Blog

Aug 28, 2023

Recalls Due to Undeclared Allergens Can Be Prevented: It Starts with Proper Controls During Packaging and Labeling

Editor’s Note: This article was written by Yale Lary, expert advisor with Active Food Safety. The current numbers of undeclared allergen recalls are close to the same as they were in 2004 when the

Editor’s Note: This article was written by Yale Lary, expert advisor with Active Food Safety.

The current numbers of undeclared allergen recalls are close to the same as they were in 2004 when the Food Allergen Labeling and Protection Act was passed,3 according to U.S. Food and Drug Administration (FDA) and U.S. Department of Agriculture (USDA) data1,2 from 2021-2023 for Undeclared Allergen Public Health Alerts (PHA) and Food Safety Recalls (FSR).

Analysis of data from the FDA and USDA shows that the top reason for undeclared allergen recalls and public health alerts is a loss of control during packaging and labeling.

In the 2004 legislation, Congress noted that the number of recalls because of undeclared allergens was 121 in the year 2000. The current PHA/FSR numbers are down compared to 2011-2019.1,2,4 The decrease correlates in part with the passing of the Food Safety Modernization Act (FSMA)5,6 and other government outreach and legislation around the control of food allergen and food intolerance substances.7,8,9,10,11,12

FOOD ALLERGEN VS. FOOD SENSITIVITY. Food allergies involve an immune system response. When a person with a food allergy consumes the allergenic food, their immune system mistakenly identifies specific proteins in the food as harmful invaders, triggering an immune response. Anaphylaxis is an immune response that can be fatal if not treated right away. Studies have shown that the eliciting dose of peanut protein to cause a systemic allergic reaction in the vulnerable population is estimated to be greater than or equal to 5 milligrams.13,14,15 To illustrate just how small a milligram is in the world of mass measurements, a 1/4 teaspoon of sugar has a mass close to 1,000 milligrams.

Food sensitivities or intolerances do not involve the immune system. They occur when the digestive system has difficulty processing certain compounds in foods. Sulfiting agents are examples of food additives that some consumers are intolerant to. Sulfite must be included in the ingredient statement if the amount of sulfiting agent is 10 parts per million (ppm or mg/kg) or more in the finished food.16

Since people with a food allergy or food sensitivity react to an amount ingested from a mislabeled food, whether undeclared presence in such a food product will provoke a reaction depends on how much food is consumed and the concentration of unintentionally present compound in that food. The eliciting dose (ED) differs by compound and the individual.

THE DANGER OF UNDECLARED ALLERGENS TO CONSUMERS. According to the CDC National Center for Health Statistics,17 6.2% of American adults had a food allergy in 2021. And 8% of all children have a food allergy.18 Moreover, approximately 40% (2 in 5) of children with food allergies in the United States have been treated in the emergency department.

There is a hierarchy of risks (see Fig. 2)19 faced by adults and children susceptible to food allergy that range from very mild symptoms to death,14 depending on host factors, level of sensitivity and the amount consumed.

ROOT CAUSES FOR UNDECLARED ALLERGEN ALERTS AND RECALLS. To better understand the sources and reasons for Undeclared Allergen PHAs and FSRs from 2021 until present, data from the FDA and the USDA Food Safety Inspection Service (FSIS) public health alert and recall reports were individually reviewed and subjectively interpreted to find clues as to what caused the failures resulting in mislabeling.

Mislabeling occurred in large, midsized and small food manufacturing facilities and in large and midsized retail grocery stores. Excluding food sensitivity data, milk (about 30%) and tree nuts (about 18%) topped the list for the most frequent undeclared allergens, followed by egg, peanuts, wheat and soy (about 10% each). Fish and shellfish were less than 5% and sesame was less than 2%. Undeclared/excessive sulfite (more than 10 mg/serving), although not an allergen, was also part of the data set, and it was shown to cause about 4% of the total PHAs and FSRs.

Most of the individual reports include the reason why the undeclared allergen recall was initiated. For example: “The recall was initiated after it was discovered by a company team member that the pecan-containing product was distributed in packaging that did not include pecans in the ingredient list. Subsequent investigation indicates the problem was caused by an isolated error in the company's packaging processes.”

However, some reports were vague; for example: “The recall was voluntarily initiated after it was discovered due to a customer complaint, who experienced a reaction, that the lot contains the undeclared tree nuts and was distributed in packaging that did not reveal the presence of the allergen. The cause of the contamination is being investigated.”

I’ve witnessed firsthand how packaging and mislabeling errors occur in food manufacturing facilities working as a food technologist, supervisor, manager and director in large and small establishments to develop and oversee food safety and quality activities. It’s because of these experiences that I was able to interpret the FDA and USDA data and estimate/categorize the most likely root causes and provide scenarios and commentary about the subject of undeclared allergen prevention.

The clear majority (about 70%) of the total PHAs and FSRs were due to loss of control during the packaging and labeling processes. Approximately 15% of the mislabeling problems were from inadequate label approval. The rest were from allergen cross contact during processing or from an ingredient supplier failing to disclose an allergen.

Approximately 92% of the PHAs and FSRs were from food manufacturing establishments, and 8% were from retail grocery stores. The root causes were categorized into four process groups: packaging and labeling; label development and approval; sanitation, scheduling and batching; and mislabeled ingredient from a supplier.

PACKAGING AND LABELING PROCESS. The packaging/labeling activity of an approved product is relatively straightforward. At this point, the heavy lifting associated with the product development, approvals and manufacturing systems has been completed. The allergen-containing ingredients and work in process have been color-coded, the product is made correctly and all that is left to do is to put the approved label on the package of the approved finished product. However, in the multifaceted, fast-paced environment of manufacturing, even a straightforward process can occasionally fail.

The following are two mislabeling scenarios associated with packaging/labeling process failures, along with excerpts from the FDA and USDA PHA and FSR data that can give insights for better preventive controls.

Scenario No. 1: A company was notified via customer complaints that the wrong product was in the wrong carton. It was a new plant, and this was the first mislabeled product complaint. The end users notified the company about the mislabeling mistake by calling the 1-800 number on the cartons. Both products contained the same allergens, and both pre-labeled cartons had the allergens listed in the ingredient statements, so there was not a health risk. But there could have been.

An investigation concluded that the reason was that a partial case of the wrong cartons was accidentally intermixed on a partial pallet of the intended cartons. The wrong cartons were not noticed during the handoff to production. The carton machine operator did not notice when loading the machine, the packers did not notice when loading the cases and the quality assurance (QA) checks were not frequent enough to catch the mislabeled product.

Interviews with the warehouse associates revealed that they understood how to issue pre-labeled cartons to packaging based on the production schedule. However, carton inventory control was not accurate. There were no verbal or written instructions to verify the accuracy of what was being handed off to production or brought back into inventory, no records of specific warehouse employee training for preventing mislabeling, no manager verification records and no process to audit.

On the production side, other than production records, there was no documented system for ensuring that the correct labels were received from the warehouse. There were no documented employee work instructions, no records of specific carton loader employee training for preventing mislabeling, no manager verification records and no audits of the processes. The only monitoring and verification records were the QA finished product box checks that were prerequisite programs referenced in the HACCP plan.

Why were these weaknesses unnoticed before the failure? The establishment’s regulatory compliance results and third-party audit scores were good. The HACCP plan had current signatures from all HACCP team members.

In this scenario, the company was able to do a market withdrawal instead of a recall. Leadership reacted appropriately, immediately implementing short-term process improvements primarily focused on clearly informing the warehouse and production packaging operators that they were expected not only to prevent downtime by delivering labels on time and keeping the machines loaded, but they were also expected to make sure product labeling is correct.

Cross-functional leaders and QA developed a set of easy-to-understand work instructions and employee expectations training materials, then began auditing the new quality system and coaching the team. The new system wasn’t required to be in the Food Safety Plan, so it was managed separately. The system relied more on continuous communication than it did records. The auditors became respected expert trainers across departments. In addition, the company made plans for long-term improvements like color-coding carton graphics and adding bar code readers on the finished product conveyors to automatically identify and reject incorrect cartons. An ideal system would combine manual and automated detection systems.

On the day of the mislabeling, a prerequisite program was not completely followed. Quality Assurance (QA) checked one case of the finished product, and it showed that the labels were accurate at the time of the check. However, the Labeling Standard Operating Procedure (SOP) monitoring form record in the label storage area had not been filled out. The person that normally monitored the label warehouse transfers and made sure old labels were removed from production was absent that day. This person also helped watch over labeling in production, although it was not in their job description. The label machine operators were not trained to check the labels for accuracy when they loaded them onto the machines. Their job was to keep the packaging machine loaded with labels and running/sealing efficiently.

The system broke down for several reasons. The person that was filling in to monitor the label process SOP was not properly prepared, and there was no weekly verification of the system that day. The system depended too heavily on one monitoring point and its management for such an important process. The system for preventive control at packaging was not robust, and it did not include more team members in the ownership of the labeling element.

In hindsight, the system should have easily been improved before the mislabeling, and the costs of the market withdrawal and lost customer confidence could have been avoided. Fortunately, both products contained the same allergens, so it was not a food safety recall.

LABEL DEVELOPMENT AND APPROVAL PROCESS. For new and revised products, the labeling process starts with a product concept, R&D tests, sales approval and a regulatory review. Sourcing and approving ingredients, ingredient suppliers and developing labels is technical and multifaceted. Therefore, a comprehensive label approval and inventory control system is required that is best led by person(s) trained in the art and experienced with the interconnection of labeling regulations, food safety plans, procurement, ingredient specifications (including all sub-ingredients), production scheduling and key company/plant functions.

Because there are so many factors to consider, and so much at stake, multiple top leaders should sign off on approval before final labels are printed and before the plant begins production of products intended for sale.

Manual cleaning, clean out of place and clean in place sanitation practices should be validated. Validation studies should include the chemicals, concentrations, water temperatures, disassembly, mechanical force steps and time for each process. There are various types of allergen rapid test kits, including enzyme-linked immunosorbent assays (ELISAs), lateral flow devices (LFDs) and polymerase chain reaction (PCR) tests. Companies should select allergen rapid test kits capable of reporting results that can be cross-referenced to scientific literature.

Cross-contact of allergens occurs when the production schedule does not account for allergen-containing formulas and finished products. Production scheduling is an incredibly challenging task when the product mix has an assortment of allergen-containing formulas. If the schedule cannot be set to run in the correct sequence for allergen control, then the facility must build in the cost of down time for sanitation clean breaks. Production schedulers must be truly knowledgeable about cross-contact control. Schedules should not be changed without cross-functional management approval.

Cross-contact from using a wrong allergen-containing ingredient or from using the wrong product handling equipment must be prevented. Using an allergen-containing ingredient by mistake can result in high concentrations of the undeclared allergen. Ingredient containers should be color-coded and identified per a facility allergen control program or standard operating procedure, and the employees working in the formulation and batching areas must be very aware of the cross-contact preventive controls. Only experienced, well-trained employees should be allowed to batch products. The process should be verified frequently throughout each production day and audited often.

Using the wrong allergen-handling tool can also add undeclared allergen residue into a product, although at a much lower concentration than a scheduling or batching error.

In a plant using ingredients with allergens, it may be unrealistic to achieve absolute zero allergenic protein on food contact surfaces after cleaning. A recent study20 concluded that the methods of allergen detection and risk management should be guided by the “tolerable risk” of the susceptible population. Table 1 (see below) presents estimates of how much allergen protein (mg), if ingested by an allergic person, would cause an objective allergic response, and of those who would have an allergic response, what percent would be anaphylaxis.

From Table 1, for cow’s milk, the effective dose to elicit a measurable allergic response in at least 5% of the allergic population is 2.4 mg of the milk protein. Of those individuals allergic to milk, 9% will have an objective symptom, and 2.5 per 1,000 are estimated to have an anaphylactic reaction. For walnuts, 14% who consume 0.8 mg of the walnut protein will have allergic responses, and 2.7 per thousand are expected to have anaphylaxis.

Anaphylaxis is a medical emergency, and prompt treatment is essential to prevent severe complications or death. The primary treatment for anaphylaxis is the administration of epinephrine (adrenaline), which helps to reverse the symptoms by constricting blood vessels, relaxing airway muscles and reducing allergic inflammation.

It should be noted that published literature does not contain cases of fatal reactions at less than 5 mg total protein for the priority food allergens examined to date15.

Before testing food-contact surfaces or finished product, it is important for food manufacturers and processors to be prepared for unacceptable results. For example, be ready to delay start up to re-clean and re-test before running production or make sure the complete lot of an ingredient or finished product is under the establishment’s control and not in commerce or distributed to another step in the supply chain before testing.

It should be noted that the elicit dose (ED) thresholds presented may be used when monitoring and managing allergens in the supply chain or production facilities. However, the ED values are not appropriate, nor intended to be used to define “allergen-free” labelling.

Manufacturers and grocery stores should maintain open and ongoing communication of expectations with suppliers regarding allergen control. This is part of a well-developed supplier management program. Companies should provide ingredient suppliers with ingredient specifications that include but are not limited to minimum allergen control parameters such as routine environmental and finished product testing. Ingredient suppliers should be expected to prove compliance with the specifications by providing the company with allergen test results. Ingredient labeling and records showing allergen-free test results should be verified for compliance prior to approval and should be accompanied with a supplier letter of guarantee.

IN SUMMARY. USDA and FDA labeling requirements7,21,22,23 extend to retail and food-service establishments so that persons with food allergies can protect themselves or their loved ones by avoiding foods labeled with ingredients they are allergic to. Our industry is full of capable leaders and food handlers that can do what it takes to drive the number of PHAs and FSRs for undeclared allergens below the levels we had in the early 2000s.

The top reason for public health alerts and recalls associated with undeclared allergens in foods continues to be from the loss of control during the packaging and labeling processes. The food industry has made progress (Figure 3) in lowering the rate of PHAs and FSRs per billion dollars spent on food at home from about 0.5 in 2015 to about 0.16 in 2022.1,2,24 To put these rates in perspective, in 2022, U.S. consumers spent $1,047 billion on food consumed at home.24

Robust systems and prepared people are two key factors to successful process control. Labeling control systems that heavily weigh the reliability on one person, are not in sync with production practices and are not verified each day have caused labeling failures in the past.

Preventing undeclared allergen mislabeling requires constant vigilance, because the consequences to the allergic public are very serious. We as food safety leaders must continue to study the current state of the processes even when there is no indication of a problem, we should make time to identify gaps proactively and act before there is a problem.

PRIORITIES FOR PREVENTING UNDECLARED ALLERGENS.

Yale Lary has years of hands-on management and leadership experience across a wide variety of supply chains, manufacturing processes and facility/organization cultures. He is a past chairman of IAFP’s Food Hygiene and Sanitation Professional Development Group and has co-authored several food safety research studies that have been published in peer reviewed scientific journals. Yale did his graduate studies at Texas A&M University, where his M.Sci. food safety research was funded by USDA / FSIS and his M.AgEco. was supplemented by the Agricultural Food Policy Center.

References:

1. FDA Recalls, Market Withdrawals, & Safety Alerts: https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts.

2. USDA Recalls and Public Health Alerts: https://www.fsis.usda.gov/recalls; https://www.fsis.usda.gov/food-safety/recalls-public-health-alerts/annual-recall-summaries.

3. Food Allergen Labeling and Consumer Protection Act of 2004. 21 USC 301 note Food Allergen Labeling and Consumer Protection Act of 2004 (FALCPA): https://www.fda.gov/food/food-allergensgluten-free-guidance-documents-regulatory-information/food-allergen-labeling-and-consumer-protection-act-2004-falcpa.

4. Food Allergy Research & Resource Program (FARP) - University of Nebraska – Lincoln: https://farrp.unl.edu/reg-sit-usa.

5. FSMA Final Rule for Preventive Controls for Human Food: https://www.fda.gov/food/food-safety-modernization-act-fsma/fsma-final-rule-preventive-controls-human-food.

6. Food Safety Modernization Act (FSMA): https://www.fda.gov/food/food-safety-modernization-act-fsma/full-text-food-safety-modernization-act-fsma.

7. FSIS Compliance Guidelines Allergens and Ingredients of Public Health Concern: Identification, Prevention and Control, and Declaration through Labeling 2015: https://www.fsis.usda.gov/sites/default/files/import/Allergens-Ingredients.pdf.

8. Hazard Analysis and Risk-Based Preventive Controls for Human Food: Guidance for Industry: https://www.fda.gov/media/100002/download?attachment.

9. USDA Food Safety Inspection Service Guidebook for the Development of HACCP Plans: https://www.fsis.usda.gov/sites/default/files/media_file/2021-01/Guidebook-for-the-Preparation-of-HACCP-Plans.pdf.

10. Federal Register 10562 Vol. 82, No. 29 Tuesday, February 14, 2017: https://www.fsis.usda.gov/sites/default/files/media_file/2020-08/2017-0005.pdf.

11. Food Allergy Safety, Treatment, Education, and Research (or FASTER) Act of 2021: https://www.congress.gov/117/plaws/publ11/PLAW-117publ11.pdf.

12. Title 21 Chapter I Subchapter B Part 101 - Food Labeling: https://www.ecfr.gov/current/title-21/chapter-I/subchapter-B/part-101.

13. Hourihane JO'B, S A Kilburn, J A Nordlee, S L Hefle, S L Taylor, J O Warner -. An evaluation of the sensitivity of subjects with peanut allergy to very low doses of peanut protein: a randomized, double-blind, placebo-controlled food challenge study. J Allergy Clin Immunol. 1997 Nov;100(5):596-600. doi: 10.1016/s0091-6749(97)701611: https://pubmed.ncbi.nlm.nih.gov/9389287/.

14. H A Sampson , L Mendelson, J P Rosen - Fatal and near-fatal anaphylactic reactions to food in children and adolescents N Engl J Med 1992 Aug 6;327(6):380-4: https://pubmed.ncbi.nlm.nih.gov/1294076/.

15. Food and Agriculture Organization (FAO) of the United Nations World Health Organization (WHO). 2022. Risk assessment of food allergens – Part 2: Review and establish threshold levels in foods for the priority allergens. Meeting Report. Food Safety and Quality Series No. 15. Rome: https://doi.org/10.4060/cc2946en.

16. Title 21 – Food and Drugs PART 130 -- FOOD STANDARDS: GENERAL Subpart A - General Provisions Sec. 130.9 Sulfites in standardized food: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=130.9.

17. NCHS Data Brief - No. 460 - January 2023. U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Centers for Disease Control and Prevention National Center for Health Statistics: https://www.cdc.gov/nchs/data/databriefs/db460.pdf.

18. Center for Disease Control & Prevention – Healthy Schools – Food Allergies: https://www.cdc.gov/healthyschools/foodallergies/index.htm.

19. Dubois, A.E.J., Turner, P.J., Hourihane, J., Ballmer-Weber, B., Beyer, K., Chan, C.-H., Gowland, M.H. et al. 2018. How does dose impact on the severity of food-induced allergic reactions, and can this improve risk assessment for allergenic foods? Report from an ILSI Europe Food Allergy Task Force Expert Group and Workshop. Allergy, 73(7): 1383–1392: https://doi.org/10.1111/all.13405.

20. Turner, P.J., Patel, N., Ballmer-Weber, B.K., Baumert, J.L., Blom, W.M., Brooke- Taylor, S., Brough, H. et al. 2022a. Peanut can be used as a reference allergen for hazard characterization in food allergen risk management: a rapid evidence assessment and meta-analysis. The Journal of Allergy and Clinical Immunology: In Practice, 10(1): 59–70: https://www.jaci-inpractice.org/article/S2213-2198(21)00907-7/fulltext.

21. Major Food Allergen Labeling and Cross-contact Draft Compliance Policy Guide Center for Food Safety and Applied Nutrition (CFSAN): https://www.fda.gov/media/168000/download?attachment.

22. Food and Drug Administration (FDA) – Federal Register - Hazard Analysis and Risk-Based Preventive Controls for Human Food; Draft Guidance for Industry; Availability - Sept. 26, 2023: https://www.federalregister.gov/documents/2023/09/27/2023-20738/hazard-analysis-and-risk-based-preventive-controls-for-human-food-draft-guidance-for-industry.

23. FDA U.S. Food & Drug – Food Labeling and Nutrition: https://www.fda.gov/food/food-labeling-nutrition.

24. USDA, Economic Research Service, Food Expenditure Series, nominal expenditures: https://www.ers.usda.gov/media/rpwbxbmf/expenditures.xlsx.