The Canine Ledger

Managing Pain in a Terminal Dog

Written by thecanineledger

In crisis right now? Canada and US: call or text 9-8-8.

Once the goal stops being cure, one question replaces all the others: is he comfortable?

It is harder to answer than it sounds. A terminally ill dog is often already on medication, already slowed by illness, and already different from himself in ways that make the usual signals unreliable. Meanwhile you are watching more closely than you have ever watched anything, and being watched by someone who loves them turns out to be one of the least accurate ways for a dog’s pain to be measured.

This page is about that question, and only that question. It does not cover drugs or doses: those belong with the veterinarian who has examined your dog. It covers what to look for, what the evidence says about how reliably you can look, when to go back, and one honest thing about terminal pain that most sources leave out.

1. Start here: get a baseline

Before anything changes (a new medication, a dose adjustment, a decision) score your dog and write it down.

This matters more than it sounds, and the reason is uncomfortable. In a randomised, double-blinded, placebo-controlled trial of 58 dogs, owners registered improvement that force-plate measurement did not show 39.7% of the time. Veterinarians did the same thing 44.8% of the time. And the effect was “significantly enhanced with time” (Conzemius & Evans, JAVMA 2012).

Roughly two in five, growing as the weeks pass, in trained clinicians under trial conditions. Hoping for improvement produces the appearance of improvement.

You cannot think your way out of this. You can write things down before you start, so that “he seems better” can be checked against something.

The instrument to use is the Canine Brief Pain Inventory, owner-completed, free, and validated in osteoarthritis and separately in canine bone cancer, which few instruments can claim.

→ Download the CBPI from Penn Vet

It asks you to rate pain at its worst, least, and average over the last seven days, and how much it interferes with general activity, enjoyment of life, rising, walking, running, and stairs.

If your dog has cancer, there is also a Canine Symptom Assessment Scale, twelve symptoms, owner-completed, developed for dogs with solid tumours (Penn Vet). It is a symptom scale rather than a pain scale, but it is the most relevant owner-usable instrument for a dog with a tumour.

Our note: All WSAVA quotes here are from the 2014 edition, and a 2022 edition exists. The owner questionnaires we point to are published free by a veterinary school, but share the results with your vet.

2. What you are looking for

Terminal illness complicates the usual signs, because so much of what pain does (reduced activity, appetite loss, withdrawal) is also what the illness does. Some signals stay useful anyway.

The best-evidenced one is night-time restlessness. In a double-masked placebo-controlled crossover trial, dogs with osteoarthritis on pain medication showed decreased night-time activity between 11pm and 6am and increased daytime activity, compared with placebo (Gruen, Samson & Lascelles, Scientific Reports 2019). The same direction was found in dogs with bone cancer, where “least active actimetry significantly decreased after treatment… indicating improvement in night-time restlessness” (Monteiro et al., PLOS One 2018).

This is causal evidence, not a list item: treat the pain, the nights improve. Which makes the nights getting worse again one of the more meaningful signals that pain has stopped being controlled.

Posture and movement. Per the 2022 AAHA Pain Management Guidelines: “at rest, awkward limb positions may indicate the presence of joint pain”; “when standing and moving, dogs shift weight away from a painful limb”; “a ‘tucked up’ appearance can be an indicator of abdominal pain”; “postural straining can be an indicator of lower urinary tract or lower gastrointestinal tract pain.”

Reaction to being touched, especially somewhere he did not previously mind.

Vocalisation, but with a specific framing. The IAAHPC guidelines put it better than the generic advice does: “vocalization of any sort that is out of the ordinary for the patient may indicate that their pain and/or anxiety has become too much for them to bear” (IAAHPC). The change from his baseline is the signal, not the sound itself.

Licking a site, or, a distinct and less-known finding, copious licking of surfaces such as floors and walls, which is associated with gastrointestinal pain.

Two things that are not reliable signals

Panting. It is not an item in any validated canine pain instrument. It is worth mentioning to your vet, but it is at least as likely to reflect heat, stress, cardiorespiratory disease, or steroid treatment.

Heart rate, breathing rate, pupil size. Directly contradicted: “physiological data such as tachycardia, tachypnea, systemic blood pressure, mydriasis… have proven to be ineffective as specific indicators of acute or chronic pain” (Sander et al., Frontiers in Veterinary Science 2025).

And one thing that misleads in the other direction. Normal sociable behaviour does not rule pain out. From the same paper: “indicators of pain may be masked by behaviors characteristic of the species being observed, such as a dog wagging its tail and greeting an observer at the cage door despite being in significant pain.”

Fuller detail: → Reading Pain in an Old or Sick Dog

3. The clinic is a bad place to judge this

Worth knowing, because it explains a frustrating experience many owners have: your dog perks up at the vet, or conversely goes rigid, and the assessment there does not match what you see at home.

“Acute stress can override pain responses and even lower the tolerance of pain perception… especially if the dog experiences stress during the evaluation” (Sander et al. 2025).

This is precisely why owner-completed instruments exist, and why the WSAVA guidelines state that “owner assessments are the mainstay of the assessment of chronic pain in dogs” (WSAVA 2014).

Your observations at home are not a lesser substitute for a clinical assessment. For chronic pain they are the primary evidence, which is also why writing them down matters.

4. When to go back

There is no authoritative owner-facing list of triggers in the literature. What exists is better: thresholds built into validated instruments, published reassessment intervals, and drug-safety guidance.

On the CBPI, treatment success is conventionally defined as a reduction of at least 1 point in pain severity and at least 2 in pain interference from baseline. Failing to reach that is a defensible, non-emotional reason to say the plan is not working.

Reassessment intervals. AAHA guidance is that chronic-pain patients “would then be reevaluated regularly (every 3–6 months).” In a terminal dog, that interval is too long: this is one to agree with your vet directly.

Night-time restlessness returning: per §2, the signal with actual causal evidence behind it.

On anti-inflammatory medication, the FDA lists what to watch for: “vomiting,” “diarrhea,” “bloody or black stool,” “decreased appetite,” “decreased activity level,” and “yellowing of the whites of the eyes or the gums.” Its instruction is unambiguous: “stop giving the medication and call your veterinarian” (FDA). The FDA also advises baseline blood tests before long-term use, and never adjusting a dose yourself.

Ask for your own trigger list. AAHA guidance says owners should be given “specific instructions, both verbally and in writing, including when the next assessment is recommended,” and be told “the action to take” if adverse effects appear, but leaves the content to each practice. If nobody has given you one, ask. Its absence is worth raising.

And the standard the field sets, which you are entitled to hold your veterinarian to:

“Regardless of origin and mechanism, prolonged severe pain is an unacceptable state for animal hospice and palliative care patients.” — IAAHPC

Our note: We have no source for how often to reassess a terminal patient, so we give no number. If your dog has a pain crisis or can’t settle at all, call your vet or an emergency clinic.

5. The hard part: when pain relief is not enough

Most sources will tell you that pain in a dying dog can be managed. For many dogs it can. For some, the evidence says otherwise, and an owner told “he’s on his medication so he’s comfortable” deserves to know that.

A trial of dogs with naturally-occurring bone cancer, using a stepwise oral protocol over 28 days, concluded:

“Dogs with bone cancer are affected by widespread somatosensory sensitivity characterized by peripheral and central sensitization and have a deficient inhibitory system. This severe pain is mostly refractory to palliative analgesic treatment, and the latter was only detected by specific and sensitive outcomes.” — Monteiro et al., PLoS ONE 2018

Read the study’s size honestly: thirteen dogs, open-label, single-arm. That is pilot-scale, and it is one tumour type. Treat it as a signal, not a settled finding.

But the signal matters. It says that in bone cancer the pain involves central sensitisation, a change in how the nervous system processes pain, which ordinary oral analgesia does not reach. It also found that the modest treatment effect was detectable only by sensitive objective measures, and that agreement between owners’ and vets’ pain scores varied widely across the 28 days.

What to do with this, practically:

  • If your dog has bone cancer or another painful malignancy, ask directly whether the current plan is likely to be sufficient, and what the options are if it is not, including referral to someone who specialises in pain management.
  • If you are being reassured that he must be comfortable because he is medicated, that reassurance is not automatically true.
  • And if adequate comfort genuinely cannot be achieved, that is itself the answer to a different question. The hospice guidelines are explicit that hospice does not support a natural death unless comfort can be delivered, and that “prolonged severe pain is an unacceptable state.” At that point the conversation is no longer about pain management.

→ How to Know When It’s Time · → Hospice & Palliative Care for Dogs

Our note: Section 5 rests on a small study. It is no reason to distrust a pain plan, but it is a reason to keep telling your vet what you see. We found no sources on comfort measures like bedding, footing and warmth.

6. What we could not verify

Commonly saidWhat we found
“Panting means he’s in pain.”Not an item in any validated instrument; a targeted search for a primary study returned only blogs and commercial sites. Non-specific.
“Watch for a raised heart rate or dilated pupils.”Contradicted. Those measures are described as ineffective as specific pain indicators.
“Read his face.”There is no validated canine grimace scale. Cats, horses, sheep, pigs and rodents have one; dogs do not.
“Hospice pain protocols are evidence-based.”The field’s own guidelines state there is “no ‘gold standard’ for assessing animal pain, and considerable research is still needed.”
A validated end-of-life pain instrument for dogsNone exists. The 2016 AAHA/IAAHPC end-of-life guidelines name no pain instrument at all. The CBPI, validated separately in bone cancer, is the closest available.

Sources

Always confirm with your vet before acting on anything on this page.