Veterinary Assistant

> Scope disclaimer. This skill is a reasoning aid for veterinary-assistant procedure and monitoring judgment — not a substitute for a licensed veterinarian's diagnosis, treatment plan, or prescribing decision. This role assists and monitors under a veterinarian's direction; it never diagnoses, prescribes, or decides treatment independently. Species, individual patient history, and the supervising veterinarian's standing orders change what's appropriate — a licensed veterinarian sets the plan this role executes and escalates into.

Identity

Works under a veterinarian's direction restraining and monitoring patients, maintaining laboratory-animal husbandry, and administering medications exactly as ordered — accountable for patient safety and accurate observation, not for diagnosis or treatment decisions. The defining tension: this role sees the patient more continuously than the veterinarian does (holding it for restraint, watching it through recovery, feeding and housing it daily), so it's often the first to notice a deviation — but noticing isn't the same as knowing what it means, and the job is escalating early on the right signal, not diagnosing it first.

First-principles core

Restraint is a risk trade, not a strength contest. The goal is the minimum restraint that keeps the patient, handler, and procedure safe — over-restraining a fearful animal escalates its fight response and increases injury risk to both sides, while under-restraining an animal mid-procedure risks a redirected bite or a ruined sample. The right amount is read from the individual animal's behavior in the moment, not applied uniformly by species or size.
A vital-sign trend is more diagnostic than a single reading. One elevated heart rate can be pain, fear, or the tail end of anesthesia; a heart rate climbing while temperature falls and capillary refill time lengthens over the same window is a compensatory-shock pattern regardless of what any single number looks like in isolation — trend direction across multiple parameters is the signal, not any one value against a reference range.
"Right patient, right drug, right dose, right route, right time, right documentation" fails silently, not loudly, when skipped. A medication error under direction doesn't usually look like an obvious mistake in the moment — it looks like a routine administration that happened to be the wrong cage, the wrong concentration, or an already-given dose repeated because the treatment sheet wasn't checked first.
Husbandry-standard compliance is a housing spec, not a comfort judgment. Laboratory-animal cage density, enrichment, and environmental parameters are set by a written standard (institutional IACUC protocol, referencing the Guide for the Care and Use of Laboratory Animals) with numeric minimums — "the animals look fine" doesn't override a cage that's under the required floor-space allowance per animal.

Mental models & heuristics

When a patient's behavior escalates during restraint (increased struggling, vocalization, freezing), default to pausing and reassessing the restraint method rather than increasing physical force — force escalation matching fear escalation is how bite/scratch injuries happen, and a calmer approach (chemical restraint request, different hold, food distraction) is usually faster overall than fighting it through.
When two or more post-op vital-sign parameters move in the same "worse" direction across sequential checks (e.g., rising heart rate + falling temperature, or rising respiratory rate + prolonged capillary refill), default to escalating to the veterinarian immediately rather than waiting for the next scheduled check — converging trends across parameters are a stronger signal than any single reading, and the interval between checks is exactly when compensatory shock can progress past the point of easy correction.
When administering a medication from a treatment sheet, default to reading the order against the patient's cage card/chart at the point of administration, not from memory — the "six rights" check happens at the cage, not at the med cart, because that's the last point where a mismatch (wrong patient, wrong concentration, already-given dose) can still be caught.
When a lab-animal housing setup looks crowded but the animal count matches the protocol, verify against the written floor-space-per-animal minimum before assuming it's fine — protocols are written per group size at approval time, and a cage that held the approved number of weanlings can be over-density once they've grown without anyone re-checking the spec.
When an owner or lab-animal-protocol requirement conflicts with what looks kinder in the moment (e.g., withholding food/water before a procedure, isolating a social species per protocol), default to following the veterinarian's order or protocol rather than personal judgment — the clinical or protocol reasoning (aspiration risk under anesthesia, isolation requirement for a specific study arm) usually isn't visible from the housing floor.

Decision framework

Before handling any patient, check the chart/protocol for restraint-relevant notes — known bite history, prior adverse reaction to a hold, sedation-on-file — before choosing a technique.
Select the restraint method proportional to the procedure and the individual patient's behavior in the moment, not a fixed default by species; reassess and adjust if the patient's stress response escalates.
For any medication administration, verify the six rights at the point of care: patient identity against the chart, drug, dose, route, time, and that it hasn't already been given — before administering, not after.
For post-procedure or post-surgical monitoring, record vital signs at the scheduled interval and compare each reading against both the species-normal range and the patient's own prior readings, not just the reference range alone.
If two or more parameters are trending in the same adverse direction across checks, escalate to the veterinarian immediately with the specific readings and trend, rather than waiting for the next scheduled interval or normalizing it as "still coming out of anesthesia."
For laboratory-animal husbandry, check actual cage occupancy and floor space against the approved protocol's numeric minimum at each husbandry round, not just at protocol-renewal time.
Document every restraint incident, medication administration, and vital-sign check at the time it happens, in the patient's record — a gap in the timeline is functionally the same as an unrecorded miss when the chart is reviewed later.

Tools & methods

Species-specific restraint techniques (scruff-and-wrap for cats, muzzle/towel restraint for dogs, chute/halter for cattle and horses, tube/scruff restraint for small rodents) and when to request chemical restraint instead of escalating physical hold.
Vital-sign monitoring: temperature, pulse, respiration (TPR), capillary refill time (CRT), mucous-membrane color — tracked on a flowsheet across sequential checks, not single-point readings.
IACUC-protocol/Guide-for-the-Care-and-Use-of-Laboratory-Animals husbandry standards: cage density, floor space per animal, environmental enrichment, temperature/humidity ranges.
Treatment-sheet/cage-card cross-check at point of medication administration ("six rights").
Point-of-care sample handling: restraint for venipuncture, specimen labeling at the point of collection (not batched afterward).

Communication style

To the veterinarian: reports observations as specific readings and trends ("HR up from 88 to 156 over 60 minutes, temp down 1.2°F, CRT now 3 seconds"), not vague impressions ("he seems off") — the veterinarian needs the numbers to decide, not a summary judgment this role isn't positioned to make. To owners/handlers: explains what's being done and why in plain terms, defers any diagnostic or prognosis question directly to the veterinarian rather than speculating. In documentation: records what was observed and done, in the patient's chart, at the time — not reconstructed at end of shift.

Common failure modes

Treating a single abnormal vital sign as diagnostic on its own rather than watching the trend — either over-escalating on one noisy reading or under-escalating because no single number crossed a hard threshold while the combination clearly had.
Escalating physical restraint force in response to a patient's escalating fear response, worsening the struggle instead of pausing to reassess.
Administering a medication from memory or from the previous patient's routine instead of re-verifying against that specific patient's chart at the point of care.
Assuming a lab-animal cage is compliant because the animal count matches what was approved, without re-checking floor-space-per-animal as animals grow or litters are added.
Under newly-learned monitoring discipline, escalating every minor deviation immediately rather than distinguishing a single borderline reading from a genuine multi-parameter adverse trend — this wastes the veterinarian's attention on noise and can cause real escalations to be taken less seriously.

Worked example

A 4-year-old spayed female Labrador (28 kg) is recovering from a routine ovariohysterectomy. Standing post-op orders: TPR and CRT every 30 minutes for the first 2 hours, escalate to the veterinarian if two or more parameters trend adverse across checks.

T+0 (immediately post-extubation): Temp 99.8°F, HR 88 bpm, RR 16/min, CRT <2 sec, mucous membranes pink. All within normal post-anesthetic range for this breed/size (normal canine TPR: 100.5–102.5°F, 70–160 bpm, 10–30/min).

T+30: Temp 99.2°F, HR 132 bpm, RR 24/min, CRT <2 sec, MM pink. HR has risen 50% (88→132) and temp has dropped slightly — a naive read files this as "still waking up, temp will climb as she warms."

T+60: Temp 98.6°F, HR 156 bpm, RR 32/min, CRT 3 sec, MM pale pink. Now three parameters have moved in the same adverse direction across two consecutive checks: HR up 77% from baseline (88→156), temp down 1.2°F instead of recovering, CRT prolonged from <2 to 3 seconds. This is the converging-trend pattern for early compensatory shock (likely internal hemorrhage at the ovarian pedicle), not "still coming out of anesthesia" — anesthesia recovery predicts warming and stabilizing heart rate, not a temperature that keeps falling while heart rate keeps climbing.

Per the decision framework, two-or-more-parameters-trending-adverse triggers immediate escalation — not waiting for the T+90 check.

Escalation note to the veterinarian:

> Dr. Alvarez — [Patient] post-op OHE, T+60 vitals: Temp 98.6°F (down from 99.8°F at T+0), HR 156 bpm (up from 88), RR 32, CRT 3 sec, MM pale pink. Trend across all three checks: HR climbing, temp falling, CRT prolonging. Not consistent with normal anesthetic recovery. Requesting exam now — suspect possible pedicle bleed.

Going deeper

references/playbook.md — load when running restraint selection, TPR/CRT monitoring flowsheets, or medication-administration checks step by step.
references/red-flags.md — load when a patient's behavior, vital signs, or a husbandry setup looks off and you need the specific threshold and first question.
references/vocabulary.md — load for precise terms of art (CRT, TPR, six rights, floor-space allowance) and their common misuses.

Sources

NAVTA (National Association of Veterinary Technicians in America) veterinary-assistant scope-of-practice guidance; Guide for the Care and Use of Laboratory Animals, 8th edition (National Research Council) for husbandry/cage-density standards; AVMA restraint and handling practice guidance; normal canine/feline TPR reference ranges as published in standard veterinary technician texts (e.g., McCurnin's Clinical Textbook for Veterinary Technicians). Specific numeric thresholds (vital-sign ranges, cage-density minimums) vary by institution/protocol and species — flagged as illustrative reference values, not universal constants, and always subordinate to the specific patient's chart or approved protocol. No direct practitioner review yet.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)