ISO 22000 clause 8.9.4.2: make a defensible evaluation for product release

What clause 8.9.4.2 requires in practice

Clause 8.9.4.2 requires every lot affected by a nonconformity to be evaluated, not just the batch that happened to be sampled. It then draws a sharp line between two situations. Product affected by a failure to stay within CCP critical limits is not released; it goes to disposition under 8.9.4.3.

Product affected by an OPRP action-criterion failure can be released as safe only if at least one condition is demonstrated. Either evidence other than the monitoring system shows the control measures were effective, or evidence shows the control measures acting together on that product achieved the performance intended, or sampling, analysis or other verification shows the product meets the acceptable levels for the hazards concerned. The results of this evaluation for release are retained as documented information. Nothing in the clause lets a single satisfactory test turn a CCP failure into a releasable lot, and nothing lets an OPRP failure be waved through without one of the three conditions.

Account for each affected lot

Establish which lots are affected and the evidence relevant to each. Shared evidence may apply to more than one lot where justified, but the evaluation must not lose individual lot accountability.

Check whether operating conditions, timing or other controls differed across the affected range. A conclusion supported for one lot should not automatically be extended to another whose relevant circumstances differ.

Keep the product controlled while the evaluation is pending. The general product-control requirement remains important while information is gathered and interpreted.

Preserve the CCP restriction

A CCP critical-limit failure follows disposition of nonconforming product, rather than ordinary release under the OPRP evidence routes. A satisfactory sample does not turn that failure into an OPRP event.

Competent technical evaluation can inform an appropriate further-processing or other permitted disposition. It must not be presented as a way to ignore the standard's explicit release instruction.

Why a defensible release decision matters

Release is the point of no return. Once product is dispatched, the organisation depends on whatever evidence it had at that moment. A decision based on hope, a convenient test or commercial pressure is hard to defend to a customer, an authority or a court. A decision based on documented evidence, matched to one of the permitted conditions and made lot by lot, can be defended and also helps the business recover value from product that is genuinely safe. Keeping the CCP and OPRP routes apart avoids the worst mistake, which is releasing CCP-affected product after reassuring sample results. It also trains the team to ask what the evidence actually proves, a habit that improves every other food safety decision.

How to implement clause 8.9.4.2

Set up the evaluation so that evidence, not urgency, drives the answer for every affected lot.

1. Define the lot or batch boundary the evaluation will use, and require a separate conclusion for each lot affected.
2. Write the CCP rule at the top of the instruction: affected product is not released and is passed to disposition.
3. For OPRP failures, offer evaluators a short prompt for each of the three evidence conditions and what good evidence looks like.
4. Specify who may judge sampling plans and analytical methods, so that unrepresentative results are not relied on.
5. Require the evaluator to state the evidence, limitations and conclusion in the record, before anyone is asked to approve.
6. Keep the evaluation separate from the commercial request to ship, and record who approved the final release or disposition.
7. Test the instruction with a past event and ask whether a stranger could reach the same decision from the record.

Evaluate the permitted evidence after OPRP failure

Illustrative food-manufacturing example

Illustrative allergen-control example

A manufacturer considers sampling after a failed OPRP criterion in an allergen-control arrangement. Competent personnel examine whether the proposed sampling and method can support the intended conclusion for the relevant product and hazard.

The team does not treat one convenient negative result as proof for every affected unit. If the evidence is insufficient, it follows an appropriate disposition rather than weakening the question until the result appears acceptable.

Retain the evaluation and its basis

Results of product-release evaluation must be retained as documented information. Link the conclusion with the lots, failure, evidence and competent authority. Keep the distinction between the technical evaluation and execution of the authorised release or disposition.

Common failures include using the same checklist for CCP and OPRP events, ignoring lot differences, assuming another control was effective without evidence and treating absence of complaints as proof of safety.

Connect the decision with the OPRP failure evaluation. The consequence and cause assessment provide context for the release question.