Provision of Outdoor Nature-Based Activity for Older People with Cognitive Impairment: A Scoping Review from the ENLIVEN Project

,


Introduction
Tere are 771 million people aged 65 years or over worldwide, and this is projected to rise to 1.6 billion by 2050 [1]. Globally, more than 55 million people live with dementia, a number expected to rise to over 150 million by 2050 [2].
Between 5 and 10% of people over 65 in higher-income countries live with dementia, with prevalence doubling every 5 years after the age of 65 [3]. Cognitive impairment can also arise due to other age-associated neurological conditions such as stroke or Parkinson's disease [4]. Living with cognitive impairment has implications for well-being, with many experiencing poor quality of life [5]. Tis is often not a direct result of the health condition itself but rather results from secondary consequences that could potentially be alleviated, such as loneliness, social isolation, and lack of meaningful occupation [6]. Community-based initiatives that enable people to engage in activities and connect socially in ways that accord with their interests and preferences are important for maintaining quality of life; this includes support to get out and about, take exercise, and enjoy the natural environment [7].
Te health and well-being benefts of engaging with nature and the outdoor environment for people with a range of health conditions are increasingly acknowledged [8][9][10][11], and this extends to older people with dementia and cognitive impairment [12,13] and their family caregivers [14,15]. Suggested benefts of outdoor activity for people with dementia include providing pleasure and enjoyment, maintaining independence and meaningful occupation, promoting social inclusion, stimulating memory and the senses, and enhancing identity and self-esteem [13,16]. In support of this, greater perceived availability of local green and blue spaces was associated with better quality of life among a large cohort of people with mild-to-moderate dementia [17]. It is important to ensure that people who wish to do so can continue to connect with the natural environment, take exercise, and engage in individual or group activities that they enjoy. Tis could be in familiar everyday spaces or could involve visiting places further afeld [18]. However, older people with dementia and other forms of cognitive impairment experience signifcant barriers to accessing the natural environment, with some excluded entirely [19][20][21][22][23], especially those from minority ethnic and disadvantaged groups and those living in long-term residential care [18]. Older people with cognitive impairment may be concerned about staying safe, falling, or getting lost, and caregivers may judge it too risky to take the person they care for outdoors, but many barriers are social and structural, relating, for example, to facilities, physical accessibility, signage, transport, and costs [19,20,22,24]. Addressing barriers to accessing outdoor nature-based activity requires us to consider both how to enable individuals and families to engage and how to develop supportive social and organisational attitudes and practices among those who manage and control access to outdoor green and blue spaces, provide the nature-based activities that take place within those outdoor spaces, or provide the care settings in which people spend time [25,26].
Te ENLIVEN project-Extending active life for older people with cognitive impairment and their families through innovation in the visitor economy of the natural environment-was set up as part of the UK Healthy Ageing Challenge Social, Behavioural, and Design Research programme https://www.ukri.org/what-we-ofer/our-mainfunds/industrial-strategy-challenge-fund/ageing-society/) to gather evidence and co-produce resources and guidance for providers ofering or wishing to ofer outdoor naturebased activities for older people with dementia and other forms of cognitive impairment. For the purposes of the project, outdoor nature-based activity was defned as "an activity undertaken outdoors in green or blue spaces such as parks, gardens, farms, woodlands, rivers, and beaches in which being in, attending to, or engaging with aspects of the natural environment, including animals where relevant, is a key component." A frst step was to assess the available evidence about the kinds of initiatives that have been implemented and tested and the routes through which these might confer benefts. Tere have been previous reviews focused on garden use and horticultural activities for people with dementia [16,27] and on mechanisms of beneft of nature-based activity [13,28], but we found no wide-ranging review covering the provision of any type of nature-based activity for older people with dementia or for older people with other forms of cognitive impairment. To address this gap, given the broad focus, we identifed a scoping review [29] as the most suitable approach. To our knowledge, this is the frst scoping review to examine the available research evidence about the provision of all types of outdoor naturebased activity for older people with dementia or other forms of cognitive impairment. Te aims of the review were as follows: (1) To identify and characterise the types of outdoor nature-based activity provision for older people with dementia and other forms of cognitive impairment that have been the subject of research and for which research evidence is available (2) To identify and describe the range of outcomes examined in the research studies on this topic

Methods
Tis scoping review was conducted in accordance with available guidance [30,31] and with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA-SR) Statement [32]. Te PRISMA checklist is available in Supplementary Material. Te review protocol is available online [33].

Eligibility Criteria.
Eligibility criteria were developed using the SPIDER approach [34] to specify Sample, Phenomenon of Interest, Design of the Study, Evaluation, and Research type.

Sample.
Te sample for this review was older people aged 65 and above with dementia or with cognitive impairment arising from another health condition. Where studies included a wider age range, either two-thirds of participants had to be aged 65 or above, the mean age had to be 70 or above, or data had to be presented separately for those over 65 Table 1). Tree sets of terms were generated. Te frst set covered terms related to cognitive impairment (e.g., dementia and Alzheimer's). Tis was combined with a set of terms related to nature-based activity (e.g., nature-based activities, ecotherapy, and green care) and a set of terms related to volunteering (e.g., volunteer and civic participation). Te potential relevance of the latter set of terms was identifed through interviews with representatives of provider organisations conducted as part of the wider EN-LIVEN project. Because eligible studies might include family members of the person with cognitive impairment and these could be of any age, we did not flter the search results based on age.

Screening and Eligibility.
Search results were uploaded into EndNote (Clarivate Analytics, USA), and duplicates were removed. Te screening of titles and abstracts was carried out by two researchers working independently, who then sought to resolve any diferences of opinion through discussion. Where agreement could not be reached, the record was added to the list for full-text screening. Full-text screening was conducted by two researchers working independently, with any diferences of opinion referred to a third researcher. In cases where we could not determine from the published paper whether inclusion criteria were met, further information was sought from the lead author. Authors were contacted on November 21 st , 2022, with responses received by December 19 th , 2022.

Data Extraction.
Data extraction was carried out by two researchers working independently using a bespoke extraction form prepared for this purpose in Excel and tested in pilot searches. For quality assurance purposes, 10% of the extractions were checked for accuracy by a third researcher. We extracted data on study characteristics (e.g., author, year, and country of study), participants (e.g., age and level of cognitive impairment), outdoor nature-based activities, the settings in which these were undertaken, the activity providers, and the outcomes that were examined. In keeping with the remit of a scoping review, we did not conduct an appraisal of study quality.

Synthesis of Results.
To answer the frst review question, we synthesised information about the outdoor nature-based activities described and grouped studies according to type of activity to provide a narrative account. To answer the second review question, we separately listed quantitative outcomes and the categories or themes identifed in qualitative studies and then summarised and grouped these by domain to provide a comprehensive overview.

Selection of Sources of Evidence.
After removing duplicates, we identifed 18,862 unique records. Following title and abstract screening, we scrutinised the full text of 122 articles and identifed 28 articles meeting inclusion criteria; see the fowchart in Figure 1 for further details. All involved people with dementia; no studies involving other groups such as stroke survivors met inclusion criteria. Tree articles that did not provide information about participants' ages were considered to meet inclusion criteria based on other information or contextual knowledge, as two had similar samples to those in other linked studies meeting inclusion criteria and one involved people living in residential care homes in the UK, which typically have an older population.

Overview of the Included Studies.
Te 28 articles all focused on the experiences of older people with dementia; two considered the perspectives of family carers only and one considered the perspectives of service providers only. Tree research groups contributed sets of linked studies, reported in four, fve, and eight discrete articles, respectively. Te remaining 11 articles reported discrete studies. Te studies were conducted in Norway (n = 9), the USA (n = 6), the Netherlands (n = 5), the UK (n = 5), Canada (n = 1), Brazil (n = 1), and Japan (n = 1). Tirteen studies adopted a quantitative design, nine were qualitative, and six were mixed methods evaluations. Quantitative studies were primarily observational (n = 11), using either cross-sectional (n = 8) or longitudinal (n = 3) data, but also included one quasi-experimental uncontrolled pre/postcomparison and one randomised crossover design; direct behavioural observation of participants during activities was a feature of six studies. Qualitative studies presented data from semistructured interviews and focus groups, analysed using thematic (n = 3), framework (n = 2), content (n = 3), or descriptive (n = 1) analytic methods. Mixed-methods evaluations combined direct observation with survey and interview data or feld notes (n = 3), reported survey and interview data (n = 2), or used participatory appraisal (n = 1). A small number of studies analysed or considered fndings in relation to specifc theoretical models, for example, the Lived Environment Life Quality Model [37], Gofman's theory of social interaction as performance [38], the social health framework [39], and the theory of salutogenesis [40]. Sample sizes for participants with dementia, provided in all but one article, ranged from 4 to 136. In 18 studies, the participants with dementia were day care attenders; four studies were conducted with residents in long-term care facilities and fve with people living in their own homes or assisted living, while one included people from a range of living situations. In the rare cases where ethnicity was mentioned, participants were primarily white, with the exception of one study [41] that recruited mainly African-Americans. Gender balance was reported in most studies and varied according to context and type of activity. In most cases where information about the severity of dementia or cognitive impairment was provided, participants were described as having, or had screening scores indicative of, mild-to-moderate dementia, but a few studies included people with moderate-to-severe dementia.  4 Health & Social Care in the Community With regard to the type of outdoor nature-based activity provided, the 28 studies fell into three distinct groups: green day care (15 articles, of which four were linked studies from one research group in the Netherlands and eight were linked studies from one research group in Norway); equine-assisted interventions (seven articles, of which fve were linked studies from one research group in the USA); and community nature-based activities (six articles).
Outcome domains considered in examining the efects of engaging in nature-based activities are summarised in Table  2, including details of the standardised measures used in quantitative studies. Outcomes assessed quantitatively covered participation in activities, clinical characteristics, the physical, functional, psychological, and social domains, and overall quality of life. Qualitative accounts considered the efects of green day care, equine-assisted interventions, or community nature-based activities on connection with nature and on the psychological, social, and physical status of the person with dementia. Connection with nature was considered in terms of the potential for both bringing out past memories and encouraging a future perspective through observing the passage of the seasons and cycle of life, particularly in studies of walking [39] and gardening [42]. Some nature-based activities involved interacting with animals. In these cases, qualitative accounts emphasised the importance of the bonds participants developed with the animals [43]. In the psychological domain, accounts considered the efects of either providing structure and routine, leading to anticipation of pleasurable activity, or creating novel experiences that allowed people to try something new [44]. Tey explored the impact of taking up valued roles and responsibilities in relation to a sense of purpose and accomplishment [42,45] and personal empowerment through being able to make choices, decide about taking risks, relearn skills, or develop new skills [46]. In the social domain, outcomes included the efects of joining a group, sharing experiences, and feeling part of a wider community on social isolation and feelings of loneliness [39,47,48], as well as the quality of relationships between people with dementia and their family carers or paid care staf [49]. In the physical domain, qualitative accounts focused on the efects of activity on physical health, diet, mobility, and balance [46,50].
Below, we examine the three groups of included studies in more detail, covering the nature of the activities and the outcome domains explored in the research.

Green Day
Care. Te 15 articles focusing on the provision and outcomes of green day care are summarised in Table 3  Health & Social Care in the Community green care farms in Norway and the Netherlands, one reported on an initiative to introduce the green care farm concept into Japan using rice farming which is widely practised in East Asia, and one explored nature-based adult day care provision in urban areas. Tere were nine quantitative, fve qualitative, and one mixed-methods research designs, and sample sizes ranged from 10 to 136 people with dementia. Green care farms are specialist facilities, typically linked to commercial working farms, where small groups of people with dementia, attending one or more days per week, are exposed to a care experience that is embedded in nature and the outdoors in the context of a home-like domestic environment. Some green care farms provide residential care, but for present purposes only day care provision was relevant. Day care programmes at green care farms include outdoor activities such as feeding animals, cleaning pens, yard work, and gardening, as well as domestic activities such as meal preparation.
Two linked studies [51,52] examined the characteristics of people attending day care at green care farms and the factors associated with quality of life for these individuals. Tese studies also explored predictors of withdrawal from green day care at one-year follow-up; variables considered were clinical characteristics (dementia severity, awareness, depression, anxiety, neuropsychiatric symptoms, and so on), physical health, functional ability, social support, and quality of life. Two qualitative studies explored the perceptions of people with dementia [40] and family carers [53] regarding their experiences of day care at green care farms and its impact, focusing on social relationships, occupation, the extent to which services are individually tailored, and the     Seven quantitative studies and two qualitative studies made comparisons between people with dementia attending day care at green care farms and regular socially oriented day care facilities ofering leisure and recreational activities [38,45,50,[54][55][56][57][58][59] and one study [57] additionally examined predictors of quality of life for people attending green care farms. Variables included in quantitative analyses were as follows: quality of life; activity engagement, focusing particularly on physical activity and efort and time spent outdoors; emotional well-being, social connections; physical health including co-morbidity, medication use, and dietary intake; functional ability in activities of daily living; and clinical characteristics including depression, neuropsychiatric symptoms, and medication use. Qualitative comparisons focused on the initiation of day care, the choice of day care setting, and the infuence of diferent kinds of settings and activities on social participation. Tere was a preponderance of men among the green care farm samples, some but not all with prior farming experience, and a higher proportion of women attending regular day care.
One study [49], introducing the concept of green care farms to Japan for the frst time, explored the efects of adding a rice farming programme consisting of weekly onehour sessions to regular day care provision relative to a reference group attending regular day care only. Quantitative indicators were well-being and cognition, and qualitative indicators were enjoyment and connection.
While green care farms are predominantly situated in rural or semirural areas, nature-based day care services can also be found in urban areas, in settings such as city farms, community gardens, and nursing home gardens. Naturebased day care may be provided by health and social care professionals and organisations or by social entrepreneurs and community groups. One study [43] examined the types and characteristics of urban nature-based day care services, motivations for choosing them, and their value to people with dementia and carers.

Equine-Assisted Interventions.
Te seven articles describing equine-assisted interventions for people with dementia are summarised in Table 4. Tese interventions were delivered at accredited therapeutic riding centres by specialist staf, with a high ratio of support staf and volunteer helpers to participants. Te interventions involved interacting with the horses (e.g., tacking up, grooming, leading, and feeding), and in most but not all cases, riding the horses. Tere were fve quantitative and two qualitative studies; sample sizes ranged from 4 to 26 people with dementia. Participants were drawn from residential care (three studies), day care (one study), or community (three studies) sources.
One study [60] examined the efects of an equineassisted intervention on balance, agility, muscle strength, and cognition. Direct behavioural observation was used to examine participants' reactions in four studies, focusing on quality of life indicators such as time use (e.g., gaze, conversation, and participation) and emotional well-being (e.g., pleasure and agitation). One study [37] described quality of life indicators displayed by participants during the activity. Two studies compared quality of life indicators observed during the activity to those observed during other activities in either long-term residential care [44] or regular day care [41], with the latter study also examining stress levels through analysis of salivary cortisol and care staf reports of problematic behaviour in the day care setting in a randomised crossover design. One study [42] compared quality of life indicators shown by a group of people with dementia who chose to participate in a horse-riding activity with those shown by a comparison group who chose to engage in a gardening activity. Two qualitative studies explored the perspectives of family members [61] and service providers [62], describing outcome domains of well-being, functional ability, and social relationships.

Community Nature-Based Activities.
Te six studies focusing on outdoor nature-based activities in community settings are summarised in Table 5. Two studies used qualitative methods and four adopted a mixed methods evaluation approach; sample sizes, available for fve studies, ranged from 6 to 39 people with dementia. Participants were drawn from residential care (one study), day care (two studies) or community (two studies) sources, or a mixture of these (one study).
Two studies [48,63] examined the experience and impact of participating in therapeutic gardening sessions provided in addition to regular day care. Both studies explored staf and care partner perceptions; one [63] used Dementia Care Mapping to identify instances of well-being and ill-being, while the other [48] conducted group interviews with participants, exploring the impact on identity, agency, and sense of community. One study [39] examined the efect of attending dementia-friendly walking groups on the social health of people with dementia living in the community and their family carers, focusing on social contact, being in a safe environment, and accessing the outdoors.
Te remaining three studies are all linked to Dementia Adventure, a British charity and social enterprise that works to enable people with dementia to get outdoors and connect with nature. Te frst of these [46] describes a pilot project providing days out in woodlands for care home residents; data about the physical, emotional, and social impact and perceived importance of various aspects of a woodland visit were collected. Te second [47], reports fndings from the evaluation of Dementia Adventure small group or bespoke holidays provided during 2016, considering perceived changes in emotional, social, and physical well-being, confdence, and relationships. Te third [64] evaluates a project in which Dementia Adventure worked with partner organisations to support the provision of opportunities for people with dementia to access the natural environment, examining the impact of participation on mental well-being and physical activity, enjoyment of the sessions, and perceived benefts.

Discussion
In this scoping review, we set out to characterise the research evidence on the provision of any type of outdoor naturebased activity for older people with dementia or other forms of cognitive impairment, considering both the range of activities studied and the outcome domains examined. Tis is, to the best of our knowledge, the frst scoping review to focus on formal activity provision, to consider older people with any form of cognitive impairment, and to cover a broad range of outdoor nature-based activities rather than focusing on one specifc type of activity. Systematic literature searches yielded 28 articles meeting our inclusion criteria, all relating to older people with dementia. Just over half focused on nature-based day care. Te remainder considered equineassisted interventions and community nature-based activities including gardening, walking groups, woodland visits, and adventure holidays, with participants drawn from day or residential care settings, community sources, or a mixture of these. Outcomes assessed quantitatively included activity participation, the impact of participation on clinical symptoms, functional ability, physical, psychological, and social health, and overall quality of life. Qualitative accounts focused on outcomes in terms of connection with nature and on the psychological, social, and physical status of the person with dementia, as well as impacts on family carers. Te included studies mainly involved participants already attending or residing in formal care settings. Around half of the studies reported on day care provision, and participants in over three-quarters of the studies were either attending day care or living in residential care. Most of the available evidence relates to day care provision as a whole or to the impact of adding specifc activities to those usually ofered as part of socially oriented day care provision. Studies of day care provision either explored the specifc impact of nature-based day care or compared the relative benefts of nature-based and regular day care. Tis is important because the availability of a range of evidence-based day care options allows for an element of choice and personalisation, as well as fexibility when care needs change. Similarly, ofering a wider range of options within regular day care services or for people living in residential care by providing outdoor nature-based activities could promote choice and enrich everyday experience, either reconnecting people with familiar or previously enjoyed activities, such as horse riding or gardening [42,44] or providing the stimulus of trying something new or diferent. While it might be assumed, for example, that day care on a green care farm would be suitable mainly for those with prior experience of farm work, fndings suggested that this appealed also to those without such prior experience [40].
Relatively few studies involved participants drawn from community sources or reported on community-based initiatives that were not connected to existing formal care services. Consequently, evidence in this area was limited to equine-assisted interventions, walking groups, and adventure holidays, with one further evaluation [64] covering a mixed set of outdoor nature-based activities including gardening, nature walks, visits to farms, woodlands, or beaches. Tis evaluation noted that while project aims were to train provider organisations to ofer outdoor nature-based activities, only about 70% of the activities conducted could be described as nature-based, and about 10% of the naturebased activities were held indoors, refecting some of the potential challenges of implementation. While community initiatives to promote outdoor nature-based activity are relatively unlikely to be reported in the research literature in the absence of academic involvement of some kind, the range of community-based activities our searches identifed appeared reasonably consistent with the fndings of an internet-based search of community nature-based activity provision for older people with cognitive impairment in the UK conducted by the ENLIVEN team. Nevertheless, the activities reported in research and associated grey literature may not fully refect either the range of activities that have been attempted in community settings or the creative potential for imaginative development of innovative approaches.
While interventions are designed to address clinical need, it is essential to seek robust evidence of improvement in the relevant outcomes to better understand the mechanisms involved and to guide future practice. It was noteworthy that several studies included direct observation of participants to establish how they engaged in and responded to the activities ofered and to check that participants appeared to fnd the activities interesting and enjoyable [42,44,54,58,59]. However, although some of the included studies focused on common clinical issues such as problem behaviours or depression or considered levels of well-being and quality of life [41,57,60], in no cases were participants selected based on a specifc clinical need. In most studies, participants were included due to their diagnosis or because they attended a particular day care setting. Tis could create challenges for outcome evaluation. Where, for example, quality of life scores are already reasonable, or there are few symptoms of depression, it will be difcult to evidence improvements resulting from participation, even if qualitative accounts suggest that people enjoy the activities and fnd them benefcial. In these cases, engagement, satisfaction, and perception of beneft would appear to be the most salient outcomes. Tese kinds of person-centred outcomes are the key considerations for community-based initiatives and are important to inform the iterative process of developing and refning activities that meet the wishes and aspirations of older people with dementia and other forms of cognitive impairment.

Limitations
Our inclusion criteria had some signifcant implications. We aimed to include older people with any form of cognitive impairment. However, studies of people with conditions other than dementia, such as stroke survivors, typically did not recruit based on age or cognitive impairment. Terefore, we considered but had to exclude a small number of studies that did not meet criteria for either age or cognitive impairment, for example, studies of forest walking for poststroke depression and anxiety [65], nature-based rehabilitation for poststroke fatigue, anxiety, and depression [66], and equine therapy for gait, balance, and mood in Parkinson's disease [67,68]. Tis meant that in practice, all the included studies focused on people with dementia, a condition where the presence of cognitive impairment is a given, but a small proportion of people are diagnosed below the age of 65. We excluded only two small-scale studies involving people with dementia due to our age criterion: one focused on people with young-onset dementia engaging in a gardening group [69], and in the other, which explored a pilot programme of woodland activities, fewer than two-thirds of the participants were aged 65 or over [70]. Four articles were excluded because they provided insufcient information to be sure that our inclusion criteria were met. In three studies of community gardening programmes, the ages or diagnoses of participants were not known to the researchers [71][72][73], and for one study of woodland visits, no further details could be obtained [74]. To be included, studies had to report on activities that were nature-based and specifcally consider the efects of being in nature. Some studies described outdoor activities such as neighbourhood walking groups [75], a safe walking programme for people who "wander" in long-term care [76], or sporting or physical activities that take place outdoors such as golf [77] or trekking [78], but were excluded because they did not explicitly consider the efects of being in nature.
For pragmatic reasons, our systematic literature searches were limited to studies published in the English language. Te studies identifed were conducted primarily, although not exclusively, in English-speaking or other European countries, and all but one were undertaken in high-income countries. Beyond age and gender, we noted very limited reporting of basic dimensions of diversity such as ethnicity. We were unable to comment on the extent of inequities in provision and access due to ethnicity and other factors such as socioeconomic status, geographical location, area-level deprivation, and the availability of community resources and volunteers, but consider that all these factors could be salient, raising questions about how to ensure that provision is targeted to those in greatest need.

Implications
Recommendations for maintaining well-being among people with dementia emphasise the importance of enabling people to participate in activities that match their personal preferences and needs [79]. Tis implies that older people with cognitive impairment should be able to choose something that appeals to them from among a range of options that support inclusion and participation. As the health and well-being benefts of being in nature are wellestablished and increasingly understood [9][10][11], and older people with cognitive impairment can experience signifcant barriers to access, whether intrinsic or extrinsic, it is important that nature-based activities should be part of this ofer. What, therefore, does the literature reviewed suggest about how this can be achieved and what more do we need to know?
Te review indicates that nature-based activity can serve several diferent purposes. It can be ofered as an opportunity for meaningful and enjoyable occupation to enrich daily life, as a framework for the provision of day care, or as an intervention to address clinical needs. Regarding meaningful and enjoyable occupation, this review describes the kinds of outdoor nature-based activities ofered for this purpose and demonstrates that it is feasible to provide such activities and evaluate their benefts. It also highlights the relative paucity of research evidence about community nature-based activities, especially where participants are drawn from community sources rather than care settings, and the limited set of activities studied, which may not fully refect either what is available or the potential that exists to develop provision in this area. Tese points to the need for knowledge exchange among older people with cognitive impairment and their families, providers or potential providers of activities, and researchers to realise this potential. Greater understanding of preferences and needs will help to ensure provision is tailored appropriately and allow for creative development of new approaches, and careful attention to participants' responses and reactions will support efective and inclusive provision that can overcome intrinsic barriers such as loss of confdence. Addressing the extrinsic barriers that limit access, such as inadequate facilities and transport or being resident in a long-term care setting [46], is also essential, suggesting that providers, and funders of infrastructure, care providers and policy makers should be included in the knowledge exchange process. Community initiatives are often precarious and, due to their reliance on securing short-term funding, difcult to sustain; a stronger evidence base demonstrating the outcomes of such initiatives could help support sustainability. One implication for research that arises from the review is the value of providing as full a description as possible of study participants, something that was lacking in many of the studies in this area.
Regarding day care provision, the review demonstrates the feasibility of providing nature-based day care through the green care farm model, ofering a potential alternative to regular day care, and indicates that nature-based day care is also starting to emerge in diferent forms in urban areas. Te evidence base for green care farms is relatively elaborated and could be translated to inform further development and innovation in day care provision, nationally and internationally. Tis could make nature-based activity more widely accessible, with adaptations for diferent contexts, cultures, and practices such as the use of rice farming in Japan [49] or new initiatives in urban areas [43].
In the case of interventions to address clinical need, outdoor nature-based activities can be conceptualised in one of two ways. Tey can be intended as general approaches to enriching people's experience and improving well-being that can contribute to reducing or preventing other problems; for example, exploring whether participation in an equineassisted intervention reduces subsequent levels of "disruptive behaviour" in the day care setting [41]. Alternatively, activities can be designed to address specifc clinical symptoms; for example, exploring whether an equine-assisted intervention that includes riding the horse afects balance, agility, and muscle strength [60]. Tis kind of approach could potentially ofer a more engaging alternative to standard physiotherapy exercises, with practice of relevant movements incorporated into an enjoyable activity taking account of personal interests and preferences. However, robust evidence of benefts relative to other treatment options or care arrangements is essential, and this review suggests that such an evidence base largely remains to be developed.
Te studies included in the review capture activities that are varied in terms of their nature, content, frequency, and duration, and in terms of the resources required to provide them. Broadly, however, in these studies, engagement with nature mainly involves being active in the natural environment as part of a group in one way or another, whether by walking, working with plants, or interacting with animals. Although diverse, these activities do not necessarily constitute the full range of possible options, and there is considerable scope for innovation to expand the choices on ofer, make these available to a wider range of people, evaluate their impact, and share the knowledge gained. Specifying the precise contribution that engaging in nature-based activity makes to well-being and health over time and to the trajectory of impairment in progressive conditions, however, would be a more complex undertaking requiring large samples, diverse activities, and long-term follow-up.
In summary, the review has highlighted three signifcant future research gaps. First, there is a need to extend involvement to people with dementia who are not attending or residing in formal care settings, and to address inequities in provision by including people from diverse backgrounds and circumstances, to ensure that those most in need have the opportunity to beneft. Second, there is scope to develop more innovative approaches encompassing a wider range of activities to suit diferent interests, preferences, and needs, and to identify solutions to some of the practical challenges of implementation. Tird, unless the activity is intended to address specifc clinical symptoms such as balance or gait, the main aim of developing and providing outdoor naturebased activities for people with dementia should be to promote inclusion, engagement, pleasure, and satisfaction among those participating. Tis means that the focus of change primarily rests with providers or potential providers, who may need to adapt both what they ofer and how it is ofered by making sites more accessible. Although there is growing awareness of this among businesses and other organisations that ofer access to the natural environment, there is a need for evidence to support initiatives that address this gap [26,80]. Te wider ENLIVEN project aims to contribute to this much-needed organisational and business transformation by working with providers to develop and implement innovative approaches and evaluate the resulting organisational changes from multiple perspectives that may then be scaled up to make the natural environment more dementia-inclusive.

Conclusions
Access to outdoor nature-based activity for older people with dementia or other forms of cognitive impairment is not just a matter of good practice, but a fundamental right, currently denied to many [18]. Tis lack is all the more worrying as it restricts quality leisure time that could be used to enhance quality of life. Tis scoping review has identifed pioneering research that reinforces this argument, reporting on the development, implementation, and evaluation of outdoor nature-based activity provision for people living with dementia, whether as an opportunity for meaningful and enjoyable occupation to enrich daily life, as a framework for the provision of day care, or as an intervention to address clinical needs. Our review of the extent and nature of the available evidence indicates that this is an emerging feld of both practice and research. Mainstreaming access to outdoor nature-based activity to ensure it continues to be a part of everyday life for older people with dementia and other forms of cognitive impairment is a key challenge that must be overcome to ensure those who value it and fnd it benefcial do not have to negotiate key barriers to access. To achieve greater access and to expand the reach of nature for older people living with dementia and other forms of cognitive impairment, our fndings reinforce the role of knowledge exchange among all parties involved as a basis for further development and evaluation. Te natural environment remains an underutilised resource that could make positive contributions to the quality of life and well-being of older people with dementia and other forms of cognitive impairment. As a future area of both dementia practice and policy, this is set to grow in both signifcance and be recognised for the added value it can provide in addressing isolation and giving people agency in taking greater ownership of their own well-being and support.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon reasonable request.

Additional Points
What Is Known About Tis Topic. Older people with cognitive impairment may experience isolation and loneliness and lack opportunities for social contact or meaningful activity. Engaging with nature and the outdoor environment benefts well-being. Older people with cognitive impairment can potentially beneft from being able to get outdoors and connect with nature but experience signifcant barriers to access. What Tis Paper Adds. Research has focused mainly on including outdoor nature-based activity in day care provision. Few studies have evaluated community provision of outdoor nature-based activity, focusing mainly on equineassisted interventions, walking groups, and adventure holidays. Tere is a need to develop and evaluate community provision of a wider range of outdoor nature-based activity for older people with cognitive impairment.