FDA Warning Letters This Week 2/22/17

warning-letters

The week goes by and new letters come out. Looks like we’ve only got one food letter this week so let’s get into it.

Don’t know what warning letters are? Check out this post for a brief overview of what they are and why FDA sends them out.

WL# 17-PHI-04

CDC Public Health Image library #2287 Credit Elizabeth White (2002)
CDC Public Health Image library #2287 Credit Elizabeth White (2002)

Subject: CGMP/Food/Prepared, Packed or Held Under Insanitary Conditions/Adulterated/L. monocytogenes

Ohhhhh it’s an L. mono letter, always interesting. Already we have to pause for some background.

At the end of 2013, CDC and FDA partnered up to stop foodborne outbreaks of Listeria monocytogenes through the utilization of whole genome sequencing (WGS). This technology used to be effectively impossible 30 years ago, and prohibitively expensive and time consuming even 10 years ago.

What this means is that when L. mono is found in foods or a sick person in the hospital, they can sequence it’s entire genome to determine how closely it might be related to other L. mono cultures found in the network. The network consists of genome databases Pulsenet, GenomeTrakr, and The Listeria Initiative, which are jointly managed by CDC, FDA, and local health departments.

Once L. mono has been found and sequenced, in tandem with a traditional epidemiological investigation (interviews, additional sampling of products and environments, illness reporting), the sequence is compared to those in the database to determine if the strain that was found is “closely related” to others and see if there is a link. This is how the listeria found in the Blue Bell products/plant was traced to 10 illnesses that occurred as early in 2010, even when the listeria in the ice cream wasn’t found until 2015. Across several states and 5 years no firm link was established until WGS sequencing revealed the relationship.

There’s some debate about how transparent CDC is about determining when related isolates are causal, or to what level this data should be considered definitive in the absence of confirmed positives in products. However, FDA is going full steam ahead with using WGS as a new enforcement and Epi tool, and CDC has some good data to support the effect on outbreaks it may have had. Whether that’s because manufacturers are more careful in light of the enforcement activities or because the enforcement activities have prevented outbreaks is unclear. But no matter the end result, it’s a win for public health.

The last bit of debate is as to the extent that L. mono can be considered an adulterant. Here in the U.S. we’ve made the determination that if there is any present in the food, it’s adulterated. This is different from several other countries, including Australia/New Zealand and the UK, which allow a non-zero legal limit if L. mono is not expected to be able to grow in the product over it’s shelf life. There’s debate over that as well.

So back to the warning letter:

FDA’s laboratory analysis of fifty environmental swabs collected on September 12, 2016, confirmed that eighteen of the fifty environmental swabs were positive for L. monocytogenes. Of most importance:

Three positive environmental swabs were collected from the following direct food contact surfaces in your cheese processing room during the production of your RTE cheeses:
  • The top of the cheese slicer
  • The cheese slicer string
  • The inside of a plastic crate used to store finished cheese before packaging
–    The remaining fifteen positive environmental swabs were collected from locations adjacent to food contact surfaces and from non-direct food contact surfaces.
–    One positive sample was collected from your RTE feta cheese

They also found a positive in unpasteurized cheese during the facility visit.

This is a pretty solid connection given they found Listeria in the product itself, especially a post-pasteurization product, and that they found it on product contact surfaces where it could continuously inoculate that pasteurized product.  There are other warning letters that claim the products are adulterated when “we found it in the threshold of your entryway to the plant” that don’t always feel solid, but this one is not one of them.

On September 12, 2016, the lift arm and bowl support brackets of the mixer contained areas which appeared to be rusted and contained rough surfaces. In addition, the beater shaft housing area of the mixer, directly above the bowl support brackets, was observed to contain areas which appeared to be rusted and contain food particulates and/or foreign matter.
Food processing tools like my hand mixer have hard to clean areas that can collect food particles like cookie dough, you’ve got to check those areas every time!

Rust happens, and plants have to do regular walkthroughs and replace what needs to be replaced as it ages. But even with the best of intentions, not everything always gets replaced by the time FDA walks through. This makes it a common violation (do you have any rusty or damaged bowls or pots in your cupboard that you plan to replace soon?). But the food particulates are a good find and an indicator of a less-than-robust sanitation standard. Small, hard-to-clean areas around bolts and inside moving parts like the underside of a kitchen-aid (see picture to left) don’t get properly inspected and cleaned unless you have a dedicated program to seek these areas out.

The floors in the processing room and walk-in cooler were observed to be in disrepair, containing areas where the concrete is cracked, rough, and peeling

This one is also common, as mentioned on this site before, cracked/pitted/porous floors are a potential point where environmental pathogens like listeria and salmonella can hide from getting properly cleaned, waiting to jump back out. It’s also a common finding because refinishing or extensive repair of floors is expensive, normally planned on a long-term basis (maybe the slow season so the plant can close or when the contractor has availability), and because in other manufacturing industries it isn’t always a problem (e.g. machine shops).

We have reviewed your written responses to the Form FDA-483, received October 24, 2016, and December 7, 2016…We will ascertain the adequacy of your corrective actions during our next inspection.

Also as discussed here in the past, this is a good sign that FDA knows this company is taking the findings seriously and making real change. Otherwise they would respond with “this is inadequate” and request additional action or proof of change before closing the letter.

Greater than twenty flies landing on the floor, food processing equipment, food processing utensils, and other food contact surfaces and non-food contact surfaces…Three fly catcher tapes containing multiple flies hanging in different areas of the processing room… Dead flies on the window sills near the batch pasteurizer and three bay sink areas.

Well, they had fly catchers, so they know that flies aren’t supposed to just keep flying around, so there’s that. The manufacturer responded that they would purchase a fly zapper and new tapes, but FDA wasn’t satisfied. With pest control, you’re supposed to keep them out of the building and away from your products. So your interventions need to prevent access (find access points, block holes and unscreened windows, create breezeways and air curtains, etc.), and keep them from getting to products with interventions like ILTs.

Your maintenance of the grounds is inadequate to protect against contamination of food, as required by 21 CFR 110.20(a)… Live chickens and pigs coming within approximately one foot of the main door to the production facility and what appeared to be remnants of dead chickens and goats in close proximity to the production facility…Multiple items within approximately twenty feet of the outside perimeter to your production facility which may constitute an attractant, breeding place, and  harborage areas for pests, including, but not limited to, a chicken coop, an abandoned truck, a small four-wheeled loader, wood paneling, vegetation over six feet tall, and other small items which appear to be refuse.

Yep, sounds like a small farm/dairy. I can just picture it, can’t you? Some animals wandering, tall grass, chicken coop, and a couple old trucks and tractors getting overtaken by weeds

Outside of the animal carcasses (obvious pest and disease attractant), this is FDA throwing a little muscle at this farm to clarify that they need to treat it as a food processing facility and not as just a farm. The letter indicates that the company took action to clean up all of these items, however FDA was unsatisfied because they apparently didn’t send them pictures of the cleaned up areas nor discuss how they would keep livestock from hovering around the entrances of the food plant. FDA again noted that they would confirm the changes made were effective at the next inspection, continuing to show that that the inspection itself and initial 483 response must have gone well and the company is doing the right things post-inspection.

Sounds like this company has some work left to do, and that in this case FDA found some tangible and realistic findings of sanitation and facility problems. It helps that the findings were also supported by the L. mono data to really drive home to the company why they should be doing these things. Always nice when cause and effect tie together with GMP’s to drive positive food safety changes home.

New FDA 483: Euthanasia drug found in canned dog food, misc. facility findings, and questions for FDA

fda-483-footerI would be remiss in the goal of this blog if I didn’t do some digging into the form 483 that was just released by FDA this week following a recall for canned dog food containing Pentobarbital. For information on the products recalled and company involved check out the FDA recalls page and search for the issue. As usual I’ll refrain from writing company and product names on this blog when there isn’t any pending civil or criminal action associated with an event. But that information is readily available for anyone by clicking through the links or performing a simple search.

The 483 is short, just two pages. What the goal of this post will be is to go over each of the observations and try to provide additional information that isn’t included in the document to hopefully provide a complete picture.

All FDA observations began with the heading clarifying which portion of the law (FDCA) the firm violated:

The following observations were found to be adulterated [sic] under the Federal Food, Drug, and Cosmetic Act: A food shall be deemed to be adulterated if it bears or contains any added poisonous or added deleterious substance that is unsafe within the meaning of section 402.

This wouldn’t be FF&F if I didn’t pause here for some definitions. Adulterated is a condition of food by which it cannot be sold in commerce. It includes both reasons of safety as this case demonstrates, but it could also be forms of “economic adulteration”, where something claims to be what it isn’t or has otherwise been robbed of characteristics that the consumer would expect. Like if I were to sell you caviar but it was actually flavored gelatin balls or something.

Poisonous or added deleterious substance is a substance that when added to food “may render it injurious to health; but in case the substance is not an added substance such food shall not be considered adulterated under this clause if the quantity of such substance in such food does not ordinarily render it injurious to health.” (emphasis mine)

FDA says two things there. First, don’t add anything poisonous to food (Protip). Second, if the food happens to contain something poisonous naturally, you need to make sure it occurs at a level where it isn’t toxic. This is the often cited”dose makes poison” principal. An example would be that I can’t sell food into which I accidentally spilled some cyanide (whoops) no matter how much or little it was, but I can sell fruits that may have trace amounts of cyanide precursors in the seeds, because it’s not expected to cause an issue in both the actual dose of the seeds and the expectation that people will avoid them when eating. This clarification is actually pretty critical as we try to make sense of past FDA guidance in this case and why the food was adulterated.

So, how did these dog food products cause themselves to be adulterated?

Your low-acid canned dog food product…was found by chemical analysis to contain the barbiturate drug pentobarbital.

By Harbin (Own work) [Public domain], via Wikimedia Commons
By Harbin (Own work) [Public domain], via Wikimedia Commons
Pentobarbital is a sedative that in the form sodium pentobarbital is used as a euthanasia drug. This recall/483 event was initiated when 5 dogs became sick and one subsequently died. Several new updates have occurred since then and I encourage you to follow the story on a site like food safety news.

Here’s the thing about this finding, it’s annoying that the 483 made no mention of the dose that was recovered. This is important because FDA did a study on pentobarbital in dog food in 2002. In that study, the samples (not randomized/representative, convenience samples selected for likely positives) tested positive for the presence of pentobarbital in more or less than 50% of the samples. However, in the same study FDA made a determination of dose that caused adverse effects:

Based on the data from this study, CVM scientists were able to determine that the no-observable-effect level – which is the highest dose at which no effects of treatment were found – for pentobarbital was 50 micrograms of pentobarbital per day

Dogs would have to consume 5-10 micrograms of pentobarbitol per Kg body weight to hit that dose. The highest value FDA found in their samples was 32ppb (32 micrograms per Kg of food). This means that 7 Kg (15.4 lb) dog would need to eat between 35-70 micrograms to reach the minimum dose, which would have been a little over 1Kg of the highest testing food. Pet food isn’t very dense (canned pet food is denser but contains more water that dilutes other ingredients) and 2.2 lbs of it is a lot of food for a 7Kg dog. Therefore FDA concluded:

the results of the assessment led CVM to conclude that it is highly unlikely a dog consuming dry dog food will experience any adverse effects from exposures to the low levels of pentobarbital found in CVM’s dog food surveys

Which means that FDA concluded that the mere presence of pentobarbital does not make the product adulterated because “the quantity of such substance in such food does not ordinarily render it injurious to health” per the FDA study.

Now, because there is report of adverse events and an Oregon State College of Veterinary Medicine report out there showing that the levels in this food were high enough to cause an effect, this food is clearly adulterated. But it seems like FDA should have included a note about the concentration of the drug found in the food in this 483 to clarify why it was legally adulterated, given the past study.

Now for the findings not related to the chemistry analysis and recall:

Condensate dripped throughout your processing facility from the building…including condensate dripping directly into open cans of the in-process low-acid canned dog food product…and also into multiple open totes of raw meats including beef intended for your canned dog food product

steam-hood
Example: steam hood over my stove that I apparently need to clean…gross.

Condensate is found wherever foods are heated and cooled, and FDA has been addressing it more and more. Condensate was noted in the Blue Bell 483’s as well. The logic is that while steam or vapor may be clean, once it collects on a surface like the ceiling or whatever else, it can carry bacteria from these “non product contact” areas back onto your food. Think of it this way, would you lick the underside of the steam hood/vent above the stove if you hadn’t just cleaned it? Now imagine that the steam from your stroganoff was condensing on the underside of the hood and dripping back into it, carrying all that old grease and dust. Yum.

The floors throughout your processing facility are pitted, cracked, and otherwise damaged causing pooled water in areas where food is exposed including where open cans of…dog food are staged

pitted-concrete
Source: my patio.

Uncleanable floors = environmental pathogens. While they didn’t go on a “swab-a-thon” in this facility (yet), uncleanable floors are essentially considered harborage points for things like Listeria and Salmonella. In any other business than food, pitted floors aren’t normally an issue, which makes it a common finding in plants holding themselves to a manufacturing efficiency standard rather than a “food grade” standard.

Additional sanitary conditions observed…include peeling paint and mold on walls throughout the processing facility including in areas where food is exposed, a live fly-like insect in the …hand-packing area during processing, and an open sanitary sewer within approximately 25 feet of two food storage trailers and one food processing trailer at the rear exterior of the facility.

Really just shows a lack of preventative maintenance and facility investement when there wasn’t a clear ROI. This particular company has been in business for a long time in the same location, so it’s possible they themselves put that old coat of paint in years ago to spiff it up and make it look nice and be good for food. These kinds of things are expensive preventative maintenance tasks (mold removal, repainting) because it causes downtime as well as the expense of the repair. Typically FDA will show discretion depending on risk to product (e.g. if you only have closed containers in a room with old paint), but the inspectors here probably determined that this was facility neglect and should be noted. Same thing happens in restaurants and retail establishments where facilities have aged but there’s been no spiffing up.

You lack operating refrigerated storage facilities or other means of controlling the temperature exposure of raw meats during thawing, storage, and processing.

Ding, ding, ding! We have a winner, here’s where we demonstrate the true lack of food safety commitment/appreciation at this facility. The last findings all relate to proper temperature control:

…raw beef and other meats in various stages of thawing were stored in ambient temperature inside your processing facility and also at abmient temperature inside three trailers…the exterior ambient temperatures were below freezing…there was frozen ice containing a blood-like substance across the floors of the three trailers and also on the ground…

Open cans of beef were staged on a pallet at ambient temperature during the hand packing process [from the start of operations until 2:00 PM]

So here’s what the deal is with food safety here. This product is going to be retorted, which means that as a low-acid product, it’s going to be cooked until it’s commercially sterile.

So, in theory, it doesn’t matter if your raw meat doesn’t stay refrigerated, since you’ll kill anything that might grow on it! Heck, you can pack it in a dirty facility with dirty tools if you wanted to…

That was sarcasm.

Processors who think like this fail to understand how cooking and kill steps work, and don’t have respect for your food at all stages of production.

FDA expects the thermal process for low-acid foods to provide a minimum of a 5 or 7-log reduction for spores and pertinent pathogens. What this means is that the process should destroy a minimum 99.999% of spores/bacteria in the product, or alternatively, it would sterilize meat that contained 10,000 spores/gram (bacteria are easier to kill than spores, and would have a much higher log reduction with the same process).

This would work for most “raw” products used in this process. However, if you don’t refrigerate or otherwise control raw meat to keep it out of the danger zone of 40-140ºF, bacteria will start to grow. And with the average piece of beef trim having anywhere from 100 to 100,000 bacteria/gram, if these bacteria are allowed to multiply to the ten-millions from lack of refrigeration suddenly that 5-log reduction doesn’t work anymore!

99.999% of 10,000,000=100

While 100 un-killed spores may not seem like much, one of them could be C. botulinum, and with a shelf life of years in a can of dog food, it only has time to grow.

Take this home: every cook or “kill” step in food processing has a log-reduction value. So while you can technically cook spoiled meat until all the bacteria are dead, you have no way of knowing (without testing) that your standard procedure of cook until 165ºf will work if the number of bacteria are 100 fold higher than what the cook was intended for.

If you still think you can throw away your refrigerator and just cook everything through, I recommend purchasing an autoclave to really sock-it-to-em. Don’t think what comes out will be very tasty though. Oh, and general autoclave parameters will give you a 12-log reduction. Happy cooking.

While this is a significant finding, it isn’t related to the issue causing the current recall (and subsequent enforcement). The issue with the product had to do with pentobarbital in the food, which is a supplier sourcing issue (pentobarbital didn’t make it’s way in at the plant unless it was a malicious act). This plant has had a poor history with supplier approval (sourcing duck that wasn’t actually duck for example), and also has a history of being ignored by the FDA based on inspection history.

What this warning letter serves to do is show that FDA is doing it’s job (or backtracking to do so) enforcing all the regs at this plant regardless of the specifics of the current problem. But I have a lingering problem with this timeline:

12/31/17: Dogs become sick after eating the implicated food.

1/3/17: Oregon State University receives the samples for autopsy and analysis, report indicates FDA was informed.

1/10/17: FDA shows up at the plant to perform inspection that led to the facility 483 findings

1/17/17: Michigan State University confirms Pentobarbital contamination

2/1-2/2/17: FDA continues inspection according to the 483, no new findings noted from the later dates

2/3/17: Recall initiated, presumably this was a result of the meeting with FDA from the previous two days where they informed them of the results and helped identify the scope of the recall and “recommended” a “voluntary” recall.

2/8/17: FDA continues inspection according to the 483, no new findings noted from later dates.

2/17/17: FDA releases their own independent press release through CVM updates

This facility had multiple problems in 2011 and 2012 that led to FDA action, and FDA had last interacted with them (according to the inspections database, which does not include contracted inspections through the state) on 2/28/13.

Did FDA inspect a facility, find problems, and then decide not to go back for 4 years? And from this timeline above, did they only go back to this facility because they had a potential poisoning related to it on file?

Thorough and rapid response to a crisis FDA, good job! But shouldn’t you have been inspecting a known problem facility to prevent problems like this from happening?

After all, in 2011, you said this:

The FDA Food Safety Modernization Act (FSMA), signed into law by President Obama on Jan. 4, enables FDA to better protect public health by strengthening the food safety system. It enables FDA to focus more on preventing food safety problems rather than relying primarily on reacting to problems after they occur.

We can’t say whether increased visits from FDA (which should have been every 3 years at minimum) would have prevented this from happening. But it certainly couldn’t have hurt.

USDA-FSIS Quarterly Enforcement Report: Downloadable Raw Data and 10 Year Trends

data-graphicIn my mass of emails from the government each day, every quarter I get one from FSIS letting me know what they’ve released their quarterly enforcement report. These reports detail actions taken against specific establishments, as well as details on how many enforcement-type actions are taking place ongoing.

I have a problem with this report however. It’s presented as a pdf with no metrics or historical data to put the information in context. FDA has done a good job in the past few years at releasing metadata in formats that allow analysis (CFSAN adverse event reports, inspection database, etc.), but USDA takes the time to kick out this report without showing what the trends are.

Alright raw data nerds, here at FF&F we’ve got your back. For all your analysis needs, we’ve collected a few of the more “top level” reporting numbers from reports going back to 2007, and put them all in excel for you to strip apart and analyze to your hearts content. We’ll even keep updating this spreadsheet and re-uploading every quarter ongoing.

fsis enforcement reports data (.xlsx)

Because some of us like bringing home a growler on a Friday night and entering data to do basic trending on government statistics, here are some of the trends I found interesting that aren’t visible by looking at any one of the individual reports.

verification-procedures-performed

These are the number of individual verification procedures performed by FSIS inspectors each quarter. We can’t speak as to why there was such a change in 2010-2011, as these numbers could simply have started being calculated differently in the system. But NPR had some data to suggest that overall meat consumption started declining in 2010, which could lead to fewer establishments or approved inspection hours, and thus fewer verification events.

compliance-rate

Compliance rates have always remained above 98% (meaning that 98% of all verification activities end with an inspector going “okay, you’re doing the right thing here”). But like overall verification, compliance seems to have a slight downward trend as well. It’s hard to say this decline is significant since the standard deviation of these percentage data is only 0.0016, but the whole point of getting this data organized to demonstrate the trend is to keep tabs on things like this. It’s hard to never have a downward trend when you hover near 100%, but it will be interesting to see if there is any FSMA effect in the next few years on this trend in either direction.

percent-of-nr-appeals-that-were-granted

That 2% noncompliance rate is a count of Noncompliance Records (NR’s) that are issued. Per FSIS:

An NR is a written record that documents noncompliance with FSIS regulations. An NR notifies the establishment of the noncompliance and that it should take action to remedy the situation and prevent its recurrence. Noncompliance reported on NRs varies from non-food safety issues to serious breakdowns in food safety controls.

When an establishment fails a verification activity, it is issued an NR and has the option to appeal if they think an observation was out of context, the inspector was incorrect in his/her interpretation of the regulation, or some other circumstance led them to believe the finding was incorrect. The graph above shows that historically about 1/3 appeals are granted by regional offices/DC when establishments challenge an NR. But in recent years that number has dropped from around 35% to 30%. This could have also contributed to the increased number of noncompliances observed earlier, now that fewer appeals are being granted.

percent-livestock-carcasses-condemned-of-those-inspected

The poultry carcass inspection/condemn data didn’t have any interesting trends, it swung up and down without a lot of overall variation. The livestock carcass data was more interesting in that it seems to have a sustained downward trend.  This could be to better herd health in the last decade or more efficient slaughter practices that result in fewer condemned carcasses from injury or contamination. While the cause is uncertain, as long as it isn’t the result of poor or inconsistent enforcement, this is a great trend! Fewer condemned animals is better for the animals and the environment.percent-meat-and-poultry-refused-on-inspection

I’m not sure this paints a completely accurate picture, but the data points were real, they actually wrote zeroes in the older reports. While the number of products imported has increased overall, this % refused trend shows that even though imports are increasing, we’re also rejecting more meat at the border than ever before. You can check out this FSIS directive to learn more about inspection of imported meat and poultry and what refusal criteria might be.

If you’re more intense than I am and want to do advanced statistics that excel can’t handle, you can see what’s out there in terms of software and let me know if you discover any new insights from this data! The spreadsheet and all graphs were created from source data in the FSIS quarterly enforcement reports.

 

 
ResearchBlogging.org

USDA/FSIS (2017). Quarterly Enforcement Report for Quarter 1, Fiscal Year 2017 United States Department of Agriculture
 

FDA Warning Letters this week 2/14/17

I’m back! Check out the About page for more information on the new direction of the blog. Today’s post will be a recurring one. Each week I’ll go through the FDA warning letters recently posted, read through the ones related to food, and offer a summary of any interesting findings and/or findings that we as consumers should or shouldn’t be paying attention to.

warning-letters

First, because this blog is about education and communication, let’s help out those who don’t know what an FDA warning letter is. Per FDA:

When FDA finds that a manufacturer has significantly violated FDA regulations, FDA notifies the manufacturer. This notification is often in the form of a Warning Letter.

The Warning Letter identifies the violation, such as poor manufacturing practices, problems with claims for what a product can do, or incorrect directions for use. The letter also makes clear that the company must correct the problem and provides directions and a timeframe for the company to inform FDA of its plans for correction. FDA then checks to ensure that the company’s corrections are adequate.

This is a good description of what a warning letter is, but when does FDA hand them out?

FDA sees warning letters as a method for encouraging voluntary action to correct problems they found in investigations, after-market product testing and label review, plant inspections, and to increase the pressure on companies that aren’t responding to form 483 findings (inspection violations).

In FDA-speak this means that they’re avoiding forced action that would be supported by a judge or federal officer, such as injunctions or seizing products and facilities. However as far as any reputable food company is concerned, it’s really being voluntold. After all, companies that do not respond to warning letters are almost guaranteed to subsequently receive actual enforcement actions, and because warning letters are public, if a company is in the public space in any large way (e.g. a recognizable brand) getting your warning letter closed is an additional public step to regaining consumer trust.

This is just a brief overview, for more information on FDA decision-making in regards to why a warning letter might be issued you can read their procedures. It also contains a lot of information about “prior notice” and other legal requirements the warning letter satisfies.

FDA warning letters this week

Now let’s get to the goods! In these posts I’ll be using the warning letter numbers to identify specific letters and avoid typing company names simply for the sake of making it about the information in the letter and not the addressee. If you’re curious about which companies received the letters, you can check them out yourself or google the warning letter number, that should get you there.

I picked a doozy of a week to start doing this, there are some extensive and complex letters here, I’ll keep my comments “brief”, but check out the letters for yourself to read the full scope of violations.

WL# 14-17

Subject: Seafood HACCP/CGMP for Foods/Adulterated/Insanitary Conditions

Seafood shows up a lot in warning letters for HACCP violations. Companies and produce Juice and Seafood have specific critical control points and HACCP plans spelled out in the regulations. Oftentimes small companies or importers who have never been inspected, or were never aware of what their requirements were, violate or do not put these programs in place. Because this is considered an imminent threat to food safety, FDA almost always issues a warning letter. Most of what you see are companies that just straight up don’t have the programs in place, rather than specific violations. This warning letter is actually a follow up to a previous one, where FDA reviewed the corrective actions of the company and determined that they weren’t enough.

We received your written response dated January 10, 2017 which included Product Specifications and Importer Written Verification Procedures…However, your product specifications failed to identify the hazards that are specific to the product.

Off the top of my head, I would guess that the person who wrote the specifications in response to the letter has not had formal HACCP training and didn’t understand how to carry out an actual hazard analysis, or where the information could be found in the CFR. Ultimately this warning letter represents a failure of a company both to follow prescribed regulations and understand the value of HACCP and Preventative Food Safety.

If you are interested in HACCP, I’ll probably cover many basic principles throughout this blog, but there are many people much smarter and experienced than me who have written solid books on the subject that can be very helpful for the studious. Otherwise there are a ton of classes and free online materials out there, go find what works for you!

WL# 2017-NOL-03

Subject: CGMP/Food/Prepared, Packed or Held Under Insanitary Conditions/Adulterated

I love seeing this subject on warning letters, because it usually means we get some insight into how FDA enforces current good manufacturing practices (cGMP or GMP). Other times, we just get to hear about gross conditions within a plant. It’s really a mixed bag of findings, but they’re usually interesting in one way or another.

The first set of findings for this letter concerned inspectors finding flies, bugs, and rodent excreta pellets (REP’s, and one of my favorite acronyms ever) in the bakery. Thankfully, the REP’s were found underneath a mop sink in the corner, and not somewhere closer to food (some warning letters are much more nauseating with respect to where they find these things). The flies were found throughout the facility, both on products and equipment as flies in your home would. Not a good start. I imagine this facility has either no pest control program or doesn’t use a professional. Either way the presence of the flies indicates that their presence was “normal” and that this facility needs culture change. Further demonstrated by FDA noting that they found pests in inspections for two previous years (clearly no lesson was learned).

Next time you are in a fast food restaurant, look around and see if you can spot an insect light trap (ILT). Often they’re decorative like the one in the link, and they work super well at immediately attracting bugs that make it through the door (or other openings). Facilities with preventative pest control will have these in areas where they will be most effective, attracting pests away from sensitive areas.

The second set of findings concerned handwashing practices. These weren’t the worst ones you see. The ones that are immediately concerning are when FDA notes things like employees using the restroom or dirty surfaces like pest traps and mop buckets, then going back to touching products. In this plant the inspector took note every time an employee left the area or touched non-food surfaces and returned to food without stopping to wash their hands. This is wrong, and the company again has a culture issue, but remember that doctors only follow proper handwashing 40% percent of the time as well. So while it’s the wrong behavior for handling food, it’s far from abnormal and a constant struggle in all settings.

Additional findings concerned sanitation and handling of food contact surfaces. And…well, these some things that directly inoculate your products with bacteria.

packaging material…was observed in direct contact with the floor soiled with food debris

The baking sheet pans came into direct contact with a trash can during scraping.  The sheet pans were not cleaned or sanitized after contact with the trash can before unbaked 9” pie shells were placed onto the sheet pans.

observations [of accumulated food debris on equipment] were noted after the previous day’s production run and prior to production start-up on each of the aforementioned dates.  Production commenced without removal of the observed food debris.

The rest of the findings show pretty severe deficiencies with the way the company conducts itself, handles products and materials in the facilities (including storing secondary packaging in the employee bathroom…) and in general fails to maintain a sanitary facility for food production. Honestly this warning letter is so chock full of obvious contamination issues it’s frustrating that FDA hasn’t taken action sooner. After all, more than half of the observations are followed by the note:

This is a repeat observation from the 9/20/2010 FDA inspection.

This is where I get frustrated as a member of the industry, a taxpayer, and a consumer. According to FDA’s database of inspections, they’ve visited this facility on the following dates: 4/14/09, 9/23/10, and 6/10/11. According to the database, that’s the last time they’ve been there until the inspection noted in this warning letter (10/18/16). They were also never issued a warning letter for those inspections in 2011. Either they decided the State of TN was in charge of following up at the state level, they contracted FDA inspections through the state of TN in the intervening 5 years, or no one inspected this facility for 5 years after finding all these problems in 2011.

We won’t ever know, but regardless of what happened in the intervening 5 years, we should be grateful FDA is taking action now, and found the facilities response to a 483 issued at the inspection:

…[we] have determined your response is inadequate.  You have not provided evidence such as photographs, receipts, and/or training documentation to show you have corrected any of the violations observed.

Here’s hoping some pretty good evidence was provided in 2011.

WL# 2017-NOL-05

Subject: Seafood HACCP/CGMP for Foods/Adulterated/Insanitary Conditions

This letter is for similar violations as the first one we discussed for failing to adhere to seafood HACCP regulations. However this producer actually has a properly constructed hazard analysis and HACCP plan, but failed to properly execute it.

The firm had a cook step for crab that included a time/temperature combination to ensure that any pathogens would be killed. By reviewing the companies records, FDA observed that 9 different times employees had ended the cooks too early but no notes or corrections were made. This is concerning because it means that product may have been released undercooked, and employees running the equipment didn’t have the training necessary to follow the cook instructions, or understand why ending it prematurely was unacceptable.

The letter indicates that the company and FDA have been working through this issue, FDA indicates that the response has been inadequate for lack of evidence. Protip: if you tell FDA “we modified our HACCP plan to your recommendations”, you have to send them the modified HACCP plan. It’s amazing how many follow-up letters have to tell companies this or similar.

This facility also had sanitation issues, not as severe as the bakery above, but running on dirty equipment undermines any HACCP plan or cooking step you do.

You are not monitoring the cleanliness of food contact surfaces as evidenced by previously cooked crab parts from operations on the previous day and after cleaning operations were observed on the metal conveyor were observed to come into direct contact with newly cooked crabs on the conveyor belt.

Protip: if there’s still pieces of the food from the last time you used it, it isn’t clean.

Among the sanitation violations however, FDA notes that the correction actions the company informed them of are acceptable enough that they will verify at next inspection, rather than simply reply that “response was not adequate”, which is a good sign that FDA has confidence the company will get it together and can be trusted to keep improving without direct supervision. From the letter I get the impression that the company had problems, but didn’t have the longstanding culture and facility issues we saw manifesting in the bakery discussed earlier. After all, they couldn’t demonstrate that the low concentration of chlorine in their sanitizing stations was such a pervasive problem without also noting that employees were actually using the sanitizer stations frequently.

WL# CMS Case # 516352

Subject: Acidified Foods/Prepared Packed or Held Under Insanitary Conditions

Acidified foods are foods in hermetically sealed (airtight) containers that could support the growth of microorganisms (high water activity), but are modified with acidic ingredients to bring their final pH to 4.6 or below. FDA strictly regulates these foods because they have the potential to support the growth of Clostridium botulinum which causes botulism, a truly terrifying disease.

Because of these strict regulations, we often see warning letters related to not following HACCP/regulatory requirements for acidified or low-acid foods, very similarly to those we see for juice and seafood. This producer had a lot of issues that come down to them not following these regulations and controlling the process of production, leading FDA to crack down because they present a risk of growing Clostridium species in their products. Check out the letter for those specifics.

In addition to the acidified foods violations, the firm had some interesting violations connected to glove use. Nitrile gloves are often associated with food safety, but they aren’t magic. A clean hand is better than a dirty glove, and this firm demonstrated this with a few well observed glove observations:

an employee…had a tear on the glove worn on his right hand. The tear extended from the palm to the backside on the hand across the area between the thumb and fore finger.

an employee…was observed picking up a lid which had fallen to the floor and placing it in a sink. The employee did not wash or sanitize his gloves before returning to capping the filled jars of product

The only food safety “magic” that gloves provide is that they can be changed. A glove change and sanitize can save time and provide convenience to employees who do things like pick up objects they dropped on the floor or open doors when they don’t have to walk away to a handwashing sink. Provided that the gloves themselves are also in a convenient nearby location.

And when it comes to enforcing food safety behaviors, convenience=compliance.

 

FDA officially refers consumers to Wikipedia for information on food pathogens

I was perusing the Bad Bug Book while doing some research on the recent Blue Bell outbreak and came across a hyperlink. After hearing “do you want to know more?” in my head, I clicked through on some non-L. mono species of Listeria and was…confused. I quickly doubled back, thinking that maybe I had been redirected, but there it was.

FDA Bad Bug Book linking directly to wikipedia
FDA Bad Bug Book linking directly to Wikipedia

FDA describes the reference as “current information about the major known agents that cause foodborne illness.” Descriptions also include a statement that it should not be used as a comprehensive or clinical reference. However, this isn’t an excuse for making a consumer and industry reference link to a completely uncontrolled document source. The Bad Bug Book (2nd ed.) is a wonderfully written resource, both for a lay and industry audience; but the fact that the authors of the Listeria page referred to Wikipedia as an ongoing resource, without knowing or being able to control the content presented to consumers, is irresponsible. A nefarious Wikipedia troll could at any moment have an article claiming that L. grayi is a GMO herbicide borne bacteria found in bananas that causes uncontrolled crying and hair growth, and have the full support of the FDA behind their article.

Please don’t write that article.

A  currently live example of why this was such a poor decision is that if you click through to some of the pages, they don’t exist (as of 7/27/15). I don’t know if the author intended to write them him/herself and never got around to it, or if they simply assumed the pages existed, and then didn’t bother to review the content. I’m not satisfied with either of those answers, and if alternatively the reference articles were removed at some point, that also highlights what a poor decision those links were.

Given the sheer number of PhD’s involved in the book’s creation, I think taxpayers should expect a resource with material actually reviewed and sanctioned by FDA. The poor editing here is unacceptable and a change should be made to the current edition of the book.

Many of the other pages in the book name multiple related species, but either included links to NIH or CDC or included no link at all, both of which are acceptable alternatives. I won’t name the authors and editors of the book here, anyone who wants to know can find them at the front of the document. If you’re interested in bringing this to FDA’s attention in your own way, they’re on twitter as @US_FDA and additional points of contact are available at http://www.fda.gov.
ResearchBlogging.org

Food and Drug Administration (2012). Listeria Monocytogenes Bad Bug Book, Foodborne pathogenic microorganisms and natural toxins. Second Edition, 99-100

Misinformation and selective coverage change perception of outbreaks, but can be corrected by presenting the facts

While it’s not an animal product, the Listeriosis outbreak recently traced to apples is just as relevant to the food industry as a whole as any other food-borne illness outbreak. While I was looking for more information on the outbreak, I came across this gem* of an article posted on cnn.com.

*When this post was originally written, the text on the website read: “At least seven people have died after eating caramel apples that may have been infected with Listeria monocytogenes. Followed immediately by a quote from CDC which stated ‘Thirty-one ill people have been hospitalized and six deaths have been reported. Listeriosis contributed to three of these deaths, and it is unclear whether it contributed to an additional two deaths. The sixth death was unrelated to listeriosis.'” CNN has since removed the CDC quote, but kept their original ‘7 deaths’ statement.”

I found this disturbing on two levels. First, the fact that they reported that at least six people had died after eating contaminated apples, when listeriosis was only confirmed as cause of death for 3 of the cases and ruled out for the 6th.

“Hey Jen, what’s the body count up to on that outbreak article?”

“Looks like 3 for sure, could be 3-5”

“Thats it?”

“Don’t worry, we’ll round up to 6+, if you use a Log scale, they’re practically the same number.”

Second, they used the direct quote from CDC’s 12/31 update to directly contradict themselves in the following sentence. Who wrote this article? (update, clearly they wizened up and removed the quote on Jan 15, I wonder if they saw the reddit post. This is also a rhetorical question, their name is on the article, but we also need to assign blame to their editor.).

So what sort of impact could this statement have? Young, Norman, and Humphries reviewed the impact of media coverage on how dangerous we think they are. They found that indeed, those conditions/diseases that receive more media coverage are perceived by medical students as a “worse” condition. This can actually be a very good thing for infectious disease outbreaks, as rapid media coverage of the danger encourages people to avoid contact with others, leading to exponentially fewer cases the earlier you do it. This is less good however, when non-infectious diseases or inaccurate correlations are blown out of proportion (e.g. people avoiding pork to avoid H1N1).

The literature review included in the beginning of the article shows that this isn’t necessarily new information. However, the authors also examined the effect or including additional “objective” information about the conditions when asking students to rank their risk. The result was that, as seen in the chart below, when provided additional information the study participants then changed their views of the diseases. The large separation between what they had seen in media and what they had not seen shrank and they assigned more risk to those threats that aren’t often talked about, and became less nervous of the high coverage items in comparison.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0003552As a science blogger, this is my soapbox, as this study highlighted the responsibility for those who know more than the headline to speak up and share their knowledge, because people can and will be receptive to it as long as it’s available.

Unfortunately, the study was inherently biased as medical students are more likely to be receptive to new data (especially related to disease) as opposed to other groups with stronger existing bias’ (e.g. CAM users, anti-vaccination proponents, specialist doctors, or epidemiologists who may be swayed by previous outbreak coverage). The authors specifically did not survey students on their current media usage or biases, and therefore could not demonstrate the power of providing additional information on subjects they may have already formed strong opinions on.

I’d like to see the study repeated with an older group, as student’s opinions are more likely to be malleable as they are less likely to know as much about these illnesses or had personal experiences with them. A repeated study with participants of at least 40-years-old would be more telling and help us understand what effect providing additional objective information can have.

After all, as nice as it is to know students can be taught, they’re not the ones in public office. Are they also willing to change their minds when new information is made available?
Young ME, Norman GR, & Humphreys KR (2008). Medicine in the popular press: the influence of the media on perceptions of disease. PloS one, 3 (10) PMID: 18958167

Mummert A, & Weiss H (2013). Get the news out loudly and quickly: the influence of the media on limiting emerging infectious disease outbreaks. PloS one, 8 (8) PMID: 23990974ResearchBlogging.org

Anatomy of a “Serving”

NFPIs it how much you’re expected to eat? How much you should eat? Pardon me, but if I want to eat an entire bag of potato chips, that’s exactly what I’m going to do.

I say this as the guy who helps create nutrition facts panels for food products for a living. I have read, reread, proof read, and colored red hundreds of these little tables in my time, and believe it or not, people aren’t lining up at parties to hear my thrilling stories.

I know!

Right now you’re thinking, “But Austin, I remember back in 2008 the FDA called 2,584 adults from the US to ask them questions about their diet. And 24% of the respondents said they had no idea if serving sizes were determined by government rules or by manufacturers.”

Your oddly specific observation would be correct, and I should have at least ¼ of the room hovering around me in rapt attention, waiting for me to clarify this confusing point. Time has shown however that everyone is clearly too intimidated to approach and ask the simple questions, even when I’m subtly firing off labeling trivia from the empty cracker box carelessly left by the cheese platter, or establishing my mastery of the dance floor.

Fact: Every party has a dance floor, here’s the label. dance floor Well let’s clear this up right now while I’ve got you at home/work/somewhere, on your computer/phone/tablet (scary that I know where you are, isn’t it?). Who determines serving sizes, manufacturers or the government?

Answer: The government!…ish.

Well that was unsatisfying, but it’s the most accurate answer I can give. Essentially what happens is that our government, via the FDA and FSIS (Food Safety Inspection Service), determine how serving sizes are to be calculated and presented, but also leave manufacturers options in specific situations.

So how are serving sizes determined?

Step 1, what are you eating?

he first step companies need to take when determining serving size is to determine what type of product they are selling. Back in 1993 when they had to decide all of this stuff, FDA determined that they could use data collected in the NHANES dietary surveys conducted in 1978 and 1988 to set these standards. These were nationwide surveys that collected all sorts of data, including nutritional intake and food frequency data. With this information, FDA created “Reference Amounts Customarily Consumed,” or RACC, for different categories of food.

The first step is easy, find the category that a food falls into, and look up the RACC used to determine the serving size in 21 CFR 101.12 (for non-meat items). For example, if I was making mashed potatoes, my category would be “Potatoes and Sweet Potatoes/Yams: Mashed, candied, stuffed, or with sauce” and my RACC would be 140 grams.

Step 2, how can someone measure it out?

Here’s where the variation begins! You might just want to use that 140 grams we saw above; however, not everyone has a scale in their kitchen, and let’s see you try to guess how much mashed potato makes up 140 grams. Can you think of the last time you weighed your food, much less with metric weights (provided you live in the U.S.)?

So at this juncture, the government instructs manufacturers to determine what the closest “common household measurement” to one RACC of your product is. So if we take 140g of our potatoes and see that that’s about 2/3 of a cup, our serving size becomes 2/3 of a cup!

>THIS IS IMPORTANT<

Imagine we made a new, super fluffy mashed potato with more butter, and 140 grams of these potatoes actually wind up closer to ¾ of a cup. This means that even though both potato products were based on a RACC of 140g, they might have two completely different serving sizes, and the manufacturer arrived at each using the same government reference amount!

So there you have it, two serving sizes created based on a government standard, but completely different once observed on the store shelf. How could there be even more variation?

Single Serving

Ah, right. For many products, if the entire container contains less or near 200% of the RACC amount, then there are different rules to play by. In most cases, the product will be considered a single serving, but in others, manufacturers have the choice to label them as one or two servings. This is why you see different types of labeling in small containers such as ice cream, muffins, soda, and other “single serving” containers that appear significantly larger than the usual RACC amount.

As eaten, not as sold

Ah, and this is critical. When you ask someone how much cake they eat, they typically don’t respond with, “about 1/3 of a box of cake mix”. RACC values are based on products as they are consumed. However, serving sizes are based on products as they are sold. The reasoning? Because it would be bizarre to buy a bag of flour and see “two slices of bread” for a serving size. This makes more sense for some products than others, but ultimately serving sizes for products that require further preparation are the amount of packaged product it takes to make about 1 RACC of product as eaten. And remember, this must be rounded off at a common household measure!

Final thoughts

As we realize that our beloved nutrition facts panel is now old enough to drink (enforcement began in ’94), we look back and start to wonder if that data from the 70’s and 80’s used to determine RACC values still holds water. I can’t think of anyone who eats ½ cup of ice cream in a sitting, nor leaves the potato chip bag untouched after their first 10-20 chips.

But how about we think about RACC values in a different way. These values were never intended to be an expectation, but simply a way to bring nutrition information into context using consumer data. The thousands of calories in a 20 lb. bag of rice don’t have a lot of context when I eat it one bowl at a time, but that’s also not to say that I’ll never eat an obscene amount of rice in one sitting just because I’m starving.

Instead think of it this way, if these values are simply references to what we customarily consume at a time, we’ve got a great tool on our hands. I wouldn’t expect you to eat only ½ cup of ice cream, but have you noticed that many ice cream scoops happen to portion about ½ cup of ice cream at a time? And while I’ve been known to turn a bag of kettle fried chips into a meal, I still eat them one handful at a time, which just so happens to contain approximately 10-15 chips.

If only some sort of reference was available so that I could tell about how many calories I ate with each handful…

To learn more about how serving sizes are determined for all food products, check out the labeling and nutrition documents on the FDA website, this PowerPoint provided by the FSIS, or the Guide to Federal Food Labeling Requirements for Meat, Poultry, and Egg Products. Check out what consumer opinions of labels are looking like since 2008 in the FDA’s consumer research.

Choinière C. & Lando A. (2008). 2008 Health and Diet Survey, DOI:

ResearchBlogging.org

Conrad J. Choinière, & Amy Lando (2008). 2008 Health and Diet Survey FDA Consumer Behavior Research Foos Safety Surveys (FSS)