This is reprinted from the Food & Water Watch blog. I am supposed to be on vacation this week. I have been pulling weeds from my vegetable garden – not using Round Up or Agent Orange or Napalm. Pulling weeds can be cathartic, but an article that appeared in the Aug. 18, 2013, New York Times entitled, “Shipping Continued After Computer Inspection System Failed at Meat Plants,” pulled me away from my peaceful gardening and prompted me to write this blog. I have known about the dysfunctional computer system featured in this article for quite some time and have held back on railing against it, but this article is the last straw. Now I’m going to tell you the rest of the story that the Times left out. I have known that the Public Health Information System (PHIS) has not worked since USDA’s Food Safety Inspection Service (FSIS) turned on the new computer system for its meat and poultry inspectors in April 2011. Billed as an upgrade to the old IT system that inspectors had been using, Web-based PHIS would offer inspectors the ability to record data in the new system even when there was no access to the Internet. PHIS was supposed to consolidate all the databases the agency maintained and provide “real time” observations into each meat and poultry plant that was inspected. Predictive analytics would be used to evaluate data recorded by the inspectors so that potential problems with a plant’s food-safety practices could be flagged and fed back to the inspectors so that they could take action to prevent contaminated products from going into commerce. In 2010, FSIS inspectors asked to test the new information technology system warned the agency that the system was not ready for prime time. There were literally thousands of issues that they brought to their supervisors’ attention about flaws in the new system. While agency management promised to correct the problems, many of them remain today. Even though they had been warned, management decided to plow ahead and turned the program on in a select group of plants in April 2011. The system immediately blew up. This past April, I decided to attend a conference of FSIS inspectors held in Branson, Mo., in order to get more insights into PHIS and maybe even take in a show. There were more than 100 inspectors who attended, and they were from all of the Midwest states. These inspectors worked in assignments where the PHIS system had been in operation the longest. I told the inspectors that I would be available to listen to their experiences with PHIS, and a table was set up for me in the hallway outside the conference room. I sat out in that hallway for a total of 18 hours and took more than 21 pages of notes. These inspectors told me about software issues that plagued the system: the chronic locking-up of the system forcing them to reboot; the frequent outages in the system that led to a loss of data that had been transmitted; and their inability to record data when there was no access to the Internet. The most troubling comment I heard was that on some occasions, inspectors were so tied up trying to get the system to work that they weren’t able to set foot on plant floors to conduct inspections. I had to laugh when I read in The New York Times story that agency management has blamed lack of broadband access in rural areas as the major problem with PHIS. One of the inspectors whom I interviewed in April was based in St. Louis. That inspector reported chronic problems with establishing connectivity. Not only is St. Louis a significant city, but St. Louis is the location of one of the national server banks for PHIS. Needless to say, I never got to see any shows in Branson. In May, I tried to inform agency management of what I found out at the inspectors conference. I told them that none of the data could be trusted in PHIS. The agency administrator, Alfred Almanza, walked out of the meeting. The truth hurts. The New York Times story left out some facts about PHIS that need to be exposed. The article claimed that the system has cost $20 million to implement. According to the White House Office of Management and Budget’s IT Dashboard, the total projected cost for PHIS has been pegged at $141.48 million. The Government Accountability Office has identified PHIS as one of the federal government’s “troubled IT projects”. But the most important fact the article left out was that PHIS can perform magic. On June 28, 2013, the management of FSIS issued instructions to inspectors assigned to horse slaughter facilities. Regardless of where you stand on the issue of slaughtering horses for meat for human consumption, I think that everyone should be astounded when reading this section of the instructions:

Until the equine class is available in PHIS, unless directed by the DO (district office) otherwise, IPP (inspection program personnel) are to verify that the establishment profile includes the slaughter class “GOAT” and enter equine data in PHIS using the goat slaughter class. If the establishment profile does not include the goat slaughter class, IPP are to add “GOAT” slaughter class to the plant profile. NOTE: “GOAT” is being used at this time in order to capture necessary information in PHIS relative to equine. FSIS will manage PHIS results in a manner to discern goat data separately from equine data until such time that PHIS is modified to accommodate equine data entry. FSIS will rely upon the grant of inspection to discern which establishments in PHIS slaughter goat versus equine.

Here I was concerned this summer that the Food and Drug Administration (FDA) was going to approve genetically engineered salmon when PHIS is turning horses into goats! The New York Times story is just one of many revelations about the mismanagement of FSIS that has come out in recent months. The USDA Office of Inspector General (OIG) has issued a series of very scathing reports that call into question whether the management of that agency has a clue. The most recent audit report entitled, “FSIS’ and AMS’ Field Level Workforce Challenges” released on Aug. 6, 2013, made the following observations:

Despite the argument that overworked employees are more likely to commit errors, some FSIS inspectors are working many hours above a normal 80 hour per two-week pay period—more than 400 of FSIS’ approximately 10,000 inspectors averaged more than 120 hours each pay period for the entire FY 2012. Our analysis showed that 1 inspector averaged 179 hours, 3 inspectors averaged over 160 hours, and 14 averaged over 150 hours. When OIG brought this issue to the attention of FSIS officials, they stated that they were unaware of this fact, and doubted that this extended overtime would negatively affect the agency’s inspectors. In addition, due to FSIS’ outdated systems that require manual data entry processes, FSIS cannot efficiently reconcile the hours of overtime billed to industry to the overtime hours recorded in its timekeeping system. Officials explained that, although FSIS has set limits on the number of hours an inspector can work in one day, FSIS has not limited inspectors working overtime hours for extended periods of time. OIG maintains that overworked FSIS inspectors may be risking their own and the public’s health, especially if they are tired or fatigued while performing crucial food safety-related tasks. Additionally, industry should be properly billed for inspection services performed during overtime hours. According to the Occupational Health and Safety Administration, extended or unusual work shifts may be stressful physically, mentally, and emotionally. These effects lead to an increased risk of operator error, injuries, or accidents. Federal regulations state that Departments, such as USDA, shall schedule the basic work week so as to consist of five consecutive 8-hour days, although the Department may depart from the basic work week in those cases where maintaining such a schedule would seriously handicap the Department in carrying out its function. Reducing Excess Overtime Many studies, including those detailed in a report by the Centers for Disease Control and Prevention, have shown that not only is working long hours detrimental to the health and well-being of employees, but that it decreases employee productivity while on the job. The FSIS union contract stipulates that field inspectors are generally not to work more than 10 or 12 hours in one day, depending on their duties. However, we found that some inspectors are working these hours six and even seven days a week. Because of these extended hours, OIG believes FSIS inspectors could have decreased productivity, which might impair their ability to perform functions that are critical to public food safety … When we spoke to FSIS officials about the long hours some of their inspectors are working, they stated that they were not aware inspectors were working such long hours each pay period. While the FSIS officials disagreed that the hours were affecting their field staff’s work, they stated that they needed to better understand the effects of these long hours on their employees. If the results of their inquiry into this issue, based on our audit work, shows that employee fatigue was contributing to worker problems, they would be willing to make changes.

There is something drastically wrong going on at that agency. I am not one who casually calls for the resignation or the termination of people, but I think the time has come for the systematic cleaning out of the management of FSIS – and that includes the Under Secretary for Food Safety, the Administrator of FSIS, the Deputy Administrator, and all of the Assistant Administrators. The safety of the food supply is being compromised by their lackadaisical approach to managing that agency. I do this reluctantly, but enough is enough.

  • Trish Thompson

    Thank you for this in-depth analysis of a system that is malfunctioning. Until every family has a few backyard chickens and every community has a small beef/miling herd, we are doomed to eating unsafe, unhealthy food that our government not only promotes, but promulgates.

    • Martha Jacks

      That is the silliest thing I’ve read all day.

  • Mike_Mychajlonka_PhD

    Strong stuff! Do you think this problem is of recent vintage or do you suspect it has been a problem for many years?

  • Excellent writing.

    I’ve been in the IT business for 26 years. I worked on a massive system for the government at Boeing. There’s nothing in this writing I don’t believe.

    What’s sad, though, is some folks will say “See? That’s why we need small government. Let’s close it down!”

    No, we don’t need a smaller government. We need a better government. We need to make things work, and sometimes that means dealing with people in positions of authority who have no right to be there.

    I have no doubts the people in charge on this computer system are people who know nothing about how a system like this should be built. And they are the same people who refuse to acknowledge any error, any problem. Even when they stay in the rooms, they’re not really there.

  • Tired of exaggeration

    FSIS is working in the same fashion any other workplace would be. Implementation of computerized system has its own challenges and drawbacks. Only difference is that they are a public organization and have to defend themselves against everyone including those who do not understand the situation but have comments.
    They could have classified the equine meat as unicorn if that class was available in the system. Re-classifying a captured data is much easier than to enter the data at a later time not to mention falsification of data.
    Obviously the Alfred left the meeting because of the tone of writer. Alfred left because truth hurts and truth was that he was facing an individual whose job and well being depended on making false claims and creating big stories from nothing.
    Half the projects that Google launches fails regardless of honest attempts and ingenuity of the idea. Google reader is a classic example of it. IT is not as straight forward as one might think. There are compatibility issues between hardware and software, bugs, user knowledge, training and so on.
    Story on NY times was half baked and FSIS explained that inspectors knew how to work without computerized system. Just because they did not log the work in computer does not mean they did not log on paper or worse yet did not do the job.

  • Rotten still

    I personally can speak that in the Cincinnati, Ohio area the inspectors work 12 hour days 7days a week almost every pay period. We have plants in this area that rarely stop for anything. There are approximately 6 plants that have 2 shifts that work 7 days a week 12 hour days everyday.
    Our second shift only has two inspectors and those assignments are big assignments and are part of this workload. This article speaks of physical, mental and emotional tiredness believe it is such a true statement.

  • Jan

    This article is a joke. Go get your food from China if you think American food is tainted!

  • Doc

    If inspectors are not going into the plant because they’re dealing with PHIS, that’s a failure of time management and priority setting. Computer won’t log on? Walk away, get your inspection tasks done and use pen and paper to write the results, let the supervisor know what’s going on, and get the info into PHIS later. Inspection happened long before computers came into play so it is way too far a jump to say PHIS failures would hinder the Agency’s ability to preform its main function. To exaggerate like this for readers that may not know how things actually work at the plant level is not responsible.

    • jim

      Exactly right Doc, I as a CSI usually write down my tasks for the whole week on paper, I being one of those mentioned who works 120hrs a pay period on average has never had to skip tasks based on PHIS being down, even if PHIS is down and I didn’t have my tasks written down then I know to perform the higher priority tasks that day and can always reschedule ones I didn’t get done. This article and others like it are misleading, PHIS has had some growing pains but as far as public safety is concerned whether PHIS is up and running or not has no bearing on food safety.

  • oldcowvet

    enjoyed the read, almost as good as the Onion. My buddy from practice describe the good chuckle he had with his inspection team. They use it every day. When it fails to work on occasion, they just go back to work, let the gremlin work though the system, and then enter the data. He points out that is not flawless, but mucks up about as often as every other government run system. He really gets a hoot out of people who have not used the system pretending to be all knowing about it. THanks for the great laff.

  • Rena

    I usually enjoy reading but this article really grinds my gears. As a food microbiologist and current veterinary student, I aspire to be an FSIS inspector in three years when I graduate. During my internship I did realize they are short staffed, but not nearly to the extent that this article claims. And remember budget cuts that effected FSIS a few years back? Yeah, blame the big boys in the white house for that, not FSIS. I do think that budget cuts to our food safety is not only bad for our country but to others we export to. As far as PHIS goes. Entering the horse slaughter in as “GOAT” is just a computer entry, they’re not turning a horse into a goat. Good try flirting with crazy anti-GMO people on that one though buddy! This article disappointed me, please go back to weeding your garden.

  • Fly on the Wall

    Shelly is right. This story would not be so entertaining if I did not have a front row seat to PHIS implementation. You cannot roll out a new IT system like PHIS when power grabs and egos trump getting it right and then you leave it to the staffers to figure out how to make it work. The bosses did not include horses when they designed PHIS because slaughtering horses is not popular with the animal right lobby; not because they forgot that horses are an amenable species. The staffers turned horses into goats after Congress removed the roadblock to horse slaughter and the bosses told them to figure out a way field personnel to enter horse slaughter data into a system not designed to accept horse slaughter data and redesigning was not an option. I guess paper records are too low tech.

  • Marbles

    Doc, Do you know how it works? Tell them each plant has about 98 to 105 tasks in PHIS that have to be done a month. A month has about 20 to 22 working days in it. Now add the inspector has maybe 3 to 4 plants in his/her assignment to cover every day, .remember you can’t enter anything on the computer during non official hours (when your off duty) so you have a system that don’t work and connection that the DC people would refuse to work with.

    Jim, Your me hero. If your working 120 hours a pay period and have time to write everything down then you’re not looking at your plants. Tell that line to the people who don’t know the job.

    Jan, where do you think a lot of the food is coming from?

    Rena, Come on now. Isn’t your schooling being paid by the USDA if you stay on 3 years? What’s the saying “don’t bite the hand that feeds you or pays your bills with our taxes”?

    • ChknChkr

      It does not take all day to write down tasks i write them down as well because what does take a lot of time is booting up a computer at each plant in the patrol so writing down taks does not keep you from performing your job

  • quickfixfsis

    Yes, PHIS has marked challenges largely because of poor oversight and strategic planning from Almanza and his management team. They know of the difficulties and yet the same core team is attempting to bring the dream to fruition. This cannot happen.
    Inspectors are not recording all of their inspection work. PHIS data does not fully represent inspection activities and yet, policy is structured around PHIS data. Policy based on poor data begets poor policy.
    PHHRS (pay-for-performance), brought to FSIS by Almanza, failed under his guidance as he was incapable of recognizing that PHHRS required a full cast of capable supervisors. Almanza reports that at the end of the 5 year PHHRS pilot that he is going to “kill it”. Sorry, it was dead on arrival. How much money and human resources were lost due to Almanza’s experiment with PHHRS?

  • Butterfly123

    Unless they have hands on experience with working in the fields, The big guys in office will continue to make assumptions and their judgements will continue to affect the safeness of this country’s food. We work to many long hours and not getting any raises. I would love for the people in office to work for 2 weeks 12 hour shifts they would probably quit. SMH. Please get some hands on experience before making decisions for us.

  • Jurgis

    There is a big difference between FSIS inspectors at one plant slaughter establishments and inspectors that cover processing plants on a patrol. With a half dozen or so tasks at up to six different plants, plus travel time, and setup and login time on antiquated hardware (2 Gig internet cards!), the time spent performing detailed inspection is necessarily limited. Official policy is that this is a “real time” system and paper notes were initially prohibitted. Now FSIS maintains that this is an “after the fact” monitoring system and data should be entered from scraps of paper and best recollections. If this data system is not necessary to protect public health, why are we wasting time and money on implementing it? If it is used to protect public health than system failures are a vector for a public health problem and missing/incomplete or compromised data must be acknowledged by FSIS. If a struggling data system isn’t endangering the public health, sweeping the issue under the rug certainly is.

  • FormerFSISAnalyst

    FSIS inspectors love to milk overtime hours.

  • Skillet

    The whole PHIS system is flawed. Its merely a means for swelled heads in D.C. to see tasks performed to justify the flawed PHIS program they mandated.