The U.S. Department of Agriculture’s Food Safety and Inspection Service (FSIS) intends to implement a sweeping and total change to its food safety program by drastically modifying carcass-by-carcass inspection and reducing the government surveillance of meat and poultry safety. This has become known as the HACCP Based Models Project, or “HIMP.” FSIS has been trying to get HIMP to work for more than a decade, yet Salmonella contamination in poultry production continue to be a significant public health hazard.

Hazard Analysis Critical Control Points, or “HACCP,” is a change to the federal law for meat and poultry inspection that was put into effect in 1998 for large plants. HACCP puts increased responsibility for food safety on to the companies’ management and lessens government authority to take actions, in real-time, to prevent unsafe food from reaching the consumer.

The HACCP Based Models Project has been functioning in pilot plants for more than 13 years. But, during that time, FSIS has failed to effectively demonstrate the viability of the pilot program through objective and statistically sound measures.

There have been numerous reports that criticize the performance of HIMP by consumer safety groups such as the Government Accountably Project, Public Citizen, the Food and Water Project and the government’s own Government Accountability Office (GAO). The GAO reported its concerns in December 2001 and again in August 2013, with many of the problems identified in the 2001 report again identified in 2013. This means that problems with the pilot program have been occurring in selected plants for more than a decade.

The pilot plants ship product to consumers as if it had been produced under proven inspection methods, meaning that consumers have no practical way of knowing if the meat and poultry they buy is produced in one of these “experimental plants.”

In a U.S. Centers for Disease Control and Prevention (CDC) report entitled, “No Progress in Salmonella During Past 15 Years – Food safety annual report card targets hard-to-prevent infection,” CDC Director Thomas R. Frieden, M.D., M.P.H. stated:

 “Although foodborne infections have decreased by nearly one-fourth in the past 15 years, more than 1 million people in this country become ill from Salmonella each year, and Salmonella accounts for about half of the hospitalizations and deaths among the nine foodborne illnesses CDC tracks through FoodNet.”

In the same report, Elisabeth Hagen, MD., then-Under Secretary for Food Safety in the Department of Agriculture, states,

“…  far too many people still get sick from the food they eat, so we have more work to do. That is why we are looking at all options, from farm to table, in order to make food safer and prevent illnesses from E. coli, Salmonella, and other harmful pathogens.”

I was a FSIS employee for 18 years, from 1987 through 2004. I served as a Consumer Safety Inspector in beef, pork and poultry plants in more than 30 different assignments locations. I was promoted five times in my career and was active in improvements in consumer safety and in employee safety and in addressing inadequate staffing within the FSIS inspection program. I was a dedicated public servant and believed strongly in the FSIS mission of consumer protection.

In 1997, FSIS asked the National Joint Council of Food Inspection Locals (the U.S. government meat inspectors’ union) to submit proposals to redesign the carcass-by-carcass inspection system. I co-wrote and submitted a proposal that would become the precursor of the HIMP model. The union was not opposed to improving the current inspection system. It was fully understood that changes in slaughter processing through mechanization was allowing significant contamination of carcasses to occur after food safety decisions by FSIS inspectors.

The proposed reconfiguration moved government oversight to the end of the process, increased the rate of “finished product standards” testing, and greatly improved the health and safety conditions of inspectors.  The proposal also gave sensible flexibility to adapt to temporary inspection staffing changes and corrected shortcomings in the traditional configuration of inspection.

FSIS adopted the proposal, radically modified it, and installed their version of a pilot configuration in several plants. In the process of transforming the union proposal, FSIS declassified disease conditions as a public health concern and restricted control by the inspector at the end of the processing line. Using the FSIS modified model, the final inspector is restricted in taking immediate action to stop contaminated carcasses from passing to the further processing and packing steps. Plant “sorters” replaced federal inspectors for segregation of diseased carcasses. The sorters are not required to have specific training and are under the direct control of plant production managers who are responsible for maximizing production rate and volume.

I was tasked, through a negotiated agreement between FSIS management and the Inspector’s Union, with evaluating the effectiveness of the HIMP pilot. Not only did I personally observe pilot plant performance during periodic visits, I placed union representatives in the pilot plants who evaluated processing performance through independent testing conducted parallel to FSIS and plant quality control testing. Not unexpectedly, the pilot project did not perform as was hoped due to drastic deviations from the union proposal.

One key element of the union’s proposal was to maintain the regulated line speeds of 70 to 91 birds per minute in poultry plants. This was critical due to the observed contamination rates in processing at those line speeds. Increasing the line speeds would logically result in an increased volume of contaminated product. Yet, FSIS deleted this requirement of the proposal and allowed plants to run at any speed that plant management desired.

Pilot plants almost immediately doubled their line speeds and, as predicted, allowed digestive tract contamination of product affecting huge numbers of carcasses without effective remedy. Even under the scrutiny of pilot plant performance, the HIMP model failed to prevent recurring incidents of visible fecal contamination of product, as documented by FSIS Noncompliance Records. Recurring failures of the pilot plants’ HACCP plan occurred with “no effective corrective actions and preventive measures” as required under law. Failures on the part of FSIS management to respond to problems in the pilot plants that I evaluated prompted me to become a “whistleblower” through the Government Accountability Project. I also shared my findings with Public Citizen, an advocate for consumer safety and government integrity.

In 2000, I compiled a report on HIMP pilot plant performance that revealed serious failures of the FSIS surveillance and enforcement of the law. I made exhaustive efforts to get FSIS management to respond to repetitive HACCP failures that were documented on their own FSIS Noncompliance Reports. In March 2000, I presented my report at a public meeting in Washington, D.C. FSIS management officials, industry representatives, and consumer groups attended the meeting, along with members of the media. Following the presentation of my report, the manager of one of the pilot plants left the meeting and immediately withdrew his plant from participation in the HIMP pilot program.

In 2013, the Government Accountability Office published its objections over data gathering in HIMP pilot plants. The GAO report noted:

“The U.S. Department of Agriculture (USDA) has not thoroughly evaluated the performance of each of the pilot projects over time even though the agency stated it would do so when it announced the pilot projects […] Specifically, there are limitations in the agency’s data analyses in its report evaluating the pilot project at young chicken plants, and there is no report evaluating the pilot project at young turkey plants.”

The HACCP Based Models Project configuration has significant problems that must be addressed if consumer safety is to be safeguarded. In future articles for Food Safety News, I will outline serious problems in a deregulated government surveillance program, the over-reliance on anti-microbial interventions in meat and poultry production, and the evolution of a failing inspection program to protect consumers from foodborne illness resulting from contaminated meat and poultry. I will also detail specific improvements to the federal inspection system that can be made to correct shortcomings in consumer safety.

  • doc raymond

    Alvin, you have many flaws in this OpEd piece. First of all you cite the CDC report and the over 1 million cases of Salmonella food borne illnesses as if they all came from chicken. You know very well that produce is the number one cause of Salmonellosis in humans. In the HIMP plants proposed by FSIS, the inspectors can stop or slow line speeds if the plant demonstrates inability to control salmonella numbers of defects such as feathers and bruises on carcasses. They do NOT have that authority now. Salmonella contamination of carcasses in the 20 chicken HIMP plants is equal to or lower than that seen in conventional plants. You know that the inspectors in conventional plants are paid by US tax payers to sort carcasses to eliminate defects such as feathers, bruises and broken limbs–this has nothing to do with food safety but is all about protecting the brand image. You know that in HIMP plants there are more verification inspectors on the floor, doing food safety assessments, quadrupling the number of carcasses closely examined for fecal contamination and sampling for pathogens likely to cause a foodborne illness, while in conventional plants the only sampling required is for generic E coli, a bacteria that will not make consumers sick.

  • John Munsell

    It is indeed refreshing to find a former FSIS inspector who has taken the time to document historical agency actions, and to reveal them, including problems conveniently ignored by FSIS. I look forward to your future articles, discussing a deregulated gov surveillance program, over reliance on interventions, and specific improvements. Can you clarify something for me? You stated above that you prepared a report in 2000 about HIMP pilot plant performance that revealed serious failures, and presented it at a public hearing in WDC. Was the hearing hosted by FSIS? If so, how did the agency allow you to publicly reveal HIMP problems? And, were your comments the beginning of the end for your FSIS career, seeing that you retired in 2002? I encourage you to be fully revealing, for which you will incur the wrath of many people. John Munsell

  • Foster Leaf

    Alvin: Thank you for your service and active participation in the process. Given your background, you know that HIMP is a change in FSIS inspection methods to reduce inspection resource requirements. HIMP is not a food safety initiative because it does not introduce any new food safety standards. You also know that Salmonella is not a single problem with a single solution. It requires the efforts of everyone from the hatchery to the dinner plate. Mother nature will have her way. Salmonella is here to stay. You cannot control it. You can only attempt to protect yourself from it. We live in a society governed by the rule of law. Until the actors in this drama start doing what is legally required of them, stop pointing fingers at the other each other, stop promising what cannot be delivered, and stop demanding what is not owed, the negative impact of Salmonella on our food supply will not change.

    • John Munsell

      If HIMP is not a food safety initiative as you claim above, how can FSIS claim that HIMP will decrease the number of illnesses? At least FSIS is equating HIMP with improved food safety. I agree that HIMP does not introduce any new food safety standards, but merely introduces new THEORY that HIMP plants will produce safer food. Well, Alvin Sewell’s comments introduce common sense facts which overrides theory. I greatly look forward to his subsequent articles. John Munsell

  • BB

    More sampling for Salmonella and Campylobacter doesn’t fix the problem-it only let’s you know you have one. What’s the point of more sampling when these pathogens are not considered adulterants and allowed to enter commerce? You’ve already allowed the plants to run 140 birds per minute which is too fast to begin with. That gives the inspector 1.7 seconds to inspect each carcass as it whizzes down the line-hardly enough time for a good inspection. Even when fecal is observed offline, the plants corrective actions don’t fix anything ( i.e. turn on post-chill chlorine sprays or adjust a nozzle on the inside/outside bird washers….come on man!). FSIS NRs for pre-chill fecal are only a mere fraction of the amount of fecal that actually goes in the chiller. Chlorine in the chiller is inactivated by organic matter making it unavailable for it’s intended purpose. Incoming pathogen load at recieving needs to be reduced. Transport cages need to be cleaned on a regular basis. Evisceration equipment needs to be adjusted for bird size. Lines need to be slowed down across the board. A lot of chillers were put in and designed for the smaller birds of years past. Now birds are pushing 6-7 lbs which increases demand and overloads the chiller. These immersion chillers aren’t reducing Salmonella like they should be. Then we move right along to the cross-contamination of chicken parts in 2nd (further) processing. Garbage in….garbage out. I’m not sure if one is better than the other. Both HIMP and conventional systems are broken. FSIS will continue to tolerate pathogens in chicken. Big chicken will continue to make a profit and cheap chicken, but NOT safe chicken. Like Consumer Reports said, it’s the cost of cheap chicken. You had better cook it!!!

    • John Munsell

      BB: you make an erudite point in your first sentence questioning the value of increased sampling for Salmonella and Campylobacter. BB, all the adverse lab reports are valueless unless (a) FSIS implements enforcement actions when it observes ongoing pathogen failures (process control failures), and (b) the industry implements truly meaningful corrective actions to prevent recurrences. Neither a nor b is being accomplished, by intentional agency design. HACCP has provided FSIS a “robust, science-based” excuse to adopt a semi-retired non-involvement role in the industry. HIMP merely takes the agency’s retirement role up one more notch. Think of the beauty of it all: reduced inspection equals safer food! It thus is logical that if we merely totally defund FSIS, our food will be much safer. John Munsell

  • John Munsell

    Mike, as always, you ask pertinent questions for which truthful answers would constitute a PR disaster for FSIS. When the detection of a greatly increased percentage of contaminated carcasses is the result of a moderate reduction in line speed, it clearly reveals that visual inspection of carcasses is much more successful when the chain is slow enough that individual birds can actually be inspected. A primary problem here is that visual inspection for visible fecal is a throw-back to pre-HACCP days, when organoleptic inspection ruled the roost (pun not intended). Organoleptic inspection has been impaled on the cross of “Science-based HACCP”, never to be recognized again as valuable by FSIS. Thus, organoleptic inspection was thrown out in the bathwater. It should have been kept as a vital component of HACCP. FSIS disagrees. Instead, FSIS concludes that visual inspection has virtually no place in HACCP or HIMP, because meat plants now have a veritable array of cutting edge multiple hurdle antimicrobial interventions which will (theoretically) subsequently remove all visible and invisible fecal down the line. The agency endorses the procedure to allow visible fecal to appear on carcasses, because the carcasses will eventually be cleaned downstream at the abattoir, prior to the chiller. Pretty nifty, eh? We need a contemporary Upton Sinclair, and perhaps Alvin Sewell is that person! John Munsell

  • John Munsell

    Alvin, I appreciate ALL your comments, regardless of length. Ironic that Tom Billy called you a liar, which is an adjective used by FSIS officials in MT to describe plant owners who have revealed the truth about agency misdeeds in our state. To have Tom Billy make that false accusation constitutes a remarkable compliment for you. Am looking forward to the rest of your articles. The agency can thank Tom Billy for botching the Supreme Beef litigation efforts, for which FSIS was spanked by the judge. John Munsell

  • John Munsell

    My perception is that our food is ALREADY the safest available on the globe. Our focus on the HIMP issue should be that FSIS is cooking the books, making food safety claims that are fabricated by the agency’s spin meisters. Also, FSIS is totally closed-lipped, unwilling to precisely identify exactly what activities its future HIMP inspectors will be required to fulfill. The agency merely states that HIMP inspectors will be liberated to do “other” activities (such as microbial testing) which will allegedly prove the plant’s success or failure in producing safe food. What amount of testing? Will FSIS testing protocol at HIMP plants be meaningful, or merely an occasional necessary evil? Therein lies the rub. FSIS will adroitly avoid any meaningful incidence of testing, because when the agency becomes fully cognizant of the abnormally high incidence of Salmonella, well, consumers would expect the agency to implement MEANINGFUL enforcement actions against the plant, which the agency is fearful of doing. Also, consumers would demand that FSIS closely surveil a plant’s corrective actions, to independently validate that the corrective actions will guarantee the prevention of recurrences. Although (a) Prevention & (b) Corrective Actions (to prevent recurrences) form the basis of HACCP theory, FSIS refuses to require large source slaughter plants to implement either (a) or (b) when ongoing pathogens are observed. Thus, in order for the agency to be relieved of the delicate task of using enforcement actions at noncompliant plants, the agency minimizes its sampling frequency, allowing the agency to conveniently be ignorant of the true incidence of pathogens. Ahhhh, the beauty of “science-based” deregulation of the largest source slaughter plants! John Munsell

    • doc raymond

      John, currently there is only one Verification Inspector in conventional poultry plants. This VI will pull ten carcasses off the line and inspect for fecal material, blemishes such as bruises and feathers, and swab for bacteria. S/he will do this twice per 8 hour shift. IN the HIMP plants this is done EVERY HOUR as their is a VI for each separate line. Also plants will have to identify pathogens likely to occur that will make us sick and test pre-chill and post-chill for them. Currently they are required to test only once, and the testing is for generic E coli which will not make us sick. I know FSIS has not answered you directly, and I hope this helps just a little bit. Your distrust is well-earned, unfortunately. BTW, did you get my check for the losing Super Bowl bet?

      • John Munsell

        Yes, this does help. And yes, FSIS has not answered me directly, nor through media releases. And yes, I’ve received your check. My memory tells me that you mentioned in a blog perhaps 3 years ago that when poultry plants know that FSIS sampling will occur, that the plant greatly increases the chlorine level in their rinse in order to produce acceptable lab microbial test results. Is my memory correct? And did you say that the increased chlorine levels are strong enough to sting one’s eyes? I’ve had one current inspector describe to me how such chlorine treatments stung his eyes and his lungs, requiring him to transfer out of poultry plants. My impression is that FSIS is fully aware of this temporary increase in chlorine levels, designed exclusively to artificially lower the incidence of bad lab reports. As a former USDA Under Secretary for meat inspection, do you feel that such diversionary tactics by the industry are ethical? Another question for you: since the meat plant realizes that increased chlorine levels are required to obtain acceptable lab results, shouldn’t those higher chlorine levels be used day in and day out? I wonder if OSHA even cares. John Munsell

  • John Munsell

    Katy, the full implementation of HIMP is already guaranteed, because it has the full support of the large source slaughter plants (many transnational in scope) and global gov meat inspection agencies, which would prefer to be retired at the slaughter behemoths. John Munsell

  • doc raymond

    Alvin, FSIS does make NRs public information when asked for them, and I think you know that. Also, you mention the Union benefits from HIMP by creating better paying jobs that are more mentally stimulating. Nice try. You also know that modernization of poultry inspection will delete Union membership by 7-800 dues paying members and that is why they are so vocal in their opposition .

    • Well, and so goes any hope of a rational, thoughtful discussion. Thank you, Doc Raymond.

  • John Munsell

    Katy, call me anytime at 406-234-1877, email is
    I strongly encourage you to bounce my thoughts off some FSIS folks first, to obtain a “balanced” report. Who knows, perhaps I am a liar? John Munsell

  • John Munsell

    Mike, I must remind myself to think what the downstream consequences would be if poultry abattoirs implemented a kill step: namely, the price of chicken would greatly increase. And perhaps we’d have to build more poultry abattoirs because the production in each plant would decrease. My wife & I buy boneless chicken breasts at our local Albertsons, which advertises them every 3 weeks for $ 1.79 – $ 1.99 per lb. Perhaps the best protein buy in the marketplace, realizing that with today’s elevated beef prices, we can easily spend $4/lb on ground beef. If we implement kill steps in poultry plants, I’m confident I’ll be spending $ 2.99 per lb for chicken breasts. We gotta ask ourselves: realizing the tough times faced by many Americans, is it ethically correct for us to require a poultry kill step when it means that a goodly % of our citizens will be priced out of the marketplace? I’m fully convinced that consumers are “comfortable” with existing pathogen loads on meat (most of us don’t get sick), and would prefer to maintain this load rather than pay moderately higher prices at the retail counter. If I am correct, the solution is simply for poultry plants to purchase a much higher level of liability insurance, and accept the current pathogen status quo. John Munsell

    • overseaschinese

      Great discussion. Thoroughly enjoy it.

      On consumer price, I think it’s a very separate issue, because I know that retailers tend to grab the bulk of the margin which directly affects the consumer price significantly. Over here, I’ve heard they take 30% and up to 50-60% margin. If you don’t like it, they’ll just choose another supplier. Even so-called organic and farm-friendly ones seem to demand such large margins!

    • Mike_Mychajlonka_PhD

      John, sorry about the response delay. It seems I am being pulled in several different directions right now. Anyway, I can tell you that here we are paying $6.45/lb. for boneless, skinless chicken breasts (with free Salmonella thrown in). So, I wonder about your numbers. You surmise that the cost of meat will increase 50% with a kill step. I seriously doubt it, at least with the methods I would like to see implemented. Neither can I say how “comfortable” consumers are going to be about an ongoing Salmonella risk. My guess is that, given the choice, they would prefer to not run that risk at all. Problem is, they currently are not being given any choice. The other side of the question is one I raised earlier but was never answered. If one third of a plant’s production can be deemed unfit for human consumption and must be sold cheap, what kind of an economic hit is that going to make to the price of the product that was released? It seems to me that if you are looking for a 50% price increase, there it is, right there. Lastly, what happens to the rejected product? Does Man’s Best Friend get it? If so, can Fido handle it? If not, will the liability insurance cover Fido as well as humans? How much will added insurance costs add to the price? I still think the better approach would be to implement a surface Pasteurization step, deliver a safe product to consumers and let the insurance guy get a real job.

  • Alvin Sewell

    Let’s flow with the discussion. It is not about a Union and Agency clash. It is only about food safety at the consumer level. There are remedies but as long as there is an agenda fed resistance to moving forward, we won’t move forward. The remedies are attainable. This is my position and I feel it is important to focus on forward movement in food safety. There are practical methods. There is opportunity. This is the direction to more toward. It is important to face the issues, head on, with fixing plant sanitation and processing control To deviate from that is a disservice to everyone concerned.

  • Alvin Sewell

    Thanks for your comments Dr. Mychajlonka. Line speed and production quality are inextricably connected. With increased line speeds, processing errors have an impact on a larger volume of product. As a matter of fact, I was assigned as an inspector that was bought by Perdue Farms. The plant was beset with processing problems at 91 birds per minute. Perdue elected to discontinue their “NELS” program and drop back to 70 birds per minute. The result was a more profitable yield. Lower line speeds no only allow for more timely correction of processing issues but give the process momentum. By that I mean that they are less prone to line stoppages, retained product, and other issues that interrupt the flow of processing. Of course, under HIMP, the lines will not be stopping so anything goes. It’s all about line speed and the lack of freedom of plant personnel to address consumer related issues. It’s all about higher production rates at the expense of consumers, worker safety, and good manufacturing practices. These are provable measures.