Who needs palliative care? The surprise question, SPICT, NECPAL and prognosis
- Assessment
Most patients who would benefit from palliative care are never identified. A few simple questions at the bedside find them years earlier.
What it is
Identifying palliative needs early is a screening task, and a different task from predicting the date of death. The surprise question asks the clinician: «Would I be surprised if this patient died in the next 12 months?». A «no» flags the patient for a needs review. SPICT (Supportive and Palliative Care Indicators Tool, University of Edinburgh, with a Spanish version) is a one-page list of general indicators of declining health, such as unplanned admissions, falling function, weight loss and persistent symptoms, plus clinical indicators for cancer, organ failure, dementia and frailty; it says plainly that it is not a prognostic tool. NECPAL CCOMS-ICO, developed in Spain from the Gold Standards Framework guidance and SPICT, starts with the surprise question and counts a patient as positive when the answer is «no» and at least one indicator of severity or progression is present; it includes psychological and spiritual items and is written in Spanish. Function scales describe where the patient stands. The Karnofsky performance status rates function from 100 (normal) to 0 (dead); the Palliative Performance Scale (PPS) adapts it for palliative care, adding ambulation, self-care, intake and consciousness in steps of 10. Their trend over time says more than one value. All these tools are moderately accurate and most false positives are still patients with needs, so they open a conversation and a plan, never a countdown.
What to assess
- Unplanned hospital admissions, especially repeated ones in the last months.
- Falling function: in bed or chair for much of the day, needing help with self-care.
- Unintended weight loss and poor intake.
- Persistent symptoms despite optimal treatment of the underlying illness.
- Disease-specific markers of advanced illness (for example heart failure with breathlessness at rest, oxygen-dependent COPD, advanced dementia).
- The patient or family asking about palliative care, or about limiting or stopping treatment.
- Psychological, social and spiritual distress, and caregiver strain.
How it is done (and why)
- Ask the surprise question at clinic reviews, at admission and at discharge of any patient with advanced chronic illness.
- If the answer is «no», look for SPICT or NECPAL indicators of decline: a «no» plus an indicator means unmet needs are likely.
- Score function (PPS or Karnofsky) and record it, so that the next score shows the slope.
- Assess needs in every domain (symptoms, mood, family, money, spiritual) rather than stopping at «is he dying?».
- Estimate prognosis as a range and decide what it changes in the plan now.
- Ask how much the patient wants to know, then share the range with honest uncertainty.
- Turn the result into action: a care plan, advance planning and, if needed, referral to the palliative team.
Identification tools vs function scales
| Tool | What it asks or measures | What a positive or low result means | What it is not |
|---|---|---|---|
| Surprise question | Would I be surprised by death within 12 months? | «No»: review palliative needs | A survival prediction for the individual |
| SPICT | General indicators of decline plus clinical indicators by disease | Indicators present: assess needs and plan care | A prognostic tool (it says so itself) |
| NECPAL CCOMS-ICO | Surprise question plus indicators of severity and progression, including psychosocial items | «No» plus at least one indicator: likely palliative needs | A test that excludes need when negative |
| Karnofsky performance status | Global function from 100 (normal) to 0 (dead) | Lower scores: shorter survival on average | A palliative-specific scale |
| Palliative Performance Scale (PPS) | Ambulation, activity, self-care, intake and consciousness, 100% to 0% | Low and falling: death likely within weeks | Meaningful as a single isolated value |
Findings that change the plan
- A falling PPS over weeks is a stronger signal than any single value; a low and falling PPS points to weeks or less.
- Repeated admissions for the same organ failure despite optimal treatment move the patient onto the organ-failure trajectory's later part.
- A «no» to the surprise question in a patient who has never discussed goals is the cue for a goals-of-care conversation.
- Prognosis in non-cancer illness is the least certain; uncertainty is a reason to plan earlier, not to wait.
- When the patient asks «how long?», first ask what is behind the question (a wedding, a trip, a fear), then answer in a range.
Next step
- Record the surprise question answer and any SPICT or NECPAL indicators in the notes.
- Score and date PPS or Karnofsky at each review to see the trend.
- Complete a needs assessment in all domains and address the most distressing need first.
- Arrange a goals-of-care conversation with the patient and family.
- Refer to the palliative team if needs are complex or the decline is fast.
Common traps
- SPICT and NECPAL predict when the patient will die: they identify likely palliative needs, and their prognostic accuracy is only moderate
- Waiting until prognosis is certain before raising palliative care: in organ failure certainty comes too late, so uncertainty is the reason to start
- A single low PPS score fixes the prognosis: the trend over time and the clinical context carry most of the information
- Giving one exact number when asked how long: clinicians usually overestimate survival, so speak in ranges and plan for the worst and the best
- A false positive on the surprise question is wasted effort: most flagged patients still have needs worth addressing
Test yourself
Think of the answer, then tap the question to see it.
An 82-year-old with COPD has had three admissions in six months and is breathless at rest. You would not be surprised if he died this year. What does that answer mean?
It flags likely palliative needs: look for indicators of decline, assess needs in every domain and start a goals-of-care conversation. It is not a prediction of the date of death.
How does NECPAL CCOMS-ICO decide that a patient is positive?
The surprise question is answered «no» (not surprised) and at least one indicator of severity or progression is present.
Why is the trend of the PPS more useful than one value?
A single score can reflect a transient illness; a steady fall over weeks shows the trajectory, and a low, falling PPS points to death within weeks.
A patient asks «how long do I have?». What is a good way to answer?
Ask what lies behind the question and how much she wants to know, then give a range (weeks to months) with honest uncertainty, and say what you hope for and what to prepare for.
Elsewhere in DurgiPedia
Connections
Educational content; it does not replace supervised training, your institution's protocols or the anaesthetist's or intensivist's decision. It contains no doses and no ventilator or infusion settings.
Sources: The SPICT (Supportive and Palliative Care Indicators Tool) (SPICT, University of Edinburgh, 2026); Screening tools for the early identification of palliative care needs in patients with advanced chronic conditions: an updated systematic review (Journal of Clinical Medicine (PubMed Central), 2026); Last Days of Life (PDQ), Health Professional Version (National Cancer Institute, 2025); Illness trajectories and palliative care (BMJ (PubMed Central), 2005); Palliative care (fact sheet) (World Health Organization, 2020).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.
