Cold-Smoked Salmon: One Weak CCP and a Proposal to Test the Product
A small smokehouse produces cold-smoked salmon, sliced and vacuum-packed, with a 30-day shelf life at 4°C. The process is dry-salt/brine for 18 hours, equalize for 12 hours, cold smoke at 22–24°C for 8 hours, then slice and pack. Measured product has water-phase salt of 3.1% and water activity of 0.965. The hazard analysis lists one CCP: "brine concentration ≥60 degrees salometer, checked each batch." It states that Listeria monocytogenes "is controlled by good hygiene and the salt level." No other CCP is listed. The plant has had two environmental Listeria positives in six months, both on a floor drain, each time "cleaned and re-swabbed until negative." To satisfy a customer audit, management proposes adding finished-product Listeria testing (five 25 g samples per lot) as a new CCP.
1. Critique the hazard analysis. What is missing or wrong, and why is brine salinity a weak choice as the only CCP?
Read the worked answer
The hazard analysis is deficient in three ways.
It does not really analyse Listeria monocytogenes, it asserts control. Cold-smoked salmon is a ready-to-eat product with no lethality step, a long refrigerated shelf life, and a known association with L. monocytogenes. That makes it a hazard reasonably likely to occur, and the analysis must show a control strategy, not a sentence.
The brine CCP is not tied to a validated limit. Degrees salometer measures brine strength, not what the product achieves. The control that matters is growth prevention in the product, which depends on water-phase salt and water activity in combination with time and temperature. With WPS 3.1% and aw 0.965, the product sits in a range where L. monocytogenes can grow slowly at refrigeration temperature, and a 30-day shelf life gives it time. Salinity may be a legitimate control measure, but the critical limit should be expressed as a product measurement (WPS by titration, aw) with a limit supported by growth modelling or a challenge study for this product. As written, a passing brine check does not tell you the product is safe.
The environmental positives are being treated as an inconvenience. Two drain positives in six months is a signal. "Re-swab until negative" is not a corrective action; it is a search for evidence. The response should include root-cause investigation, intensified cleaning of harbourage points, drain treatment, and a defined escalation path.
Also missing: the smoke step is not a kill step and should be stated as such, and the plan should record the shelf life and storage temperature assumptions that the growth control relies on.
2. Management wants to make finished-product Listeria testing a CCP. Evaluate that proposal.
Read the worked answer
It should be rejected as a CCP, while being allowed as limited verification.
Why it cannot be a CCP. A CCP is a step where control is applied to prevent, eliminate, or reduce a hazard to an acceptable level. End-product testing does not control anything; it reports on product that already exists. It is retrospective, it cannot prevent the hazard reaching consumers, and at five 25 g samples per lot it has very low sensitivity for low-level, non-uniform contamination. A negative result says little; a positive result says the lot is contaminated, which is too late.
What it does do, badly. As a verification activity it has a place in some regulatory frameworks and customer expectations, but it must be understood as a check on the system, not a release gate. If the plant uses it for release, it creates a false sense of security and an incentive to test into compliance.
The real problem it is being asked to solve. The audit wants evidence that L. monocytogenes is controlled. The right evidence is a validated formulation or process control, a shelf-life justification, and a functioning environmental monitoring program with meaningful corrective actions. Offer the auditor those instead.
The judgment: agree that the current plan is inadequate, disagree with the proposed fix, and propose a control strategy. Do not simply refuse the customer requirement; redirect it to where it does some good.
3. Recommend a control strategy for L. monocytogenes and state what evidence would be needed to support each control measure.
Read the worked answer
A workable strategy usually combines two or three measures, because any one alone is thin for this product.
Formulation control (a genuine CCP). Set a product-level critical limit, such as water-phase salt ≥3.5% or a validated aw/WPS combination, measured by titration or aw meter each batch, with a corrective action for out-of-limit brine batches. Validation evidence: a challenge study with a cocktail of L. monocytogenes strains in this product at this WPS/aw, or predictive growth modelling plus published data, showing no growth or acceptable growth over 30 days at 4°C with a temperature-abuse scenario.
A listericidal step, if feasible. High-pressure processing at a validated pressure and hold time, or a validated post-pack heat treatment. Validation evidence: a challenge study demonstrating the target log reduction in the actual product matrix, plus process records showing the equipment delivers it.
Time/temperature control. A validated shelf life with a defined maximum storage temperature and a labelled use-by date. Validation evidence: shelf-life study including a modelled abuse profile, and distribution temperature data.
Environmental monitoring as a prerequisite, not a CCP. A written program with site maps, sampling frequency, alert and action levels, and defined responses to a positive that go beyond re-swabbing: source investigation, intensified cleaning, equipment teardown, and product risk review. Verification evidence: trend reports reviewed at management review.
Present the combination with the reasoning that each measure covers a different failure mode: formulation slows growth, a kill step reduces numbers, cold chain limits growth, and environmental monitoring keeps the plant from seeding product.