Pain in Somebody Who Cannot Tell You

Pain in somebody who cannot tell you: the myth, the tools, and the baseline nobody writes down

Let us start with the belief that does the most damage, because it is still held and it is wrong.

People with intellectual disability do not experience less pain than anybody else. They are less able to report it. Those two things are completely different, and conflating them has led to decades of under-treatment — surgery with inadequate analgesia, fractures managed with paracetamol, dental abscesses left for months, and dying people whose distress was recorded as behaviour.

The same error, in a different costume, applies to people with advanced dementia, to people who have lost speech after a stroke, and to anybody whose face and body do not produce the signals a clinician was trained to look for.

Pain self-report is the gold standard in every pain guideline in the world. When self-report is unavailable, the guidelines do not say to assume there is no pain. They say to use a hierarchy of other methods. In practice, what frequently happens instead is that the absence of a complaint is treated as the absence of a problem.

This guide is about doing it properly. It matters throughout somebody’s life and it matters most at the end of it, when pain is more likely, the person is less able to communicate than ever, and dose decisions depend entirely on somebody watching carefully.

We support people with disability in homes across New South Wales. We are not a clinical service, and nothing on this page is medical advice. Pain assessment tools are clinical instruments and analgesia is a prescribing decision — what a support team can do is observe, document and escalate well, which is the part that most often fails.

The hierarchy of pain assessment

When somebody cannot self-report, pain assessment works through five levels, used together rather than in sequence.

One: self-report in any form. Before concluding somebody cannot self-report, check properly. Can they indicate yes and no? Point to a body part? Use a faces scale? Use their communication device? Many people assumed to be unable to report pain can do so with the right method and enough time. This step gets skipped constantly.

Two: look for a likely cause. If the person has a condition that is known to be painful, assume pain until proven otherwise. A hip that has subluxed, a pressure injury, a fracture, an infection, a mouthful of untreated decay — these hurt, and the fact that nobody has complained is not evidence against them.

Three: observe behaviour. This is where the tools below come in.

Four: proxy report from people who know the person well. A parent or a long-term worker saying “that is her pain face” is clinical information. It is not a feeling. It is pattern recognition built over years, and it should be weighed accordingly.

Five: an analgesic trial. The workhorse of this field and the most under-used step. Where pain is suspected but cannot be confirmed, a clinician can prescribe regular analgesia for a defined period and observe whether the behaviour changes. If the agitation stops when the person is given regular paracetamol, that is diagnostic information.

Asking for an analgesic trial is one of the most useful requests a family or provider can make. It is a specific, concrete thing to put to a GP: this behaviour started eleven weeks ago, we think it may be pain, can we trial regular analgesia for two weeks and see whether it changes?

The tools

Three matter, and they do different jobs.

The Abbey Pain Scale

Australian, developed for measuring pain severity in people with late-stage dementia who cannot verbalise, and published by Abbey and colleagues in 2004. It is an observational tool completed by a caregiver or observer and takes about a minute.

Six domains, each scored from absent (0) through mild (1) and moderate (2) to severe (3):

1. Vocalisation
2. Facial expression
3. Change in body language
4. Behavioural change
5. Physiological change
6. Physical change

Total score 0 to 18, with bands of 0–2 no pain, 3–7 mild, 8–13 moderate, and 14 or above severe. The rater also records whether the pain appears chronic, acute, or acute on chronic.

One honest caveat worth knowing: the physiological change item has shown psychometric weakness in evaluation, and in some studies so has the body language item — removing the physiological item improved reliability in one analysis. That does not invalidate the tool, which is widely used and genuinely useful, but it is a reason to treat the total as a prompt and a trend rather than a precise measurement.

PAINAD

Pain Assessment in Advanced Dementia. Five items — breathing, negative vocalisation, facial expression, body language, and consolability — each scored 0 to 2, giving a total of 0 to 10. Shorter than the Abbey and widely used in acute and residential settings.

DisDAT — the one that matters most here

The Disability Distress Assessment Tool is different in kind from the other two, and it is the one built for this population.

DisDAT is not a score. It is a baseline document. You record, in detail, what the person looks like when they are content, and what they look like when they are distressed, across domains including appearance, vocal signs, habits and mannerisms, posture, body language, and observable signs.

Then, when something changes, you have something to compare it to.

That is the entire insight and it is the thing that unlocks pain assessment in people with severe or profound intellectual disability and communication disability. A score of 9 on a scale means very little in somebody whose baseline presentation includes rocking, vocalising and a tense posture. A documented difference from their own recorded content state means a great deal.

If you do one thing after reading this guide, complete a DisDAT-style baseline for the people you support who cannot report pain. It takes a conversation with the people who know them best, it costs nothing, and it is the document that makes every subsequent clinical assessment possible. Australian work examining pain and distress in people with severe and profound intellectual disability — including research in Victoria published in the Journal of Policy and Practice in Intellectual Disabilities in 2022 — has looked specifically at the utility of behavioural assessment tools in this group, and the recurring theme is that detection depends on structured observation rather than intuition.

Others worth knowing

FLACC and its revised version, used in children and in some adults with cognitive impairment. The Non-Communicating Adult Pain Checklist and the Chronic Pain Scale for Nonverbal Adults with Intellectual Disabilities are both designed for this population and may be used by specialist services.

The specific tool matters less than the fact that one is being used consistently by people who know the person, against a recorded baseline.

What pain looks like when it cannot be said

Behaviours that commonly signal pain in people who cannot report it:

• Facial expression — grimacing, frowning, a tightened or frightened look, eyes closed or screwed up, a look the family will recognise and name
• Vocalisation — crying, moaning, groaning, calling out, noisy breathing, new or changed sounds, or in some people the opposite: going silent
• Guarding — protecting a body part, resisting movement or touch in one area, flinching
• Repeated action toward one place — hand going to the same spot, hitting the same part of the head, rubbing
• New or increased self-injury
• Agitation, restlessness, pacing
• Aggression, particularly during care tasks that involve moving the painful part
• Withdrawal — going quiet, not engaging, sleeping more, refusing activities previously enjoyed
• Resisting care that was previously accepted
• Changed posture or position preference
• Reduced appetite or refusing food
• Disturbed sleep
• Increased spasticity or tone
• Autonomic changes — sweating, pallor, flushing, raised heart rate
• In spinal cord injury at or above T6, autonomic dysreflexia — which may be the only signal available
• Simply being “not themselves”, which from somebody who has known the person for years is a legitimate observation

Where to look for the cause

If pain is suspected, this is the systematic list. It is long because the common causes are mundane and get missed.

Mouth and teeth. The single most commonly missed cause. Dental decay, abscess, gum disease, broken teeth, ulcers, oral thrush, ill-fitting dentures. Dental access for people with intellectual disability in Australia is poor, and an abscess can produce months of distress that gets managed as behaviour.

Ears. Infection, impacted wax, foreign body.

Constipation and faecal impaction. Extremely common and extremely painful.

Urinary. Infection, retention, stones.

Reflux and oesophagitis. Common in people with cerebral palsy, dysphagia and in those who are tube fed. Genuinely painful and often silent.

Pressure injuries, including deep tissue injury that looks like a bruise.

Feet. Ingrown toenails, fungal infection, poorly fitting footwear, pressure under an orthosis.

Hips. Subluxation and dislocation in cerebral palsy is a classic cause of severe, chronic, commonly missed pain. If somebody with cerebral palsy has new pain behaviour and nobody has imaged the hips, that is the question to ask.

Fractures the person cannot report, particularly where bone density is low — after spinal cord injury, in people with limited mobility, on long-term anticonvulsants.

Spasticity, contractures and positioning. Being left in a position that hurts.

Scoliosis and its consequences.

Gastrostomy site problems — infection, granulation, leakage, buried bumper.

Shunt malfunction in people with hydrocephalus, which can present as headache and change in alertness.

Menstrual pain, which gets overlooked routinely.

Neuropathic pain, including after spinal cord injury, after a stroke, and in diabetes.

Headache, including medication overuse headache.

And the cause relevant to this library: at the end of life, the underlying condition itself.

At the end of life specifically

Three things change, and they all make assessment harder at exactly the point it matters most.

Pain is more likely and more severe. The underlying disease progresses, mobility reduces, pressure areas develop, and existing pain often escalates.

The route of administration changes. As swallowing fails, oral analgesia stops being an option, and medications move to subcutaneous delivery — often by syringe driver. That transition needs planning before it is needed, which is what anticipatory prescribing is for.

The person becomes less able to communicate than they already were. Reduced consciousness removes even the limited signals that existed. At that point, dose titration depends almost entirely on observation by the people at the bedside, which makes the quality of that observation a clinical input.

Which produces the single most important operational point in this guide: in the final days, the support workers and family at the bedside are the pain assessment instrument. Their observations, recorded with times, are what the palliative care nurse uses to decide whether the current dose is adequate. Vague reports produce unchanged doses. Specific ones produce adjustments.

And the concept worth knowing: total pain. Not all distress at the end of life is physical. Fear, unresolved grief, loneliness, spiritual distress and unfinished business all produce suffering that looks like pain and does not respond to analgesia. A person who is frightened and alone at 2am needs a different intervention from a person whose morphine dose is too low. Distinguishing them is clinical work, but the people at the bedside are the ones who can report the context that allows it.

What a support team should actually do

Complete a baseline. DisDAT-style, for anybody who cannot report pain. What they look like content, what they look like distressed.

Describe, do not interpret. “Grimaced and pulled his right leg away each time it was moved during washing, four times this morning” is clinical information. “Agitated behaviour, appeared distressed” is not.

Record with times and dates. Pattern matters. Pain that is worse in the morning, or on movement, or after meals, or at night, points somewhere.

Note what helps and what does not. Repositioning. A warm pack. Being left alone. A dose of analgesia already prescribed.

Report changes as changes, stating what the baseline was.

Use the same tool consistently if one has been introduced, and make sure the people completing it know the person.

Escalate to the GP or palliative care service, not to a behaviour support referral, when the change is new and unexplained. Behaviour support has a role; it is not where an undiagnosed pain problem belongs, and the weeks it takes are weeks of untreated pain.

Ask for an analgesic trial where pain is suspected and unconfirmed.

Push back on the sensitivity myth if you hear it. Politely, and with the correction: the issue is reporting, not sensation.

What not to do

Do not conclude there is no pain because nobody complained.

Do not route new unexplained distress to a behaviour support plan first. Medical causes should be excluded before a functional behaviour assessment begins, and a plan built for a symptom that was an abscess wastes months.

Do not give PRN sedation for distress of unknown cause. Sedating somebody in pain removes the only signal available and leaves the pain untreated. It also raises a restrictive practice question, since chemical restraint is medication used primarily to influence behaviour rather than to treat a diagnosed condition.

Do not assume the dose is right because the person has stopped making noise. Exhaustion and sedation both look like comfort.