Allergen Changeover Validation: Methods and Verification

Allergen changeover validation is the documented process of proving that the cleaning performed between two different products reliably removes allergen residue to a level that is safe for the next product run. When a line switches from a peanut-containing recipe to a peanut-free one, the changeover clean has to be shown to work, not just assumed to. Validation is the up-front scientific proof, while verification is the routine check that the proven procedure was actually followed on the day. Get either wrong and you risk an undeclared allergen, one of the most common causes of food recalls in the EU and worldwide.

Validation versus verification: two different jobs

These terms are often used loosely, but they carry distinct obligations under HACCP and allergen management standards such as BRCGS and FSSC 22000.

• Validation is a periodic study, run once and repeated when something material changes (new equipment, new recipe, altered clean). It typically requires three consecutive successful cleans under worst-case conditions to demonstrate the procedure is capable.
• Verification is the ongoing evidence, gathered at every changeover, that the validated clean was executed correctly. It usually combines a documented visual inspection with a rapid allergen test at defined swab points.

Put simply: validation answers "can this clean remove the allergen?" while verification answers "did it, this time?" A changeover clean is also unplanned-feeling downtime that erodes the availability term in your overall equipment effectiveness calculation, so a validated, right-first-time procedure protects both food safety and throughput.

Set the acceptance limit: a worked example

You cannot verify a clean without a numeric target. Reference-dose frameworks such as VITAL convert a safe protein dose into a concentration limit for your specific product.

1. Take the milk reference dose (ED01), roughly 0.2 mg of milk protein, the amount tolerated by about 99% of allergic consumers.
2. Take the serving size of the next product, say 30 g.
3. Divide: 0.2 mg / 30 g = 0.0067 mg/g, which is 6.7 mg/kg, or about 6.7 ppm milk protein.

So for this product, residual milk protein above roughly 6.7 ppm would trigger a precautionary allergen label; a validated clean must reliably get you below it, ideally with a safety margin. A smaller 15 g serving halves the tolerance to about 3.3 ppm, which is why the acceptance limit is always product-specific and must be recalculated when serving sizes change.

Lateral flow and ELISA: protein-specific detection

Both methods detect the actual allergen protein, which is what regulators care about.

• Lateral flow devices (LFDs) are the workhorse of the factory floor: swab a surface, elute, apply, and read in around 10 minutes. They are qualitative or semi-quantitative, with detection limits commonly in the low single-digit ppm range, ideal for per-changeover verification.
• ELISA is the laboratory reference method. It is fully quantitative, sensitive to sub-ppm levels, and takes hours plus a lab, which makes it the tool of choice for the validation study and for confirming any LFD failure.

A practical pattern is ELISA to validate and periodically audit, LFD to verify daily. Because both rely on a measurement, treat the test itself as a measurement system: a quick gauge R&R style check that different operators read the same swab consistently prevents false confidence, and an acceptance sampling mindset helps you decide how many swab points make a clean line "accepted."

ATP swabs: fast cleanliness checks, not allergen proof

ATP bioluminescence swabs measure adenosine triphosphate from organic residue and give a relative light unit (RLU) reading in about 15 seconds. They are excellent for confirming general cleaning effectiveness, catching a poorly cleaned surface before you even bother with an allergen test.

But ATP is not allergen-specific. A denatured or highly processed allergen protein can still be immunologically active while carrying little ATP, so a low RLU does not guarantee the allergen is gone. Use ATP as a first-pass cleanliness gate and rapid feedback for operators, then confirm allergen removal with an LFD or ELISA against your calculated ppm limit. The two answer different questions and neither replaces the other.

Build the validation protocol and line clearance

A defensible protocol nails down the following, in writing, before the first test.

• Worst case: the hardest-to-clean allergen, the stickiest soil, the longest dwell, and the most complex equipment path.
• Sampling points: named high-risk contact surfaces (dead legs, gaskets, augers, blades) plus final rinse water where relevant.
• Sample count and method: how many swabs, swab area, elution, and the pass/fail ppm target.
• First product through: testing the leading portion of the next batch, which is often diverted or held.

Line clearance, the formal sign-off that the previous product and its allergens are gone, is the gate that releases the line. Mapping the changeover with a value stream mapping exercise exposes where residue hides, and treating the routine clean as autonomous maintenance puts consistent execution in operators' hands.

Documentation that survives an audit

An auditor will ask for the chain: the cleaning SOP, the validation report with its three runs and ELISA results, the per-changeover verification records with LFD outcomes and swab locations, deviation handling, and the resulting precautionary labeling decision. Folding these expectations into a formal control plan keeps the acceptance limits, methods, and reaction plan in one controlled document, and scheduling the cleans and checks through a CMMS turns them into traceable, time-stamped work orders instead of paper that goes missing.

Where Fabrico fits

Fabrico is the real-time data foundation that makes allergen changeovers visible and traceable. Its field-ready CMMS lets you schedule changeover cleaning and verification as preventive work orders, attach the SOP and required swab points, and capture LFD or ATP results as time-stamped records against the specific asset and line. Real-time OEE and production monitoring show exactly how long each allergen changeover takes and how that downtime hits availability, so you can target the slowest cleans without guessing. For lines with no PLC, Fabrico's computer vision captures machine state directly. Everything is EU-built with EU data residency, which matters for factories operating under GDPR. Explore the CMMS solution and the MES and OEE solution to see how the pieces connect.

Frequently Asked Questions

Can ATP testing replace allergen-specific testing?

No. ATP swabs measure general organic residue and cleaning effectiveness, not allergen protein. A processed or denatured allergen can persist with a low ATP reading, so ATP is a useful first-pass cleanliness gate but cannot confirm that an allergen has been removed to your ppm limit. Confirmation always needs a protein-specific method such as a lateral flow device or ELISA.

How many runs are needed to validate an allergen changeover clean?

Common practice, aligned with major food safety schemes, is three consecutive successful cleans performed under worst-case conditions. Three passes demonstrate the procedure is repeatable rather than a lucky one-off. Revalidate whenever you change equipment, recipe, allergen profile, or the cleaning method itself.

Does allergen changeover cleaning affect OEE?

Yes. The clean and its verification are planned or changeover downtime that reduces the availability component of OEE. A validated, right-first-time procedure minimizes reruns and re-cleans, protecting both food safety and line uptime, which is why tracking changeover duration alongside test results is so valuable.

Ready to turn allergen changeovers into traceable, time-stamped records and see their real cost to availability? Book a Fabrico demo and watch your cleaning, verification, and OEE data come together on one EU-hosted platform.