same degree of quality assurance so that you can compare trend data, the use of rapid methods provides another added benefit, because most of the rapid methods procedures have excellent quality control standards built in.
The recommendation I would make is that before adopting any of these assays, whether it’s chicken products or something else, companies should work with the tests or systems for a period of time with their particular type of products to make sure that they perform well in their specific operations. In an industry lab, for example, where you’re running samples full-out every day, the rapid methods enable you to run many more samples and screen them in 24 hours with just one or two people, and you can probably do three times as many samples as the conventional methods would allow. This is the predominant situation in the industry.
As far as other advances in detection, real-time detection will probably require the use of some type of biosensor. However, despite a great deal of research, biosensors probably are a few years from being of practical use. I do see increased use of genetic methodology for trace- back purposes, such as for the epidemiological type of studies I mentioned earlier. It’s not really enough anymore just to say you’ve detected Salmonella; you want to know where that Salmonella came from in order to develop effective prevention and control strategies. I think that we’ll see a lot of developments in this area coming along in the next few years.
A number of the newer assays, particularly PCR-based assays, that are currently being developed will offer the ability to reasonably quantitate the pathogen of interest. This will enable the user to detect the presence of the pathogen and, at the same time, tell how much there is, which is a better indicator of how big the problem might be. For example, if we are developing Salmonella intervention procedures, a quantitative test might show that we have reduced the level of Salmonella in the birds by 2 or 3 logs, which might be significant; whereas, if we had used a qualitative test alone, we might not have seen a difference. From many perspectives the ability to quantitate is very important, although the existing performance standards don’t dictate that this be done.
Food Safety Magazine: What kinds of research, government or industry efforts with regard to the prevention, control and detection of Salmonella in poultry do you think are on the horizon?
Bailey: I see a more concerted effort on the whole farm continuum—from the breeder farm through the hatchery, to the grow-out and transport, through the processing end. In recent years, the majority of efforts have been placed solely on the processing plant, and that’s been good, but we’ve been able to reduce Salmonella contamination rates from 50%, to 20%, to 10%. If we’re going to get down further than that we’re going to have to look at the whole farm continuum and attack it from all directions.
If continued pressure to remove subtheraputic antibiotics from the feed comes to fruition, there will be many new challenges to general bird health. The most significant will be the increased susceptibility of the birds to some of the pathogens, which will have to be addressed through research to find alternatives. A more immediate challenge will be to identify the cause(s) of and prevent the sporadic high levels of Salmonella that we see from time to time in various grow-out areas.