There is a moment familiar to almost anyone caring for a person with dementia: a door to a garden, and a decision. Is it worth the coats, the shoes, the walking frame, the twenty minutes of negotiation, the risk of a fall on wet paving? Care homes advertise sensory gardens. Charities run walking groups and supported holidays. Families are told, gently and often, that being outside is good. What has been missing is a straight answer to a simple question: what does the research actually cover, and where does it simply stop?
A scoping review accepted on 3 August 2026 in Frontiers in Dementia sets out to draw that map. Manpreet Kaur Gill, Shareefa Thekkan and Danine Irwin, all at Dementia Adventure in Chelmsford, England, a charity that itself provides nature-based breaks, searched PubMed, Scopus, Web of Science, Emerald Insight and the Cochrane Library, screened the results with two independent reviewers, and included 108 studies. The method is stated plainly: the Arksey and O'Malley framework, reported to PRISMA-ScR standards, with narrative synthesis and thematic categorisation. That wording matters more than it looks. A scoping review counts and sorts the literature. It does not pool results into a single effect size, and it does not issue a verdict on how reliable each study was. This is a map, not a measurement. At the time of writing it is the provisional accepted version; the final formatted paper is still to come.
The shape of the map is the story. The 108 studies fall into six themes, and they are wildly uneven. Therapeutic gardens and environmental design account for 33 studies. Horticultural and gardening interventions account for 30. Green care farms and nature-based services account for 22. Nature exposure, access and participation account for 16. Then the map thins out sharply: five studies on technology-based nature interventions, and two, just two, on the development and implementation of nature-based services. Decades of interest in dementia-friendly gardens have produced a substantial body of work on what a good garden looks like, and almost nothing on how a service gets built, funded, staffed and kept running.
On benefit, the authors are careful, and so are we. Their reading is that nature-based interventions may offer something beyond wellbeing alone, supporting identity, social participation and meaningful engagement for people living with dementia and for carers, with accessibility, environmental design and the chance to keep taking part in everyday life running through the findings. Read that as a description of what a scattered, largely small-scale literature reports, not as proof that a garden changes anything by itself. The review does not separate results by type of dementia, so nothing here tells a family whether Alzheimer's, vascular, Lewy body or frontotemporal dementia responds differently. Nor is this the only map of the territory: a conceptual systematic review published in Dementia in July 2024 by Harmony Jiang and colleagues found 52 articles reporting 51 studies and concluded that researchers define "nature-based intervention" so differently that a taxonomy was needed, while a scoping review in Archives of Gerontology and Geriatrics in January 2026 charted 32 gardening studies and reported that evidence for cognitive benefit was mixed, even where mood and psychosocial measures looked better.
The gaps the Frontiers authors name are the ones a family would name too. Limited research on implementation. Limited research on blue spaces, meaning rivers, coasts and lakes. Limited research on technology-based approaches, the virtual woodland on a screen for someone who cannot easily get outdoors. And limited research on underrepresented populations, which in practice means people whose culture, language, income, housing or mobility differs from the participants who happen to have been studied. An earlier scoping review from the ENLIVEN project, published in Health and Social Care in the Community in July 2023, made the same point from the other end: seniors with cognitive impairment face significant barriers to getting outdoors at all, whatever the research says about the benefits once they are there.
So what does a carer do with a map that has blank quarters? Not much that is dramatic, and that is the honest answer. Going outside is not a treatment, and nobody in this literature is claiming it slows disease. What the review does is give a name to something families already sense: that time in a garden can be about who a person still is, and who they are still with, rather than about scores on a test. Whether a particular garden, farm, walk or holiday is safe and suitable for one particular person, with their particular balance, eyesight, stamina and medication, is a conversation for the clinical team who know them, not a conclusion that can be drawn from 108 studies on a page.