The short answer

  • Serious harm is rare in trials, but reminiscence therapy risks are real: painful or traumatic memories, grief, repetitive bitterness, and stress for family carers.12
  • When someone gets upset, pause, name the feeling, don't push for detail, offer to change topic or stop, and never leave them on the hard note.
  • Validate, don't correct. Never force a topic. Consent is something you check again every session, not once.
  • Trauma processing belongs with trained clinicians, not in an activity session.10
  • Talk of wanting to die, any abuse disclosure, or distress that won't settle goes to a named staff member the same day.

Anyone who has run reminiscence sessions has seen it: a question about a childhood home, and suddenly a resident is in tears about a brother who died seventy years ago. Usually a hand on the arm and a few kind words are enough. But reminiscence therapy risks deserve planning, because the people you serve are often grieving, sometimes carrying trauma, and in many cases living with dementia.

This guide is part of our series on reminiscence therapy. It covers why distress happens, which reminiscing patterns are linked to worse well-being, when reminiscence is the wrong tool, what to say in the moment, and a printable team checklist.

What the research says about reminiscence therapy risks

The good news first. The 2018 Cochrane review of reminiscence therapy for dementia, covering 22 randomized trials, reported: "We found no evidence of any harmful effects on people with dementia."1 Reminiscence is not a risky intervention in the way a drug can be.

Three caveats keep that from becoming "it can't hurt":

The honest summary: harm is uncommon and usually brief, but it is foreseeable, and a good program plans for it.

Why memories sometimes hurt

Distress in a reminiscence session usually has one of a handful of causes:

The UK Alzheimer's Society acknowledges this directly: life story and reminiscence work "may sometimes bring back difficult or sad memories, and the person may become upset. If this happens, the person should be supported to express their feelings and to address the memory if they are comfortable doing so."13 Note the direction of that advice. The aim is not to stop sadness at all costs. It is to support it.

Maladaptive reminiscence: bitterness revival and obsessive rumination

People reminisce for different reasons, and the reasons matter. Psychologist Jeffrey Webster's Reminiscence Functions Scale sorts them into functions such as identity, problem-solving, teaching others, conversation, preparing for death, boredom reduction, keeping a lost person close ("intimacy maintenance") and "bitterness revival," which means going back over old grievances.5

Later research grouped these. Identity, problem-solving and death preparation were positively associated with well-being. Bitterness revival, boredom reduction and intimacy maintenance were associated with lower life satisfaction and more depressive and anxiety symptoms.78 In a longitudinal study of 411 older adults, the positive functions predicted better physical health and lower psychological distress eight months later.9 An earlier study of adults aged 65 to 95 found that "obsessive" reminiscence, persistent guilt, bitterness or despair about the past, was less common among people who were aging successfully.6

Two practical points follow, and one caution.

Trauma: when reminiscence is not the right tool

Traumatic memories can be worked with therapeutically, but look at what that took in the research. A randomized trial of life review therapy for Holocaust survivors reported benefits on PTSD symptoms at follow-up. It was a clinician-delivered therapy that included narrative exposure, and it excluded people with probable dementia, acute psychosis and acute suicidality.10 A review of reminiscence among Holocaust survivors adds that survivors can use reminiscence well but "are increasingly more vulnerable as they reach very old age."11

That is not an activity program, and your reminiscence group should not try to be one. In practice:

For residents with known trauma histories, it is often better to steer reminiscence toward safer ground altogether: food, music, pets, skills and small pleasures. Our list of 120 reminiscence questions marks the service and hard-times questions as optional for this reason.

Grief

Talking about people who have died is one of the most valued parts of reminiscence, and you should not avoid it just because tears are possible. Tears are not failure. But grief deserves a watchful eye. Prolonged grief disorder is now a recognized diagnosis,12 and, as above, repeated "keeping the lost person close" reminiscing is associated with lower well-being in research samples.7

Two gentle rules. Let the resident decide how much to say about the person. And if someone with dementia asks where a spouse is and seems not to know they have died, don't force the news on them again in the middle of a session; talk about the person ("Tell me about him. What was he like?") and involve the care team, who should have an agreed approach for that resident.

How to respond in the moment

Practice guidance converges on a simple sequence.1317 Think of it as pause, acknowledge, offer, settle, record.

StepWhat it sounds likeWhat to avoid
Pause and stayStop asking. Sit closer. Silence is fine.Moving straight to the next question.
Acknowledge the feeling"That still means a lot to you." "You really miss her." "That sounds like it was frightening.""Don't be sad." "That was a long time ago."
Don't probe"You don't have to tell me any more than you want to.""What exactly happened?" "Then what did he do?"
Offer a choice"Would you like to keep talking about him, or shall we talk about something else for a while?"Deciding for them in either direction.
Settle before you leave"Tell me something about him that makes you smile." Or shift to a cup of tea, a song, the garden.Walking away while they are still upset.
Record and tellA short note for the care team: what came up, how they were when you left.Keeping it to yourself, or retelling it at the nurses' station as gossip.

Validation over correction

If a resident's story doesn't match the facts, the feeling is still real. Federal nursing-home guidance lists "validating the resident's feelings and words" as an activity approach, and for people experiencing things that are not real to others, "acknowledging that the resident's experience is real to her/him."15 Formal validation therapy has too little trial evidence for firm conclusions,16 so this is a professional and regulatory norm rather than proven science. It is still the right default: correcting a distressed person almost never helps.

Never force a topic

"Not today" is a complete answer. Nursing homes must support residents' "choice of activities,"15 and even outside those rules the principle holds. Don't re-ask a question someone has declined, and don't make them explain why. Life story work "may be overused" with people with dementia, one study warns;17 some days, a walk or a card game is the better offer.

Consent and capacity are ongoing

A signature at move-in is not consent to every conversation after it. People with dementia may be able to make some decisions and not others, and their capacity can vary day to day. The widely held approach in dementia care is "process consent": check again each time, watch for non-verbal signs of reluctance, and involve family or a legal representative where the person can no longer decide. (This is expert consensus; we have not tied it to a single guideline.)

In practice: ask "Would you like to talk for a bit?" every session, stop when the answer is no, and treat turning away, going quiet or getting agitated as a no.

Whose story is it?

Reminiscence produces private material. Researchers studying life story work found that "private memories were sometimes recalled" that were not meant for any written product, and summed up the challenge in their title: "You have to be mindful of whose story it is."1718

Escalation triggers: when to bring in a named staff member

Every reminiscence program needs a named person (a nurse, social worker or supervisor on each shift) and clear triggers. These triggers are practice consensus rather than trial findings, but they mirror the exclusion criteria used in the clinical trauma trial above.10

Printable checklist: safe reminiscence sessions

Before

  • [ ] I know this resident's known losses, trauma history and topics to steer around (from the care plan or family).
  • [ ] I know who today's named escalation contact is and how to reach them.
  • [ ] My questions are invitations ("Tell me about…"), not tests ("Do you remember…?").
  • [ ] I've asked, "Would you like to talk for a while?" and the answer was yes.

During

  • [ ] I follow their lead and don't correct their version of events.
  • [ ] I don't go looking for trauma, and I don't probe if it comes up.
  • [ ] If they get upset: pause, acknowledge, offer a choice, settle.
  • [ ] I treat turning away, silence or agitation as "not now."

After

  • [ ] The resident was calm or content when I left.
  • [ ] I noted anything distressing for the care team, factually and privately.
  • [ ] Anything about self-harm, abuse or acute distress went to the named contact today, in person or by phone.
  • [ ] Nothing private is shared with family or put on display without the resident's OK.

What Porchlight does, and doesn't do, here

We make a tablet app for recording life stories, so we owe you a precise answer.

What it does. A resident answers one question at a time, read aloud and shown in large type, and any question can be skipped with one tap. Nothing is scored or graded. Once a day, before the first question, there is a one-tap mood check-in; a low score sends an email alert to a coordinator suggesting a friendly check-in. It is a single question, not a validated scale or a screening tool.

What it does not do. Porchlight does not listen for distress in what residents say. It will not notice tears, flag a comment like "I wish I were dead," or recognize an abuse disclosure. It does not keep a list of topics a resident found painful, and a question skipped today may come up again another day. There is no setting to mark a story private: recordings go to the family members invited to the resident's page, so talk with the resident about who will hear them.

So the responsibility described in this article stays with your staff. Porchlight can lower the labor of structured, individual reminiscence and give staff a head start through the "Know Your Resident" briefing, but people remain responsible for listening and responding. If you read a transcript that worries you, treat it exactly as you would a disclosure made to you in person. For more on what software can and can't do, see digital reminiscence therapy.

Frequently asked questions

Can reminiscence therapy make dementia worse or upset someone?

There is no evidence that reminiscence makes dementia itself worse; the Cochrane review found no harmful effects on people with dementia. But a memory can upset someone in the moment, especially grief or trauma, and joint groups with family carers were linked to more carer anxiety. Upset is usually manageable with a calm, validating response.

What are the contraindications for reminiscence therapy?

There is no formal list, but practice guidance points to caution with recent or unresolved trauma, active suicidal thoughts, psychosis, severe or persistent depression, and grief that is not easing. In those situations a clinician should be involved, and an activity session is not the place to explore the painful material.

What should you do if someone cries during a reminiscence session?

Pause, stay with them and name the feeling kindly, for example: "That still means a lot to you." Don't push for details and don't rush to cheer them up. Offer to keep talking, change the subject or stop. Before you leave, bring the conversation to a calmer place, then tell the care team what happened.

Should you correct a person with dementia when their memory is wrong?

Generally no. Respond to the feeling behind the story rather than the facts. Federal nursing-home guidance names validating the resident's feelings and words as an activity approach. Trial evidence for formal validation therapy is limited, so this rests on professional consensus, but correcting usually adds distress without any benefit.

When should reminiscence staff escalate to a nurse or clinician?

Escalate the same day if a resident talks about wanting to die or harm themselves, discloses abuse (past or current), shows acute distress that does not settle, or seems to be re-living a traumatic event. Also flag persistent low mood or grief that is not easing. Tell a named staff member directly rather than leaving a note.

A tool that keeps people in the loop

See how Porchlight handles skipping, the daily mood check-in and staff briefings, and where your team stays in charge.

Explore a live demo Book a demo

Sources & notes

  1. Woods B, O'Philbin L, Farrell EM, Spector AE, Orrell M (2018). Reminiscence therapy for dementia. Cochrane Database Syst Rev 3:CD001120. PMID 29493789. doi:10.1002/14651858.CD001120.pub3
  2. Woods RT, Orrell M, Bruce E, Edwards RT, Hoare Z, Hounsome B, et al. (2016). REMCARE: pragmatic multi-centre randomised trial of reminiscence groups for people with dementia and their family carers: effectiveness and economic analysis. PLoS One 11(4):e0152843. PMID 27093052. doi:10.1371/journal.pone.0152843
  3. Melunsky N, Crellin N, Dudzinski E, Orrell M, Wenborn J, Poland F, et al. (2015). The experience of family carers attending a joint reminiscence group with people with dementia: a thematic analysis. Dementia (London) 14(6):842-59. PMID 24381218. doi:10.1177/1471301213516332
  4. Butler RN (1963). The life review: an interpretation of reminiscence in the aged. Psychiatry 26:65-76. PMID 14017386. doi:10.1080/00332747.1963.11023339
  5. Webster JD (1993). Construction and validation of the Reminiscence Functions Scale. J Gerontol 48(5):P256-62. PMID 8366271. doi:10.1093/geronj/48.5.p256
  6. Wong PT, Watt LM (1991). What types of reminiscence are associated with successful aging? Psychol Aging 6(2):272-9. PMID 1863396. doi:10.1037//0882-7974.6.2.272
  7. Cappeliez P, O'Rourke N (2006). Empirical validation of a model of reminiscence and health in later life. J Gerontol B Psychol Sci Soc Sci 61(4):P237-44. PMID 16855036. doi:10.1093/geronb/61.4.p237
  8. O'Rourke N, Cappeliez P, Claxton A (2011). Functions of reminiscence and the psychological well-being of young-old and older adults over time. Aging Ment Health 15(2):272-81. PMID 21140308. doi:10.1080/13607861003713281
  9. King DB, Cappeliez P, Canham SL, O'Rourke N (2019). Functions of reminiscence in later life: predicting change in the physical and mental health of older adults over time. Aging Ment Health 23(2):246-254. PMID 29110517. doi:10.1080/13607863.2017.1396581
  10. Forstmeier S, Zimmermann S, van der Hal E, Auerbach M, Kleinke K, Maercker A, Brom D (2023). Effect of Life Review Therapy for Holocaust Survivors: a randomized controlled trial. J Trauma Stress 36(3):628-641. PMID 37155933. doi:10.1002/jts.22933
  11. Zimmermann S, Forstmeier S (2020). From fragments to identity: reminiscence, life review and well-being of holocaust survivors. An integrative review. Aging Ment Health 24(4):525-549. PMID 30522330. doi:10.1080/13607863.2018.1525608
  12. Killikelly C, Smith KV, Zhou N, Prigerson HG, O'Connor MF, Kokou-Kpolou CK, et al. (2025). Prolonged grief disorder. Lancet 405(10489):1621-1632. PMID 40254022. doi:10.1016/S0140-6736(25)00354-X
  13. Alzheimer's Society (UK). Therapy and approaches for memory loss support (life story work, reminiscence work). https://www.alzheimers.org.uk/about-dementia/stages-and-symptoms/dementia-symptoms/approaches-therapy-memory-loss
  14. Alzheimer's Association (US). Reminiscence and Reminiscence Therapy. https://www.alz.org/help-support/caregiving/daily-care/reminiscence-and-reminiscence-therapy
  15. 42 CFR 483.24(c) Activities; CMS State Operations Manual, Appendix PP (Rev. 225, 08-08-24): F679 (Activities), F699 (trauma-informed care), F742 (PTSD). These bind Medicare/Medicaid nursing homes; assisted living is regulated by the states.
  16. Neal M, Barton Wright P (2003). Validation therapy for dementia. Cochrane Database Syst Rev (3):CD001394. PMID 12917907. doi:10.1002/14651858.CD001394
  17. McKeown J, Ryan T, Ingleton C, Clarke A (2015). 'You have to be mindful of whose story it is': the challenges of undertaking life story work with people with dementia and their family carers. Dementia (London) 14(2):238-56. PMID 24339102. doi:10.1177/1471301213495864
  18. Gridley K, Birks Y, Parker G (2020). Exploring good practice in life story work with people with dementia: the findings of a qualitative study looking at the multiple views of stakeholders. Dementia (London) 19(2):182-194. PMID 29688046. doi:10.1177/1471301218768921
  19. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO (1982). Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 17(1):37-49. PMID 7183759. doi:10.1016/0022-3956(82)90033-4
  20. Alexopoulos GS, Abrams RC, Young RC, Shamoian CA (1988). Cornell Scale for Depression in Dementia. Biol Psychiatry 23(3):271-84. PMID 3337862. doi:10.1016/0006-3223(88)90038-8

Porchlight is a conversation and life-story tool, not a medical device or a clinical treatment. This article is educational and is not medical advice. No fabricated customers, testimonials, or outcome metrics appear here.