The continuing attention to the needs and comfort of a dying patient is as important as the care given to any other patient and part of that care includes reducing the distress of relatives. Many issues may be raised by relatives that pose ethical dilemmas and these may make you question the therapy that has been or should be given. Amongst the most frequent scenarios is the role of intravenous hydration, evaluating the balance between painful cannulation and restriction of mobility versus an uncomfortable dry mouth and thirst. To address these questions you should consider whether death from the cancer is now inevitable, whether interventions would relieve symptoms and whether treatment would cause harm. Careful explanation to the relatives is essential; in this circumstance, for example, they need to be reassured that the patient is not dying because of dehydration but rather because of progression of the cancer.

Symptom control in the dying patient often requires a different route of administration as swallowing may be difficult and agitation and restlessness are often prominent features as death approaches. A number of factors may contribute to terminal agitation including physical causes such as pain, sore mouth and full bladder or rectum, along with emotional factors including fear of dying and the distress of relatives. The physical causes should be addressed appropriately and unnecessary medications should be stopped. Often the best method of delivering analgesia, antiemetics and sedation, if appropriate, is via a subcutaneous syringe driver. Similarly, oral secretions accumulating in a patient who is too weak to cough may be distressing to the patient and family alike. Drug treatment for terminal secretions includes hyoscine hydrobromide and glycopyrronium bromide, which is less sedating. It is important to recall that not all patients wish to be sedated and this should be discussed with them and their families.

For many patients with cancer the last hours and days are heralded by a deterioration to semiconsciousness. At this time patients are usually unable to take oral medication and prescriptions need to be reconsidered. Many medicines may be stopped altogether and alternative routes of administration, including subcutaneous, rectal and transdermal routes, may be employed for other necessary medications including analgesia. Although patients may no longer be receiving medicines by mouth, oral hygiene remains an important part of overall care. It is particularly important to avoid unnecessary unpleasant interventions at this time and to adopt a practical problem-oriented approach to symptom control. A practical guide to the care of the dying patient in a hospital was developed at the Royal Liverpool Hospital in conjunction with Marie Curie Cancer Care to transfer best practice learnt from hospice care. The Liverpool Care of the Dying Pathway (LCP) helps members of the multidisciplinary team in making a decision about which medical interventions should be stopped and which one should be continued (including anticipatory prescribing) and what comfort measures should be started. It also promotes psychological support of the patients, family and carers as well as addressing spiritual needs and bereavement. There has been debate recently about the value of the Liverpool Pathway, with some critics suggesting that it is a one-way road that sanitizes and precipitates the process of dying without allowing thought or revision of decisions. In 2012, half of all NHS trust were to receive financial rewards under the Commissioning for Quality and Innovation (CQUIN) scheme for hitting targets associated with this pathway. This perhaps inevitably led to a furore about the pathway with media critics accusing clinicians of hastening death with a “sedation and dehydration” tactic. In July 2013 the Department of Health decided to phase out the LCP in favour of more individualized end-of- life care.

When death is inevitable, as it is for all of us, and is approaching rapidly it is the policy in many UK hospitals to discuss resuscitation policies with patients and their relatives. Under these circumstances resuscitation is rarely appropriate and, if deemed futile, the lead clinician may make a “do not attempt resuscitation” (DNAR) decision. These DNAR decisions should be discussed with patients who wish to engage in advanced care planning. However, prolonged discussions about DNAR policies with patients who do not wish to contemplate their future are, in the view of these authors, distressing and irrelevant. The reason for our autocratic view is that cardiac resuscitation cannot return the patient who has died from cancer from his journey across the River Styx, and it causes distress in the relatives and the arrest team.

Just as different cultures, regardless of the scientific evidence, have developed distinct explanations for the origins of life ranging from Big Bangs and evolution to creationist genesis, similar cultural variations affect attitudes to death. For example, Christians, Jews (Box 41.1) and Sufis believe in resurrection whilst Hindus, Buddhists (Box 41.2) and Sikhs believe in reincarnation. These cultural discrepancies must be recognized and respected, particularly where patients’ and carers’ views differ.