The dog detects trace amounts of a specific allergen (most commonly peanut, tree nut, egg, dairy, gluten, shellfish) in food and objects, and communicates the find with a passive trained indication — allowing a handler with a severe (often anaphylactic) allergy to avoid exposure.
The three-test protocol in §5.5 is the canonical Service Paws allergen assessment. It predates this manual and is preserved as the standard used for all certified allergen dogs.
5.1 Certifiable tasks
| Task | Trigger | Standard |
|---|---|---|
| Controlled product/food check ⚠ | Cue ("check") on a presented item | Clear passive indication if target odour present; clear "all clear" behaviour (disengage/return) if absent |
| Room / area search | Cue, systematic search | Locates accessible target-odour sources in the area within the time standard |
| Trained indication | Odour find | Passive, unambiguous, non-destructive (e.g., nose press to handler's leg or arm, sit-and-stare); must not touch/damage the item |
| Refusal of unchecked food (recommended) | Food offered/reachable | Does not take food not released by handler (overlaps PAT P10/P14) |
5.2 Candidate suitability
- Strong food/toy drive convertible to search motivation; sound-stable per Chapter 1.
- No chronic nasal/respiratory disease (§1.2).
- Any size; small-to-medium dogs often suit table-level and travel work well.
5.3 Training methodology requirements
- Odour imprinting on the verified target allergen using clean-sample protocol: target samples prepared and stored in dedicated airtight containers; separate tongs/gloves per sample class; hot and blank containers never share storage or handling.
- Blank and distractor training from the start: unscented containers, non-target foods, and packaging materials, so the dog learns the odour — not the container, the routine, or the handler's expectation.
- Passive indication selected and fixed early. Vocal or destructive indications are not certifiable.
- Handler-blind sessions must be part of training (a third party places samples) to prevent unintentional cueing before the blind test.
- Handler safety protocol: the allergic handler never handles hot samples; during training/testing involving the handler, sealed secondary containment and a second person managing samples are required. For anaphylactic handlers, emergency medication must be on site through all hot-sample work.
5.4 Live exposure log (recommended)
A dated log of real-world checks (item, result, outcome) is recommended for the dog's file — it supports attestation letters and re-assessment decisions.
5.5 Task test — the Service Paws allergen protocol
All three tests are scored by the assessor; samples are placed by someone other than the handler. Tests 1–2 are typically conducted on one day; Test 3 in real public venues on a separate day.
Test 1 — Controlled test ⚠
- 15 presentations, of which 5 contain no trace of the allergen (blanks), presented in randomized order at 5-minute intervals to prevent cross-contamination or confusion.
- The dog must show its clear, trained indication when the allergen is detected, and respond within 60 seconds of presentation.
- No failed presentations across all 15 (a miss on a hot item or an indication on a blank is a failure).
- An item may be re-presented to the dog, but no more than 3 total attempts on that item.
Test 2 — Room search
- A room free of other allergen sources is prepared with 3 scented items of varying intensity: a large allergen-containing food source; a medium source; and a small trace affixed to a tap or a piece of paper. All placed so scent can escape.
- The team has 15 minutes to locate all 3 items.
Test 3 — Public product check (grocery store + restaurant) ⚠
Conducted with real-world distractions, with precautions preventing handler exposure.
Grocery store:
- Seven items pulled from shelves for the dog to check — food, cleaning supplies, toys, dishes, or hygiene products.
- At least two items contain the allergen, confirming correct indication.
- The dog's indication and all-clear behaviour must not damage products.
- A false alert on an allergen-free product is tolerated in this environment only if, on re-presentation, the dog gives 3 consistent indications in a row (treated as a find for scoring).
- The dog must not "free-indicate" on shelf products — only on the controlled search.
- The dog must remain focused and responsive; if distracted, the handler must refocus the dog within 1 minute.
Restaurant:
- Three dishes are presented, one containing the allergen; the dog must correctly indicate the hot dish.
- No spilling or mess during indication or all-clear behaviour.
- The dog stays calm and controlled throughout a full meal.
Pass: Test 1 clean (within the re-presentation rule) + all 3 items found in Test 2 within time + Test 3 completed with correct indications and conduct as above.
5.6 Maintenance
- Weekly odour refreshers with verified samples; monthly blank sessions to keep the false-alert rate honest.
- Samples refreshed regularly; expired or contaminated sample stock discarded.
- Recommended annual re-run of Test 1 (informal, logged) to confirm detection reliability.
5.7 Records for this chapter
Scoresheet (Ch. 16) for Tests 1–3 with dates and venues; sample-handling notes (who placed, blind status); video when overseen remotely. On pass, Service Paws records the task testing date.