Skip to content
Field guide8 Curing Problems: How to Diagnose, Fix, and Prevent Them
Concept

Physical Hazards

Physical hazards are foreign or natural objects whose material, size, shape, hardness or location can cause choking, cuts, puncture injury, broken teeth or other harm when cured meat is eaten.

Identity and injury mechanism

Physical hazards include metal fragments, glass, brittle plastic, stones, wood, rubber pieces, needles, wire, hard bone, clips and packaging components. A natural part of an animal or ingredient can become hazardous when its size, hardness or context makes injury reasonably foreseeable. The relevant question is not simply whether material is foreign, but whether the object can harm the intended consumer under the product's normal use. Ready-to-eat sliced meats may receive no preparation capable of removing an object. Vulnerable consumers can also be harmed by smaller material. FDA's hard-or-sharp-object policy uses numerical ranges in a particular enforcement context, but these are not universal safe limits and should not be copied into a meat plant's release rule. Soft filth, pests and quality defects may require action for other reasons even where puncture risk is low.

Sources across cured-meat production

Raw meat can carry bone, shot, needles or material introduced during slaughter and transport. Spices and dry ingredients can contain stones, plant fragments, wire or packaging debris. Knives, saws, grinder plates, mincer blades, injectors, hooks, racks, clips and slicers can shed metal when worn, damaged or incorrectly assembled. Brittle guards, scrapers, seals, gloves, pens, lights, thermometers and cable ties create plastic, rubber or glass risks. Wood pallets, maintenance work and construction add further routes. Packaging contributes film, foil, clips, absorbent pads or hard fragments, particularly at slicing and sealing. The hazard analysis should follow each material through the actual line and distinguish likely source, object characteristics, point of introduction and whether a later step can reliably prevent or detect it. A generic foreign-body list without line-specific evidence is not an adequate analysis.

Prevention by design and maintenance

Prevention is stronger than relying only on end-of-line detection. Hygienic equipment design should avoid loose fasteners, metal-to-metal contact, brittle components above exposed product and inaccessible locations where damage cannot be seen. Preventive maintenance sets inspection and replacement criteria for blades, plates, needles, seals, belts, scrapers and guards. Tool and parts control accounts for items before and after work; temporary repairs require authorisation, food-safe materials and a defined removal date. Glass and brittle-plastic programmes identify protected fixtures, inspect them at risk-based intervals and prescribe breakage response. Knife and needle counts are useful only if identities and discrepancies are investigated. Maintenance release should include reassembly, debris removal and inspection of the product zone. A detector downstream cannot justify operating damaged equipment that continues generating fragments or objects outside the detector's capability.

Supplier, receiving and handling controls

Supplier specifications should address plausible foreign materials for meat, casings, spices, ingredients and packaging. Receiving inspection can detect damaged containers, pallet debris, broken seals or gross contamination, but it cannot see every object within bulk material. Supplier controls may include slaughter and deboning practices, sieving, magnets, optical sorting, detector records and complaint trends. The receiving business must understand what was actually controlled and at what sensitivity. Decanting ingredients, removing staples, controlling bag knives and keeping outer packaging away from exposed product prevent new contamination. Bone-management criteria should match the product: a whole cut, coarse sausage and finely emulsified ready-to-eat product present different expectations and detection challenges. Supplier certificates and a clean top layer do not eliminate the need for in-process prevention, and repeated findings require supplier corrective action rather than routine sorting around the defect.

Screening and detection technologies

Sieves, filters, magnets, visual inspection, metal detectors and X-ray systems detect different materials. A magnet captures suitable ferrous particles but not glass, bone or many stainless-steel fragments. Metal-detector sensitivity depends on metal type, size, shape, orientation, product effect, aperture, speed and package format. X-ray performance depends on density contrast, thickness, position, equipment settings and product structure; low-density plastic, wood or thin bone may be difficult to detect. Visual inspection is limited by coverage, fatigue and objects hidden within the meat. Equipment placement matters because contamination introduced after the device is not controlled by it. The detection limit established for one pack size or recipe cannot be assumed for another. Detector capability is not the same as absence of foreign material; it is evidence that defined test pieces can be found under stated operating conditions.

Validation, routine challenges and rejected product

Validation establishes that the chosen prevention or detection measure addresses the identified object in the real product and configuration. For a detector this includes the least detectable relevant material, test-piece dimensions, product temperature, orientation, package position, line speed and rejection mechanism. Routine challenge checks then verify continuing operation at defined frequencies, including startup, changeover, shutdown and after intervention where appropriate. The challenge should test detection and physical rejection, not merely illuminate a panel. Test pieces require identification and control so they cannot enter product. Rejected packs must be secured, reconciled and investigated; repeatedly passing them through until one clears destroys evidence and can return contaminated product. Loss of challenge performance defines an affected time window from the last satisfactory check, which must be held and assessed before release.

Incident investigation and product scope

When an object is found, stop and preserve it without altering edges, residue or location evidence. Record the product, pack, lot, line, time and discovery point; photograph and identify the material; and inspect likely source equipment. Determine whether the event was a single intact object or fragmentation capable of producing multiple pieces. Product scope should extend across the credible generation and detection window, not merely the pack containing the discovery. Detector history, maintenance, tool counts, earlier complaints and adjacent-line work can refine that window. Removing the visible fragment does not establish that the remainder is clear, and a negative scan may be uninformative if the material is not detectable. FSIS inspection procedures emphasise detection, segregation and disposition after foreign-material contamination; establishment evidence must support each of those stages.

Disposition, verification and learning

Disposition considers object identity, injury potential, product use, consumer, distribution, quantity, fragmentation, detector capability and jurisdictional requirements. Reinspection or reprocessing is acceptable only when a supported method can control the specific material throughout the lot and does not create another hazard. If product has shipped, traceability and recall procedures may be required. Corrective action removes the source, repairs or replaces equipment, and examines why preventive inspection or detection failed. Verification reviews challenge records, reject counts, maintenance findings, supplier incidents and complaints for recurring patterns. Trends by material and source are more informative than a single foreign-body total. The programme should also confirm that operators stop the line and preserve evidence rather than quietly removing objects to protect performance figures. A zero-complaint period cannot prove absence when consumers may not report every event.

Related in the Codex

References

  • https://openknowledge.fao.org/server/api/core/bitstreams/6866dc55-d2c0-48dd-a528-a4d634f1b0b4/content
  • https://www.fsis.usda.gov/policy/fsis-directives/7310.5
  • https://www.fda.gov/media/71953/download
  • https://www.fsis.usda.gov/guidelines/2018-0005
  • https://www.fao.org/fao-who-codexalimentarius/sh-proxy/pt/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcodex%252FStandards%252FCXG%2B69-2008%252FCXG_069e.pdf