Sunday, February 3, 2013

Food Allergy Cross Contamination: In the Trenches

I want to write a quick follow up to the article about the FARE survey and food allergy thresholds. I assumed people knew there are unlabeled allergens in our food. Based on the article comments, however, that doesn't appear to be common knowledge.

A study done several years back by Sicherer showed that 2% of the samples they tested contained some level of milk, egg, or peanut, even though the foods did not mention them on the label at all. The reality is that, if you use manufactured foods, you have most likely used a contaminated food at some point without knowing it.

Yes, this is technically out of compliance with FALCPA. BUT (and this is an important but), the food actually has to cause a problem before the FDA will take action. That means the allergen has to be at a threshold high enough to cause a reaction, a reaction has to occur, it has to be reported, and the FDA has to decide it's important enough on which to act. Since the process can break down at any of those points, the reality is that not many food recalls happen in the real world. (The FDA is so stressed and food inspections so infrequent that the chances of a company proactively being cited are slim to none.)

We have gotten burned by cross-contaminated foods so many times over the years that I assumed this was a common experience among FA parents. I've always assumed no food is 100% guaranteed to be safe, despite the label. (That's why we used to have our son try a little of everything before we went on camping trips.)

Scratch an "old timer" like me and you'll probably find a bunch of subjective coping strategies to try to avoid cross-contamination in foods. Not all of them probably make total sense. But, I'm going to note them here because they can at least add to the discussion about how very bad U.S. food labeling laws currently are.

1. The Liquid, Creamy, Sticky Rule.

Precautionary labels ("may contain", "made in a facility", made on a shared line") are completely voluntary on the part of manufacturers. There's no consistency at all to how they're used, or whether they're used. While I don't ignore them if they're there, I also try to think beyond them and consider the product being made:

Manufactured items that start out as creams or sticky dough are more likely to be contaminated. 
Over the years, we've had two really memorable incidences of reactions from cross-contamination. One involved a sticky candy (where the QA director told me the peanut dust "hangs in the air in the factory) and the other involved a small cookie manufacturer. These experiences have caused me to be very nervous about sticky.

2. Shared Product Lines. 

Sharing a manufacturing line is not always a big deal. After all, most of the small manufacturers in our country essentially rent time on manufacturing equipment. Even many large companies shift manufacturing operations, depending on open capacity.

However, there's sometimes an issue when a company carries two brands that are manufactured on shared equipment, where one of the brands contains an allergen. Examples of this would include almond and soy milks, different flavors of soups, or milk chocolate and dark chocolate. Because the products are similar, the companies will often run one manufacturing process in the morning, and the second manufacturing process in the afternoon. While all companies give lip service to "good manufacturing practices" and clean lines, the reality may be that the new batch is poured into the line after not-good-enough cleaning.

3. Small Companies.

When my son was little, we were excited to find a wholesome cookie that listed safe ingredients for him. However, I also noticed they carried a peanut variety so I called the company. The person I talked with assured me that the lines were thoroughly cleaned between batches and there would be no issue.

There was. My son suffered a pretty dramatic reaction that time. It made me extremely wary of small companies, both because the controls on information can be so poor and because of the issue I mentioned before: they tend to rent time on lines and may cut corners on cleaning.

I see a lot of folks who complain about the labeling practices of various large manufacturers. However, there's simply less likelihood of issues with larger manufacturers because they do not switch lines as often. National brands have enough business to run the same manufacturing batches over and over. The less switching, the less risk.

4. Country of origin.

This one's always a bit of a game because it can be difficult to tell what's really going on in any country. However, consumers protections tend to go hand in hand with economic development.

I am leery of products manufactured in China or Mexico. I trust products manufactured in Canada much more than US-made items. We do use many canned imports, but since many of these products are likely processed at a dedicated facility (i.e., coconut milk), the risk seems minimal. Others I know won't use any products processed outside the U.S.

One important warning: European chocolates have an entirely different labeling standard that U.S. chocolates. We completely avoid these and I highly recommend you do as well.

I'm sure there are other useful rules (and if you'd like to share yours, please add them in the comments). It's a shame we need these at all. But, once bitten, twice shy as they say. Hopefully my rules will help someone in the future to avoid the bite of a reaction from poorly-labeled food.

Follow me on Facebook or Twitter  


  1. Thank you for this blog post. ~e

  2. Let me add one for "false sense of security" -- a food is labeled "organic" so folks give that manufacturer/product more trust and believability so far as the ingredients and "may contains".

    Nothing like the blind leading the blind on that one.

  3. The reality of unlabeled allergens is one that I was forced to accept years ago. After countless "mystery reactions" despite my own care to prevent cross-contamination at home, it was impossible to discount! For me, acceptance of this fact was a relief.

    For those who overcome fear that their child could die from anaphylaxis by imposing what seems like total control over their child's diet - make that their child's entire environment - accepting that processed foods are cross-contaminated would be almost equal to choosing not to carry epinephrine. I am so glad to have moved past this point!

  4. Thank you for the run-down and clarification as well as the links answering my other post. I'm going to bring up this topic on my blog and link to yours as I think this info is extremely useful.

    There is definitely a false sense of security pervading the allergy warning labels. I had no idea a reaction had to be reported in order for an allergen warning to appear! This is terrible and now the 'mysterious reactions' that still ensue are not so mysterious anymore.

  5. Thank you so much for posting. I already tend to follow 1, 2 and 4 from having been around the block for a year or two but had not considered 3 as much so thank you for your insight.

  6. Hi FAB, We avoid products from companies that make product lines of food containing the allergen, too. For example, Late July has several similar products lines, crackers with peanut butter or crackers with cheese. Or Plum Organics, which makes peanut butter sandwiches. However, I was unaware that there were foods out there that are contaminated with allergens (without warnings). Do you know if the contamination is between product lines of one brand - Hershey's for example, which doesn't have any labeling although they have many almond-containing products - or is the contamination more random?

  7. If you click on the link in the second paragraph, it will take you to the actual study where the 2% number was cited. I do know many of the items involved milk contamination with candy. Beyond that, I think it was somewhat random.

    Keep in mind as well that the study was very limited (400 products, I think?). There is no comprehensive industry sampling of foods for unlabeled contamination. The process to find them is essentially to use us as guinea pigs; once a report is filed, the FDA looks into the situation.

    This is the one lab in the country that I know of that's trying to test some food proactively to see what problems are uncovered. (They will also test foods if your child has a reaction.)

  8. I completely agree with 2 and 4, we are very aware of those as well. I did know that reactions due to cross contamination must be reported to start an investigation, and I know that few complaints result in an actual recall. I know that fewer people complain, as many allergic consumers have experienced reactions due to cross contamination and simply stop buying the product in question.
    A rule that I also abide by is contamination type. In a shared plant where dairy and egg products are produced on separate lines I am far less stressed. This is because both items are heavy and wet, so do not become airborne as easily or travel as far when they do. In plants that have dust- nut dust especially- I say no. The dust travels farther more easily and leads to greater likelihood of cross contamination.
    I acknowledge that my standards are more relaxed than many of the allergic families I talk with, but I do not find their strategies to be incorrect, just different from my own.
    I also know that 90% of the food in my house is made by me, from scratch. The few processed foods we eat I have carefully verified to meet my levels of safety. My daughter is allergic to 17 different foods, including 7 of the top 8 and rice, so even if I wanted to enjoy manufactured snacks... it just isn't an option. There are few that are made without her allergens. Because of this I have less experience with cross contamination in the foods we eat. (Though we have had a few reactions where that is the only explanation.)
    I think that given the number of times there have been "mystery" reactions we need to work on better labeling for cross contamination. Labeling in the US is not great, and the few cautionary labels we have are voluntary and inconsistently applied. I would love to see tightened requirements on labeling practices.

  9. Thank you for covering these allergen issues. I have recently started a new company to make gluten-free healthy snacks for kids like my 3, who are gluten-free, sometimes dairy-free, sugar sensitive and attend nut-free schools! I was frustrated with the options I had for snacks, which included a lot of carbs and little protein that I started making my own energy bars. We just launched the company--ZEGO or

    I am discovering just how hard it is for a small company to be as allergy-sensitive as possible. Our manufacturer follows almost all of the FARRP recommendations for avoiding cross contact and we are having our product tested post production for the Top 8 as well as gluten.

    Navigating the world of foods "produced in a shared facility..." has been difficult, though, as the FDA doesn't provide us with any definitions or standards to use. Instead, I've done a lot of research with different allergy organizations to pull together as much of that as I can, though I still can't find generally accepted definitions for terms commonly used in disclaimers.

    For example, we have not been able to find a copacker that is free of the Top 8 allergens but our manufacturer is wheat-free and breaks down all the equipment at the end of the day, cleans and sanitizes, reassembles and produces our bars as the first run the next morning (greatly minimizes dust contact potential). Does that mean we use "shared equipment" or not? It wasn't shared that morning before we used it but it is generally shared. If you or any of your readers know a place I can get more information on that, I would really appreciate it. We want to be as allergy-friendly and responsible to our customers as possible.

    I would also like to know if your allergic readers would find it helpful if I post the PPM for the top 8 allergens on our website? Most people don't know what their tolerance is, so I dont' want to provide useless or confusing information, but I would like the customer to have as much info as possible in helping them make their decision on whether ZEGO bars would be safe for them or their kids to eat.

    Thank you

    1. I commend you for making zego bars, but I would not be comfortable having peanut and tree nut allergic son eat your bars. your bars are still produced on machinery which produces bars with nuts in it. It's wonderful you are making those bars, but I will stick to the Enjoy Life bars as those absolutely do not have nuts in them.

  10. Colleen, you are asking questions for which there are no firm answers! Personally (as you can tell if you read this blog post), I like more information and would love it if manufacturers would post PPM, even knowing that the data is only representative. But a LOT of FARE members essentially expressed that they did not want this information when they commented during the FDA open comments period.

    This thread has a lot of good opinions on all of this:,6714.0.html

    I think the bottom line is that people are asking for as much information as possible from manufacturers. BUT...if I were a business owner, as you are going to be, I would be thinking about the following:

    - If I give PPM information, will it just scare people off because they really don't understand the issues and believe that foods currently contain NO allergens?

    - If I give a quantifiable cut-off and someone has a reaction, especially if they prove that the batch that caused the reaction had a higher percentage of the allergen, would I be more legally liable?

    - What's the cost of the added testing vs. the potential market return?

    If I were making this decision as a business owner, I frankly would not choose to give this information.

    To answer your question about shared equipment, yes, that is still shared equipment. There are several good industry statements out there, though, that can help your marketing. Most of the statements seem to go something like "Manufactured on equipment shared with X, Y, and Z. Good Manufacturing Practices are used, including running batches on separate days, to minimize contact with allergens." However, just know that many FA parents will reject any food that's done on a shared line, no matter how you phrase it.

    These guys do a really nice job with their allergy statements if you need someone to copy:

  11. Lets get this stuff labeled. Sign the petition and help us get the word out!


Note: Only a member of this blog may post a comment.