Palliative care is specialist help with symptoms, worries and decisions, and can be given alongside cancer treatment, not only near the end of life. Guidelines agree on early referral in advanced cancer but differ on how: one approach refers everyone early in the illness, another refers when a person has severe symptoms, distress or decisions to make. You can also ask for a referral yourself.
This page explains when a referral to specialist palliative care is recommended, so that patients and families know what to ask for; your own team's pathway and local services come first. ASCO (2024 update, read as an abstract) says oncology clinicians should refer patients with advanced solid tumours and haematological malignancies to specialised interdisciplinary palliative care teams early in the course of the disease, alongside active treatment, and that cancer programmes should provide specialist palliative care for patients with unaddressed physical, psychosocial or spiritual distress. ESMO's 2018 position paper says supportive and palliative care should start at diagnosis. Neither text, as read, gives a number of weeks; the full ASCO guidelines could not be read. The second position refers by need. An international Delphi panel of 60 experts (Hui 2016) agreed 11 major criteria, any one of which could be enough to trigger a referral; nine are needs and two are times (within 3 months of an advanced cancer diagnosis where median survival is a year or less, and progression despite second-line treatment). A severe symptom in these criteria is one scored 7 to 10 on a ten-point scale. NICE NG142 sets no time: it asks services to identify adults approaching the end of life and to assess their holistic needs, and its committee concluded that the evidence on early referral was not sufficient for a recommendation on timing. The trials: in Temel 2010 (151 patients, metastatic non-small-cell lung cancer, one specialised centre) early palliative care improved quality of life and depressive symptoms at 12 weeks, and median survival was 11.6 against 8.9 months; a median is the middle of a group, so half of each group lived longer than its figure. A Cochrane review (2017, seven trials, 1,614 participants) found a small improvement in quality of life on low-certainty evidence and no clear difference in survival. A Danish trial that referred by screened need (DanPaCT, 297 patients) found no effect on its primary outcome. A US trial of stepped palliative care (507 patients with advanced lung cancer) found quality of life no worse with visits stepped up by need than with visits every 4 weeks, with a mean of 2.4 visits against 4.7.
Why: In several randomised trials, people who saw a palliative care team from soon after diagnosis reported better quality of life than those given standard cancer care; the trials are set out below.
“Oncology clinicians should refer patients with advanced solid tumors and hematologic malignancies to specialized interdisciplinary palliative care teams that provide outpatient and inpatient care beginning early in the course of the disease, alongside active treatment of their cancer.”
American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 · Recommendation (rating not given in the abstract)
Why: The Commission lists this belief among the barriers to bringing palliative care into cancer care, and a review by authors of the consensus criteria calls the stigma attached to palliative care a major reason referrals are delayed.
“Other barriers include the common misconception that palliative care is end-of-life care only, stigmatisation of death and dying, and insufficient infrastructure and funding.”
The Lancet Oncology (Kaasa et al.) 2018 · Commission report (not graded)
Why: The panel wrote that, to make the best use of scarce palliative care resources, patients need to be referred at the right time for the right reasons. It also wrote that the criteria should add to clinical judgement and not replace it.
“Panellists reached consensus on 11 major criteria for referral: severe physical symptoms, severe emotional symptoms, request for hastened death, spiritual or existential crisis, assistance with decision making or care planning, patient request for referral, delirium, spinal cord compression, brain or leptomeningeal metastases, within 3 months of advanced cancer diagnosis for patients with median survival of 1 year or less, and progressive disease despite second-line therapy.”
The Lancet Oncology (Hui et al.) 2016 · International Delphi consensus (60 experts; agreement of 70% or more)
Why: Patient request is one of the 11 major criteria (55 of the panellists who answered, 95 per cent, agreed), and the authors say any one criterion would be reason enough.
“Panelists reached consensus on 11 major criteria for referral (Table 2)—the presence of any criterion would be sufficient reason for a patient to be referred.”
Cancers (Hui, Heung and Bruera) 2022 · Narrative review reporting the Delphi consensus
Why: The review says many symptoms remain under-reported and under-diagnosed, and that referral often happens haphazardly, with some patients not referred at all and many others too late.
“It requires four components: (1) routine screening of supportive care needs at the oncology clinics, (2) establishment of institution-specific consensual criteria for referral, (3) having a system in place to trigger a referral when patients meet criteria, and (4) availability of outpatient palliative care resources to provide timely access (Figure 1).”
Cancers (Hui, Heung and Bruera) 2022 · Narrative review (not a guideline)
Why: ASCO addresses this to cancer care programmes, so that specialist help exists for distress that is going unaddressed; its abstract gives no further reason.
“For patients with cancer with unaddressed physical, psychosocial, or spiritual distress, cancer care programs should provide dedicated specialist palliative care services complementing existing or emerging supportive care interventions.”
American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 · Recommendation (rating not given in the abstract)
Why: NICE says this will enable the right support to be provided when it is needed. It describes the assessment as one that considers all aspects of a person's wellbeing, and their spiritual, health and social care needs.
“If it is thought that an adult is approaching the end of their life, carry out an initial holistic needs assessment with the person and document this.”
NICE 2019 · Not graded (NICE NG142 1.2.1)
Why: ASCO says clinicians may do this to give carers additional support. A review by authors of the consensus criteria says family carers benefit from a lasting relationship with the palliative care team, which can give them education, psychological care and resources.
“Oncology clinicians from across the interdisciplinary cancer care team may refer the caregivers (eg, family, chosen family, and friends) of patients with cancer to palliative care teams for additional support.”
American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 · Recommendation (rating not given in the abstract)
Why: The 2024 update names this group and says clinicians may refer them. Its abstract gives no reason, and the full text could not be read.
“Clinicians caring for patients with solid tumors on phase I cancer trials may also refer them to specialist palliative care.”
American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 · Recommendation (rating not given in the abstract)
How care is usually carried out, read from the guidelines and standards cited. Your own team's protocol and their judgement come first; this page is for understanding it, not for doing it yourself.
Every source read agrees that specialist palliative care belongs alongside cancer treatment for people with advanced cancer and should start early rather than at the end. They disagree on what triggers the referral, and both positions are shown here. Position one, by time and for everyone: ASCO's 2024 update, read as an abstract, says oncology clinicians should refer patients with advanced solid tumours and haematological malignancies to specialist teams beginning early in the course of the disease, alongside active treatment; ESMO's 2018 position paper says supportive and palliative care should start at the time of diagnosis. Neither gives a number of weeks in the text that could be read, and the full ASCO guidelines of 2017 and 2024, which might set one, were not reachable. The trials behind this position enrolled people soon after diagnosis: within 12 weeks of a new primary tumour or a diagnosis of progression in Vanbutsele 2018, and newly diagnosed in Temel 2010 and Temel 2017. Position two, by need: the international Delphi consensus (Hui 2016, 60 experts) defines 11 major criteria, nine of them needs and two of them times, says the presence of any one could be sufficient to trigger a referral, and its authors argue that there are not enough specialist teams to see everyone from diagnosis. NICE NG142 sets no time and asks for systematic identification of adults approaching the end of life and a holistic needs assessment; its committee, having reviewed studies of early against late referral, concluded that the evidence was not sufficient to recommend a time. The 2024 ASCO abstract carries both ideas, suggesting early involvement especially for patients with uncontrolled symptoms and quality of life concerns. The evidence for referral by time: Temel 2010 (151 patients, at what the Lancet Oncology Commission calls a highly specialised institution) found better quality of life (98.0 against 91.5 on a 0 to 136 scale), fewer patients with depressive symptoms (16 against 38 per cent) and longer median survival (11.6 against 8.9 months; half of each group lived longer than its median). Zimmermann 2014 (461 patients, 24 clinics) missed its primary quality of life endpoint at 3 months (p = 0.07) and found differences favouring palliative care at 4 months on all but one of its measures. Temel 2017 (350 patients) found improvement at 24 weeks but not at 12. Vanbutsele 2018 (186 patients, Belgium) found better quality of life at 12 weeks. ENABLE III (207 patients) compared starting at once with starting 3 months later: patient-reported outcomes did not differ, and one-year survival was 63 against 48 per cent. The Cochrane review (2017) rated the quality of life benefit small and of low certainty and found no clear survival difference (hazard ratio 0.85, 95% CI 0.56 to 1.28, very low certainty). The evidence for referral by need: DanPaCT (297 patients screened for need) found no effect on the primary outcome; a phase 2 study of symptom screening with triggered referral (STEP, 116 patients) found it feasible; and stepped palliative care (507 patients with advanced lung cancer) was non-inferior to visits every 4 weeks for quality of life at 24 weeks with 2.4 against 4.7 visits, though non-inferiority was not shown for days in hospice (19.5 against 34.6). No trial read compares referral for everyone at diagnosis with referral against the consensus criteria.
American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 · ESMO (Annals of Oncology, Jordan et al.) 2018 · The Lancet Oncology (Kaasa et al.) 2018 · The Lancet Oncology (Hui et al.) 2016 · Cancers (Hui, Heung and Bruera) 2022 · NICE 2019 · NICE (developed by the National Guideline Centre) 2019 · New England Journal of Medicine (Temel et al.) 2010 · The Lancet (Zimmermann et al.) 2014 · Journal of Clinical Oncology (Temel et al.) 2017 · The Lancet Oncology (Vanbutsele et al.) 2018 · Journal of Clinical Oncology (Bakitas et al.) 2015 · Cochrane Database of Systematic Reviews (Haun et al.) 2017 · Palliative Medicine (Groenvold et al.) 2017 · Journal of the National Comprehensive Cancer Network (Zimmermann et al.) 2021 · JAMA (Temel et al.) 2024
Where it differs: One palliative care visit within 4 weeks of enrolment, then visits only at a change of cancer treatment or after a hospital admission, with quality of life measured every 6 weeks; a fall of 10 points or more from the starting score steps the patient up to visits every 4 weeks. Tested in people diagnosed with advanced lung cancer within the previous 12 weeks.
Evidence: Randomised non-inferiority trial, 507 patients, three US academic centres: quality of life at week 24 non-inferior (adjusted mean 100.6 with stepped care against 97.8 with early palliative care), mean visits 2.4 against 4.7, end-of-life care communication non-inferior, and "noninferiority was not demonstrated for days in hospice (adjusted mean, 19.5 with stepped palliative care vs 34.6 with early palliative care; P = .91)".
Where it differs: Symptoms are scored at each oncology visit; a moderate to severe score sends an email to a palliative care nurse, who telephones the patient and offers early palliative care (STEP). In DanPaCT (Denmark) consecutive patients were screened with a quality of life questionnaire and those above a threshold were randomised to referral or standard care.
Evidence: STEP phase 2 (116 patients; the comparison is between people who accepted and received palliative care and people who did not): 70 screened positive and 39 of them received early palliative care during the 6-month study; mood and symptom control improved over time in those who received it and worsened in those who did not, with no difference in quality of life or satisfaction with care. DanPaCT (145 against 152 patients): no effect on the primary outcome (-4.9 points, 95% CI -11.3 to +1.5, p = 0.14); the authors wrote "We did not observe beneficial or harmful effects of early specialist palliative care, but important beneficial effects cannot be excluded."
Journal of the National Comprehensive Cancer Network (Zimmermann et al.) 2021 · Palliative Medicine (Groenvold et al.) 2017
Where it differs: ENABLE III gave an in-person palliative care consultation, six weekly structured telephone coaching sessions with a nurse and monthly follow-up, started either early after enrolment or 3 months later, at a US cancer centre, a veterans' hospital and community outreach clinics.
Evidence: 207 patients: patient-reported quality of life, symptom impact and mood did not differ between the groups; one-year survival was 63 per cent in the early group and 48 per cent in the delayed group (p = 0.038). The authors call the mechanism of any survival effect an open research question.
Where it differs: NICE NG142 asks services for systems to identify adults likely to be approaching the end of life, naming the Gold Standards Framework, the Amber Care Bundle and the Supportive and Palliative Care Indicators Tool as examples, followed by a holistic needs assessment and care from a team with the skills to meet the needs found. It gives no number of weeks from diagnosis.
Evidence: NICE does not grade its recommendations. Its evidence review on the timing of referral compared early with late referral; the committee found more benefits for early referral than late but judged the evidence too limited to recommend a time, noting that all the studies were in people with cancer and that several came from health systems unlike the UK's. It recommended a holistic needs assessment instead.
NICE 2019 · NICE (developed by the National Guideline Centre) 2019
Where it differs: In Belgium, where psychologists, social workers, dieticians and specialist nurses are part of standard oncology care, Vanbutsele 2018 added early and systematic palliative care to that standard.
Evidence: 186 patients: overall quality of life at 12 weeks was 61.98 with early palliative care against 54.39 with standard care (difference 7.60, 95% CI 0.59 to 14.60, p = 0.03). The authors concluded that systematic integration was "more beneficial for patients with advanced cancer than palliative care consultations offered on demand, even when psychosocial support has already been offered".
Guidelines ask for palliative care early in advanced cancer, but there are too few specialist teams to see everyone every month. In one randomised trial in advanced lung cancer, a single early visit followed by more visits only when a person's quality of life score fell gave quality of life no worse than monthly visits, with about half the appointments. It has not yet been tested in other cancers.
| Recommendation | Grade, as printed | Source |
|---|---|---|
| Refer patients with advanced solid tumours and haematological malignancies early in the course of the disease, alongside active treatment (2024 update) | Recommendation (rating not given in the abstract) | American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 |
| Early palliative care involvement, especially for patients with uncontrolled symptoms and quality of life concerns (2024 update) “The Expert Panel suggests early palliative care involvement, especially for patients with uncontrolled symptoms and QOL concerns.” | Expert Panel suggestion (rating not given in the abstract) | American Society of Clinical Oncology (Journal of Clinical Oncology, Sanders et al.) 2024 |
| Supportive and palliative care should start at diagnosis “They should start at the time of diagnosis and continue through to end-of-life or survivorship.” | Position paper (not graded) | ESMO (Annals of Oncology, Jordan et al.) 2018 |
| Early specialist palliative care alongside tumour-directed treatment promotes patient-centred care “Early delivery of patient-directed care by specialist palliative care teams alongside tumour-directed treatment promotes patient-centred care.” | Commission report (not graded) | The Lancet Oncology (Kaasa et al.) 2018 |
| Eleven major referral criteria and 36 minor criteria, agreed by consensus “Consensus was defined by an a-priori agreement of 70% or more.” | International Delphi consensus (60 experts; agreement of 70% or more) | The Lancet Oncology (Hui et al.) 2016 |
| Any one major criterion could be enough to trigger a referral “The presence of any major criterion alone could be sufficient to trigger an outpatient palliative-care referral.” | International Delphi consensus (authors' statement) | The Lancet Oncology (Hui et al.) 2016 |
| Minor criteria support a referral together and may not be enough singly “The panellists also identified 36 minor criteria, which might be used to support referral in aggregate, but might not individually be adequate to trigger a referral in the absence of other factors.” | International Delphi consensus (authors' statement) | The Lancet Oncology (Hui et al.) 2016 |
| The criteria guide and do not replace clinical judgement “Importantly, these criteria should serve as guiding principles, and should augment rather than replace clinical judgment.” | International Delphi consensus (authors' statement) | The Lancet Oncology (Hui et al.) 2016 |
| Most of the major criteria are needs rather than times “Importantly, 9 of these 11 criteria were needs-based instead of time-based.” | Narrative review reporting the Delphi consensus | Cancers (Hui, Heung and Bruera) 2022 |
| Early palliative care for everyone from diagnosis is not possible with present resources (the authors' argument for referral by need) “Due to the scarcity of palliative care resources, it is not possible to provide early palliative care for all patients with advanced disease from around the time of diagnosis” | Narrative review (authors' view) | Cancers (Hui, Heung and Bruera) 2022 |
| Identify adults approaching the end of life with a systematic approach “People managing and delivering services should develop systems to identify adults who are likely to be approaching the end of their life (for example, using tools such as the Gold Standards Framework, the Amber Care Bundle or the Supportive and Palliative Care Indicators Tool [SPICT]).” | Not graded (NICE NG142 1.1.1) | NICE 2019 |
| The evidence on early referral in the last year of life was not sufficient for a NICE recommendation on timing “The Committee concluded that the evidence on benefit of early referral (or provision of) palliative care services for people in the last year of life was not sufficient to formulate a recommendation.” | Not graded (NICE committee's conclusion, evidence review B) | NICE (developed by the National Guideline Centre) 2019 |
| Early palliative care in metastatic non-small-cell lung cancer: quality of life, mood and survival (151 patients) “Despite the fact that fewer patients in the early palliative care group than in the standard care group received aggressive end-of-life care (33% vs. 54%, P=0.05), median survival was longer among patients receiving early palliative care (11.6 months vs. 8.9 months, P=0.02).” | Randomised trial | New England Journal of Medicine (Temel et al.) 2010 |
| Early palliative care across tumour types: primary endpoint at 3 months not met (461 patients) “Although the difference in quality of life was non-significant at the primary endpoint, this trial shows promising findings that support early palliative care for patients with advanced cancer.” | Cluster-randomised trial | The Lancet (Zimmermann et al.) 2014 |
| Early palliative care improves quality of life by a small amount; effect on survival uncertain (7 trials, 1,614 participants) “Data on survival, available from four studies enrolling a total of 800 participants, did not indicate differences in efficacy (death hazard ratio 0.85, 95% CI 0.56 to 1.28; evidence of very low certainty).” | evidence of low certainty (quality of life); evidence of very low certainty (survival) | Cochrane Database of Systematic Reviews (Haun et al.) 2017 |
| Early against delayed start: one-year survival (207 patients) “Kaplan-Meier 1-year survival rates were 63% in the early group and 48% in the delayed group (difference, 15%; P = .038).” | Randomised trial | Journal of Clinical Oncology (Bakitas et al.) 2015 |
| Stepped palliative care non-inferior to visits every 4 weeks for quality of life at 24 weeks (507 patients) “FACT-L scores at week 24 for the stepped palliative care group were noninferior to scores among those receiving early palliative care (adjusted FACT-L mean score, 100.6 vs 97.8, respectively; difference, 2.9; lower 1-sided 95% confidence limit, -0.1; P < .001 for noninferiority).” | Randomised non-inferiority trial | JAMA (Temel et al.) 2024 |
| Referral by screened need: no effect on the primary outcome (297 patients) “Early specialist palliative care showed no effect on the primary outcome of change in primary need (-4.9 points (95% confidence interval -11.3 to +1.5 points); p = 0.14).” | Randomised trial | Palliative Medicine (Groenvold et al.) 2017 |
Shares Advance care planning, Care in the last days of life: recognising dying, hydration, symptom control and anticipatory medicines, Early integrated palliative care.
Shares Palliative sedation for refractory suffering at the end of life, Advance care planning, Strong opioids for cancer pain: starting, titrating, rescue doses and preventing constipation, Care in the last days of life: recognising dying, hydration, symptom control and anticipatory medicines.
Shares Advance care planning, Hospice and end-of-life care, Early integrated palliative care.
Shares Advance care planning.
Shares Advance care planning.
Shares Advance care planning, Early integrated palliative care.
Shares Palliative sedation for refractory suffering at the end of life, Care in the last days of life: recognising dying, hydration, symptom control and anticipatory medicines, Hospice and end-of-life care, Early integrated palliative care.
Shares Palliative sedation for refractory suffering at the end of life, Care in the last days of life: recognising dying, hydration, symptom control and anticipatory medicines, Hospice and end-of-life care, Early integrated palliative care.