Palliative Care

The standing claim

Timing of specialist involvement is the primary lever, and this is the one domain where a randomised trial tests it directly. Brought in early, alongside treatment, specialist palliative care improves quality of life and mood and reduces aggressive end-of-life care in cancer. That it also converts the final months into time spent where the patient chooses to be is the hope behind it, consistent with register data on hospital use, but not something a trial has shown. As of 2022 this is also, for the first time, a matter of law: an integrated care board must arrange palliative care provision (Health and Care Act 2022, s.21, amending the NHS Act 2006). The question this domain tracks is whether a discretionary duty becomes a resourced reality.

Where practice stands

The structural position changed in July 2022: palliative care was named explicitly, for the first time, in the list of services NHS commissioners are statutorily required to arrange (Health and Care Act 2022, s.21 → NHS Act 2006 s.3). But the duty is doubly qualified — to “such extent as [the board] considers necessary” and “as the board considers are appropriate” — so it is a duty to commission with discretion as to extent, and we have not yet seen data showing it has changed funding or access. NICE NG142 (End of life care for adults: service delivery) and the national Ambitions for Palliative and End of Life Care framework set the expected model. Identification remains a weak point: the community under-identification pattern is tracked in the pattern register.

What good looks like

“Good” is early, integrated specialist palliative care, and the evidence for earlier points the same way across designs. The landmark trial (Temel et al., 2010; 151 patients with metastatic lung cancer) showed that introducing specialist palliative care soon after diagnosis, alongside oncology, improved quality of life and mood, reduced aggressive end-of-life care, and was associated with longer survival. A nationwide Finnish register study of men dying of prostate cancer found specialist contact more than 30 days before death was associated with markedly less acute hospitalisation in the final month (21% vs 38%) and far greater hospital-at-home use (Carpén et al., 2026); it had no data on disease severity or preferences, so who was referred early may explain part of the difference. A systematic meta-review found 15 of 39 reviews reporting decreased hospital utilisation in line with patients’ preferences after advance care planning (Crooks et al., 2025).

But recording a preference is not the same as resourcing it. In a Madrid cohort of patients followed by home palliative care, most died at home whether or not a preference was documented: documentation itself was not independently associated with where people died (Reguillo et al., 2026). That points to the service, not the paperwork, though the study did not test the service itself. On delivery, a realist review of 42 international publications identifies five aspects of community palliative rehabilitation that could improve outcomes: early referral, a layered model matched to complexity, holistic assessment, services accessible throughout the palliative journey rather than the last days only, and information and education for patients and carers (Manson et al., 2025).

The exemplar is the model, chosen on outcomes: early, integrated specialist palliative care introduced well before the final weeks — the Temel/Carpén pattern — backed by a resourced service that can actually honour a stated preference.

The international benchmark is unusual for this publication: here the NHS is itself the exemplar. On the two serious cross-country measures of palliative-care quality the UK ranks first — 1st of 80 in the Economist Intelligence Unit’s 2015 Quality of Death Index, and still 1st in the 2021 expert assessment of the quality of death and dying, one of only six countries graded A (with Ireland, Taiwan, Australia, the Republic of Korea and Costa Rica; Finkelstein et al., 2021). The strength is structural — palliative care embedded in a universal system, a mature hospice movement, national policy. But the honest gap is place of death: among people over 65 dying with cancer across seven developed countries, 41.7% died in hospital in England versus 29.4% in the Netherlands (and 22.2% in the US), even though England spent the least of the seven on hospital care in the last six months of life (Bekelman et al., 2016). So the lesson to import is not a whole system but a specific lever — how the Netherlands enables more people to die at home — while Costa Rica’s grade A, the only one outside the high-income countries, suggests that this quality, as experts judge it, is reachable without high-income resourcing.

Through the lenses

This is the purest use-value domain in the publication. The measure is the lived quality of the time that remains and whether a person dies where they wished — not bed-days or activity units. McGilchrist’s caution is sharpest here: a good death cannot be decomposed into measurable variables; the advance-care-planning conversation is the intervention, and the DNACPR form is only the trace it leaves — exactly the illness-versus-disease distinction (Kleinman) and the phenomenological insistence that the patient’s experience is irreducible evidence. The VAT lens asks whether the final months added value on the patient’s own terms. Cross-domain, this meets acute frailty (D01) at the frailty–palliative boundary and OPAT (D05) at the question of what treatment is worth delivering at all near the end of life.

Confidence and limits

The statutory duty is firm in word but discretionary in extent. Temel is high-quality randomised evidence but cancer-specific (metastatic lung cancer) — extending it to non-cancer and frailty trajectories is a reasonable inference, not a demonstrated equivalence. Carpén is a national register study and Reguillo a single-city cohort — observational, single-jurisdiction signals, not confirmation. The realist review describes what could improve outcomes, not what has been shown to. On the international benchmark, two honest caveats: the quality-of-death indices measure expert-assessed system quality and perception, not population outcomes directly — and on the one hard outcome compared, place of death, England is mid-table rather than leading. The place-of-death figures are from 2010 cancer-decedent data, so treat the cross-country comparison as directional.

What we are watching

Whether the statutory duty translates into funding and genuinely equal access; community identification and registration of people in the last year of life; the palliative–OPAT interface; and whether England can close the place-of-death gap — enabling more people to die at home, as the Netherlands does — while protecting its top-ranked palliative-care quality.