Food safety

FDA is rethinking Listeria control — and most monitoring programmes are designed to pass, not to find

FDA is taking comments on Listeria monocytogenes prevention until 2 November 2026, after a two-day meeting where experts debated zero tolerance and pressed for environmental monitoring that genuinely looks for the organism. Three things worth understanding, whether or not you comment.

FDA is taking comments on how Listeria monocytogenes should be controlled until 2 November 2026. The docket is FDA-2026-N-7914, and it follows a two-day public meeting held on 18–19 August at which the disagreements were more interesting than the agreements.

Most food manufacturers will never submit a comment, which is reasonable. But the meeting is still worth understanding, because what was argued there is what your auditor will be asking about in two years' time. Three things came out of it that deserve your attention regardless of whether you write to FDA.

1. Your environmental monitoring programme is probably designed to pass

The phrase that came up repeatedly was “seek and destroy”: environmental monitoring designed to aggressively find Listeria, rather than to demonstrate its absence.

That is a bigger distinction than it sounds, and it is where most programmes quietly fail. If your sample sites have not changed in three years, if you swab after sanitation rather than during production, if you avoid the drain and the floor-wall junction and the underside of the conveyor because those are where you might actually find something — then your programme is not looking for Listeria. It is producing evidence that you looked.

A genuinely useful programme finds things. It should make you uncomfortable occasionally. A year of clean results from a stable set of easy sites tells you about your sampling plan, not about your plant.

The honest test: when did your environmental monitoring last tell you something you did not already know? If the answer is “it hasn't”, that is the finding.

2. Persistence is a design problem, not a cleaning problem

The meeting spent real time on Listeria persistence, sanitation and hygienic design — and the order matters. A harbourage site is a place the organism survives sanitation because sanitation cannot reach it: hollow rollers, cracked welds, worn gaskets, equipment feet, standing water under a line.

No amount of cleaning frequency fixes a hollow roller. If the same site or the same zone keeps producing presumptive positives, the corrective action is rarely “re-clean and retrain”. It is to change the equipment, the drainage or the traffic pattern — a capital conversation rather than a sanitation one, and the conversation most plants defer.

This is also where corrective action records give the game away. A string of closed actions that all read “deep cleaned, re-swabbed, negative” is a system treating a design fault as a cleaning lapse, repeatedly.

3. The zero-tolerance debate is not permission to relax

The most contested session concerned dose-response. Models presented at the meeting suggested that low-level L. monocytogenes exposure may carry relatively low risk, which prompted genuine debate about whether a strict zero-tolerance stance is proportionate.

Two things need saying, and they pull in opposite directions.

First, it is a live scientific question and the people arguing it are serious. Second — and this is the part that matters for anyone running a plant tomorrow morning — none of it has changed any requirement you are held to. Zero tolerance for ready-to-eat products is the current position. Your GFSI scheme has not moved. Your customers' specifications have not moved. A meeting at which experts debated proportionality is not a basis for loosening anything, and “we read that low levels might be lower risk” is not a defence anyone will accept.

Watch the debate. Do not act on it yet.

What else came up

  • Growth after dispatch. L. monocytogenes can grow after product leaves your facility, which is why retail food safety management and shelf-life assumptions were discussed as part of the same problem. Your validated shelf life is a control, not a labelling decision.
  • An ageing population. US listeriosis incidence has been relatively stable, but the population most vulnerable to it is growing. A stable rate against a growing susceptible population means a rising burden — which is the underlying reason this is on FDA's agenda at all.
  • Retail is in scope. Much of the meeting dealt with ready-to-eat foods at retail, not only at manufacture. If you supply deli, foodservice or in-store preparation, the controls being discussed reach past your dispatch bay.

What this means for your next audit

Regulatory direction reaches you through your certification scheme. FSSC 22000, BRCGS, SQF and IFS already require environmental monitoring for ready-to-eat products, and scheme owners track FDA's thinking closely. When emphasis shifts — toward programmes that find organisms rather than programmes that demonstrate absence — auditors ask different questions.

The questions worth rehearsing now:

  • How were your sample sites chosen, and when were they last reviewed against actual risk rather than carried forward?
  • What is your response plan when a presumptive positive appears — and has it ever been used?
  • Can you show a corrective action where the root cause was equipment or design rather than cleaning?
  • Does your hazard analysis treat Listeria as a hazard managed by prerequisite programmes, and is the evidence for that still true?

For US manufacturers there is a second thread. HARPC and the preventive controls framework are where FDA expectations land first, before any scheme catches up — so a plant with a well-run PCQI-led preventive controls programme usually finds scheme changes undramatic.

If you do want to comment

Comments go to docket FDA-2026-N-7914 by 2 November 2026, and the meeting materials are published on FDA's site. The comments that carry weight are specific: what a control costs to implement at your scale, what a requirement does to a small plant that it does not do to a large one, what you have seen work.

Dockets like this are usually dominated by large manufacturers and trade associations, so a short, concrete account from a mid-sized processor describing what actually happens on a line fills a real gap. Specific beats comprehensive.

The uncomfortable summary

Nothing in the August meeting was new science. The repeated message was to reinforce fundamentals — sanitation, hygienic design, monitoring that genuinely looks. That is an unglamorous conclusion, and it is the right one.

Which means the useful question is not what FDA will decide. It is whether your environmental monitoring programme would survive being read by someone who wanted it to find something.

Sources

FDA, public meeting on Listeria monocytogenes prevention, 18–19 August 2026, docket FDA-2026-N-7914, comments to 2 November 2026. Meeting discussion as reported by Food Safety Magazine, “FDA's Listeria Prevention Public Meeting: Experts Debate Zero Tolerance, Emphasize Environmental Monitoring and Fundamental Controls”.

Questions this raises

What is the deadline for commenting on FDA's Listeria prevention questions?

2 November 2026, to docket FDA-2026-N-7914. FDA published the materials from its 18–19 August 2026 public meeting and is seeking comment on key questions arising from it.

Has FDA changed the zero-tolerance policy for Listeria in ready-to-eat foods?

No. Dose-response science presented at the August meeting prompted debate about whether strict zero tolerance is proportionate, but nothing has changed. Zero tolerance remains the position, GFSI schemes have not moved, and customer specifications have not moved. The debate is worth following; it is not a basis for relaxing any control.

What does seek and destroy mean in environmental monitoring?

Designing the programme to aggressively find Listeria rather than to demonstrate its absence — sampling the harbourage sites you would rather not swab, during production rather than only after sanitation, and treating a positive as useful information instead of a problem. A programme returning a year of clean results from an unchanged set of easy sites is telling you about its sampling plan, not about the plant.

Why does hygienic design matter more than cleaning frequency for Listeria?

Because persistence is usually a design problem. A harbourage site is somewhere the organism survives sanitation because sanitation cannot reach it — a hollow roller, a cracked weld, a worn gasket, standing water under a line. Increasing cleaning frequency does not reach those places. If the same site keeps producing positives, the corrective action is equipment, drainage or traffic flow rather than re-cleaning.

Does this affect our FSSC 22000, BRCGS, SQF or IFS certification?

Not immediately, but regulatory direction reaches certified sites through their scheme. All four already require environmental monitoring for ready-to-eat products, and scheme owners follow FDA's thinking. The practical preparation is being able to show how sample sites were chosen, that your response plan for a presumptive positive has actually been used, and at least one corrective action whose root cause was design rather than cleaning.

We are a small processor — is it worth submitting a comment?

Often more worthwhile than for a large one. Dockets like this are usually dominated by large manufacturers and trade associations, so a short, concrete account of what a control costs at your scale, or what a requirement does to a small plant that it does not do to a large one, fills a gap.

Begin with context

Discuss how this applies to your operation.

Share the requirement, your current situation, and the outcome you are working toward.

Contact QSE