Creative Work at the End of Life: Legacy Pieces, Memory Books and Who They Belong To

Legacy work is creative work made near the end of somebody’s life that is meant to remain afterwards: a memory book, a recording, letters, photographs or a made object. The questions that decide whether it goes well are not artistic ones, and almost all of them arrive after the person has died, which is exactly when nobody can answer them.

A family asked me to help make a recording once. Their mother, who was in one of the homes I work in, still sang whole verses of songs she had known since childhood, and her daughter wanted it kept. We did it, and the singing was the easy part. What I had not thought about beforehand was everything else on the recording. Two other residents are audible on it. A member of staff comes in halfway through and says something about lunch. Her daughter loved it and wanted to send it to the rest of the family, and I found myself working out, after the fact, that I had made an object containing three people who had not agreed to be in it, in a home with its own duties about all of them. It was fixable and I fixed it, and the point is that I fixed it afterwards, which is the wrong order and is how most of these problems arise.

Where a claim about effect appears below it carries a tier (Tier 1 supported, Tier 2 promising but limited, Tier 3 practice and anecdote), explained on reading arts in health research.

What the work covers

A wide range of things, with the same practical questions running through all of them.

Memory books and life story work. Photographs, documents, written recollections, sometimes assembled over months. Common in care homes and hospices, and often begun long before anybody is dying.

Recordings. A song, a message, a voice. The most emotionally powerful and the most complicated, because a recording captures whatever else was in the room.

Letters and writing. To be read later, sometimes at a specified time. These raise their own question about who decides when later is.

Made objects. A piece of art, a quilt, something with handprints, a piece of jewellery. Often made with family and often the least contested, because everybody involved is present at the making.

Photography. Portraits, hands, familiar objects, the room. Powerful and easy to make in quantity, which is part of the problem.

Who does it varies: hospice staff, an arts practitioner or artist in residence, a chaplain, a volunteer, an activities coordinator, or the family alone. In most cases it is not a registered therapy. Music therapy, art therapy, art psychotherapy and dramatherapy are clinical interventions delivered by practitioners holding titles protected in UK law, and in July 2026 the Health and Care Professions Council register listed 6,103 arts therapists across all four titles¹, a small workforce spread thinly across every setting. Where a music therapist is involved in palliative care, the work is assessed, recorded and supervised, and it may include legacy pieces as part of something wider². See music therapy in palliative care.

The questions to settle at the start

Six of them, and they are far easier to ask before than after.

Who is this for? A recording made for one grandchild and a recording made for a funeral are different objects, and only the maker can decide which it is. Ask, and write the answer down.

Who else is in it? Other residents, other patients, staff, visitors. Anybody audible or visible in a recording or photograph is in it, and they have not agreed to be. This is the one people miss most often and it is the easiest to prevent by choosing where and when.

Who holds the original, and who gets a copy? Name a person, not a family. Families disagree, and an organisation holding the only copy can find itself in the middle of a dispute it has no standing in.

What happens if it is unfinished? People die before projects end, often. Agree in advance who receives an unfinished piece and whether anybody may complete it, because a family finishing somebody’s work can be a comfort or a violation depending entirely on what that person would have wanted.

May any of it be used publicly? By the hospice, the care home, the arts organisation, in a newsletter, an annual report or a funding bid. The answer is very often no, and it should be asked rather than assumed, and asked of the person rather than of the family.

What happens to it in a year? Organisations hold files, and files persist. Somebody should know when a copy held by a service is deleted or handed over.

Consent, and why a relative’s agreement is not enough

The core constraint, and the one most often got wrong in good faith.

Capacity in England and Wales is decision specific and time specific: a person is assumed to have capacity, capacity is assessed for the particular decision in front of them, and it can fluctuate³. Somebody may be perfectly able to decide whether to sing this afternoon and unable to decide whether a recording of it should be kept, copied and shared indefinitely. Those are different decisions and the second is much larger.

A relative’s agreement does not reliably extend to publication. Family members are frequently the right people to consult about what somebody would have wanted, and consulting them is proper. What they cannot generally do is give consent on somebody’s behalf for their image, voice or work to be published, because publication is not the kind of decision a best interests process fits well: it is open ended, it cannot be undone, and it benefits the publisher.

The practical rule follows from that. Where a person can consent, ask them, specifically, about each use. Where they cannot, the default for anything beyond the immediate family is not to share. Recordings and images are also personal data with obligations attached under UK data protection law⁴, which is a further reason for an organisation to be careful rather than an alternative one. The full account is on consent and safeguarding in arts in health, and it binds this site’s own imagery as much as anybody’s.

What the evidence supports

Very little directly, and the nearest controlled evidence is weak.

No effect size for legacy work is locked in this site’s source document, so none is quoted here. That is Tier 3, of the nobody has looked properly kind rather than the null controlled evidence kind. The reasons are obvious and not suspicious: the population is small, unwell and changing quickly, randomisation is close to unthinkable for most of it, and the outcomes people care about, such as how a family feels three years later, are extremely hard to measure and further still from where the funding is.

The nearest controlled evidence concerns music interventions in cancer care. The 2021 Cochrane review pooled 81 trials and 5,576 participants and reported anxiety 7.73 STAI-S units lower and pain at SMD -0.67, both at very low certainty⁵. That is Tier 2, and it is also this site’s best teaching example: eighty one trials and five and a half thousand participants can still add up to very low certainty, because certainty is about how much the pooled result can be trusted rather than how much of it there is. See art therapy during cancer treatment.

Tier 3 is not a verdict that legacy work is without value. It means nobody has tested it properly, and for this particular activity the question of whether a trial is the right instrument is a real one. What a family has afterwards is not obviously the kind of thing an outcome scale was built for.

What can be said honestly to a family

Language that is neither a promise nor a discouragement, which is a narrow path.

Say: many families value having something afterwards; this is something to do together while it is still possible; there is no right way to do it and it does not have to be good; and it can stop at any point.

Do not say: that it will help with grief, that it will bring closure, that it improves quality of life at the end of life, or that research shows anything specific about it. None of that can be supported, and a family who were promised comfort and did not find it will carry that as a further loss.

The other thing worth saying plainly is that it is allowed not to. Legacy projects are offered with such warmth that declining one can feel like a failure of love, and some people do not want to make anything, do not want to be recorded, and would rather the last months were not organised around producing an object. That is a complete answer.

Practical points that prevent most problems

Nine, drawn from doing this badly once and better since.

1. Choose the room and the time so that other people are not captured.
2. Ask before you record, every time, even if you asked last week.
3. Test the equipment before the person is in front of it. Asking somebody who is exhausted to sing it again is a real cost.
4. Keep sessions short. Fatigue arrives quickly and dominates everything.
5. Make more than one copy immediately, and do not let the only copy live on one person’s phone.
6. Name who gets it, in writing, and give them a copy while the person is alive if that is what they want.
7. Do not edit out the ordinary parts without asking. The pause, the cough and the interruption are frequently what families treasure.
8. Plan for unfinished.
9. Write down what may and may not be used publicly, and default to nothing.

For anybody setting up this kind of work as a programme rather than doing it once, starting an arts programme in a care setting covers insurance, safeguarding and who owns it internally. The sector’s own case studies are gathered in one place⁶, usefully read as descriptions of practice rather than as evidence of effect.

Where this sits

Alongside care, never instead of it.

Nothing on this page is a reason to change, delay or decline any part of anybody’s treatment or palliative care, and no creative project substitutes for symptom control, for a conversation with the palliative care team, or for support a family needs. If somebody is in pain, distressed or frightened, that is a clinical matter first and a creative one afterwards.

The related pages are arts in hospitals for the wider setting, music therapy in palliative care for the registered version of this work, arts in care homes for the setting most of it happens in, and therapeutic writing and photography for two forms that come up constantly here and are neither of them registered therapies.