Health-promoting and preventive interventions for community-dwelling older people published from inception to 2019: a scoping review to guide decision making in a Swedish municipality context

Background Despite the promising evidence of health-promoting and preventive interventions for maintaining health among older people, not all interventions can be implemented due to limited resources. Due to the variation of content in the interventions and the breadth of outcomes used to evaluate effects in such interventions, comparisons are difficult and the choice of which interventions to implement is challenging. Therefore, more information, beyond effects, is needed to guide decision-makers. The aim of this review was to investigate, to what degree factors important for decision-making have been reported in the existing health-promoting and preventive interventions literature for community-dwelling older people in the Nordic countries. Methods This review was guided by the PRISMA-ScR checklist (Preferred Reporting Items for Systematic reviews and Meta-Analysis extension for Scoping Reviews), the methodological steps for scoping reviews described in the Arksey and O′Malley’s framework, and the Medical Research Council’s (MRC) guidance on complex interventions. Eligible studies for inclusion were randomised controlled trials (RCTs) concerning health promotion or primary prevention for community-dwelling older people implemented in the Nordic countries. Additionally, all included RCTs were searched for related papers that were reporting on additional factors. Eligible studies were searched in seven databases: PubMed, SCOPUS, CINAHL, Academic Search Elite, PsycINFO, SocINDEX, and SPORTDiscus. Results Eighty-two studies met the inclusion criteria (twenty-seven unique studies and fifty-five related studies). Twelve studies focused on fall prevention, eleven had a health-promoting approach, and four studies focused on preventing disability. All interventions, besides one, reported positive effects on at least one health outcome. Three studies reported data on cost-effectiveness, three on experiences of participants and two conducted feasibility studies. Only one intervention, reported information on all seven factors. Conclusions All identified studies on health-promoting and preventive interventions for older people evaluated in the Nordic countries report positive effects although the magnitude of effects and number of follow-ups differed substantially. Overall, there was a general lack of studies on feasibility, cost-effectiveness, and experiences of participants, thus, limiting the basis for decision making. Considering all reported factors, promising candidates to be recommended for implementation in a Nordic municipality context are ‘Senior meetings’, ‘preventive home visits’ and ‘exercise interventions’ on its own or combined with other components.


Background
The population across the world is growing older which calls for effective health-promoting and preventive interventions in order to help older people maintain a good quality of life. In accordance with the World Health Organisation (WHO), health promotion is defined as the process of enabling the population/individual to increase control over and improve their health, while disease prevention is defined as measures taken to prevent the occurrence of disease or limit its development [1,2]. The implementation of health promotion and prevention is imperative given that increased levels of dependency in managing activities of daily living (ADLs) is related to a reduction in self-rated health [3] as well as higher societal costs [4]. In Sweden, municipalities have a responsibility to address health concerns and social care needs among older people ultimately aiming to optimize the person's quality of life by promoting independence and opportunities to participate in society [5]. Therefore, municipalities need to consider health promoting and preventive interventions besides, and to complement, the provision of social care. Such interventions can promote various aspects of the health and well-being of older people by strengthening the person's opportunities to be active and participate in society [6]. Simultaneously, a more health promoting approach to the provision of municipality services for older people could reduce the expected increase in health and social care costs.
Several studies show that health promotion and prevention in different forms have resulted in a range of positive effects such as maintenance of ability to perform ADLs [7], enhanced quality of life [8,9], prevention of functional decline [10,11], and reduced falls [12]. In addition, some interventions have shown to be cost-effective [13,14]. In all, examples in the previous literature indicates that positive effects can be achieved from both multi-professional and single-professional interventions [10,15], from both short and long-term interventions [16,17] and both group-based and individual interventions [10,18]. Even though the existing evidence is promising in improving health outcomes among older people, the range of interventions have varied considerably regarding their content, design and outcomes used, making them hard to compare [19]. Since resources (e.g. staff) are limited, not all promising health-promoting or preventive interventions can be implemented. Thus, more information than mere evidence on effects, based on single trials, is needed to provide sufficient guidance for decision-makers on what type of intervention to implement [20].
The question of which interventions to implement needs to be guided by a systematic decision-making process based on the best available evidence [21]. In this systematic process, a range of factors need to be considered, e.g. intervention design, effects, cost-effectiveness, feasibility of recruitment and intervention procedures as well as an understanding of how participants experience the intervention. The challenge with this task is that many health-promoting interventions often miss to report all such information relevant for decision making [22,23]. In addition, the issue of context should be considered when assessing how evidence can be transferred from controlled trials to clinical settings [24]. In this study, the context is focused on the Nordic countries, because these countries, to a large extent, share similar welfare systems characterized by publicly funded health and social care.
A scoping review design has been proposed as an effective tool to disseminate research findings and provide an overview of evidence for decision-makers and policymakers [25], and is especially appropriate when exploring a heterogeneous or complex body of literature [26].
Given the potentially positive effects on older peoplesh ealth and the cost-effective use of societal resources, a comprehensive overview of the existing evidence on health promoting and preventive interventions is needed. Therefore, the aim of this review was to investigate to what degree factors important for decision-making have been reported in the existing health-promoting and preventive interventions literature for community-dwelling older people in the Nordic countries.

Methods
This scoping review follows the PRISMA-ScR checklist (Preferred Reporting Items for Systematic reviews and Meta-Analysis extension for Scoping Reviews) [27] as well as the methodological steps for scoping reviews described in the Arksey and O′Malley's framework [25]. The Arksey and O′Malley's framework consists of five stages: 1) identifying the research question; 2) identifying relevant studies; 3) selecting studies; 4) charting the data; 5) collating, summarizing and reporting the results [25]. This scoping review has been conducted following an unpublished work plan.

Identifying the research question
Health promotion and prevention often include several interacting components and can, therefore, be considered as complex interventions. The Medical Research Council's (MRC) guidance for the process of developing, evaluating and implementing complex interventions was used to identify the research questions of this scoping review [28]. According to the MRC guidelines, this process includes several phases in which evaluations of feasibility, effectiveness and cost-effectiveness provide essential knowledge. In addition, the PICO framework (Population, Intervention, Comparison, Outcome) which is recommended to frame the research question but also to guide the whole process in a review, was used as an additional source in guiding the formulation of the research questions regarding the population, intervention/control and effects on possible outcomes [29]. Hence, the research questions were: 1. In which contexts have interventions been conducted? 2. For which populations have interventions been conducted? 3. How have the interventions been designed (e.g., which components, duration of interventions and mode of delivery)? 4. Which feasibility aspects have been described? 5. How have the participants experienced the interventions? 6. Were interventions effective, and on which outcomes? 7. Were interventions cost-effective?

Eligibility criteria
The eligibility criteria were defined in advance but were modified with increased familiarity with the literature. Eligible studies were: 1) interventions categorised as health promotion (HP) or primary prevention (PP) following the WHO's definition [1,2] and addressing behavioural risk factors, injury prevention, physical health, social and mental health, 2) including populations of community-living older people 65+ as of it being the lowest retirement age in the Nordic Countries, hence exclude the risk of missing relevant studies due to the age limitation, 3) implemented in a Nordic country (Denmark, Finland, Iceland, Norway, Sweden and Faroe Islands), 4) studies applying a randomized controlled trial design (RCT) for the evaluation of effects (research question six), 5) studies related to the identified RCTs addressing the remaining research question, e.g. experiences of participants, feasibility as well as studies on cost-effectiveness. Only studies written in English were included and to decrease the risk of missing relevant articles, no year limit was applied.
The exclusion criteria were: secondary prevention programmes related to a specific disease or diagnosis e.g. interventions implemented for participants with a neurological condition such as stroke or Alzheimer's disease, tertiary prevention programmes (e.g. rehabilitation, hospital discharge) as well as studies in populations with extensive needs for support in ADLs. Furthermore, interventions focusing on dental health promotion; interventions targeting older people with cognitive malfunction; programmes assessing effects of medication or evaluations of effects only focused on specific body structures [30], were also excluded.

Information sources
Seven online databases were searched: PubMed, SCO-PUS, CINAHL, Academic Search Elite, PsycINFO, SocINDEX, and SPORTDiscus. In designing the most suitable search strategy, a librarian at Umeå University was consulted on several occasions. The search strategy was based on a combination of words to capture key terms related to the purpose of this study: "health promotion", "prevention", "old people", "community-dwelling", "Nordic countries", "Randomised controlled trial" and their synonyms/alternative words. A detailed outline of the search strategy, including the full syntaxes to screen the databases and numbers of search results, is available in Additional File 1. The initial search strategy was piloted and refined in the light of early findings. The search for literature was conducted from inception to January 9, 2019 (last date searched).
Identification of studies, relevant to this review, was done in two stages. At the first stage, we identified RCTs in the field of health promoting and preventive interventions for community dwelling older people conducted in the Nordic countries. To decrease the risk of missing relevant studies during the first stage of identifying studies, we did not limit our search to only primary prevention programmes. We applied this inclusion criterion when screening titles and abstracts for study selection. In the second stage, reference lists of identified and selected studies from the first stage (the RCTs) were examined for the purpose of identifying related studies, i.e. studies evaluating the same intervention but at different follow-ups, looking at different outcomes, or addressing the other research questions.

Selecting studies
Search results were exported in EndNote reference manager, which was used to remove duplicates. In the next step, the EndNote reference manager was used to ease the process of identifying and excluding irrelevant studies through searching for key exclusion terms (hospital discharge, cognitive malfunction, dementia etc.). Titles and abstracts of the remaining studies were organised in an excel document and read independently by all authors. Studies that all authors agreed did not meet all of the eligibility criteria were removed. In cases of uncertainty, disagreement was resolved by reading the whole study and discussion among the three authors. After screening titles and abstracts and excluding studies not meeting the inclusion criteria, the remaining studies were read in full text.

Charting the data
In line with the process of identifying research questions, the MRC framework and the PICO framework were used to guide the process of data extraction. The included studies were distributed between authors SB and MZ who independently charted the data for summarizing information related to the research questions, each question targeting one of the seven factors: context, population, intervention content, feasibility, experiences of participants, effects and cost-effectiveness. Disagreement was resolved through discussion between all authors. All authors read the extracted data and discussed the results. Main results are presented in the text under a specific heading for each of the research questions. Results are presented and described by referring to either the original study/ study (At first-hand study protocol, if available. If no study protocol was identified we referred to the first published RCT), related studies (other publications related to the original study) or intervention (referring to the specific interventions evaluated in each study).
In the section below there is a description of the factors (data items) extracted to address the research questions.

Data items
To the extent available, data on context, population, intervention content, feasibility, experiences of participants, effects and cost-effectiveness have been extracted from the included studies. The extraction of data regarding intervention context focused on identifying the setting (e.g. primary care, clinical, home, physical activities facilities) in which the specific intervention was evaluated as well as the country, and if available, the municipality in which the study was conducted. Data extracted on population concerned how the target population was defined in age, frailty/morbidities, gender, and socio-economic status. The data extracted concerning feasibility was specifically focused on identifying participation rates and retention. If a pilot or feasibility study was published, the aim and main results of the study were also extracted. Information on experiences of participants was extracted from related qualitative studies, and main results on experiences of participants were summarised. Effects were examined by extracting effects on specific health outcomes at different time-points as reported in each study. In general, the data extracted regarding effects included effect sizes if reported, confidence intervals and p-values for outcomes for which a statistically significant difference was reported. No effect sizes, confidence intervals or p-values were extracted for outcomes upon which no significant difference was reported, they are mentioned in text however.
The first step in exploring cost-effectiveness was to identify if such studies had been conducted. The primary objective when looking at identified studies on costeffectiveness was to examine if evaluated interventions were found to be cost-effective and in relation to which outcomes cost-effectiveness was established. Furthermore, if available, data concerning methodological aspects of such studies were extracted, e.g. perspective used (health provider/payer or social perspective), outcome-and cost measures and how they were affected by the specific intervention, comparator (e.g. no intervention, alternative intervention) and time horizon (over which time horizon costs and effects were measured) [31].

Results
The search yielded a total of 690 studies. After removing duplicates, 381 titles and abstracts were screened and studies obviously not meeting the inclusion criteria were excluded. All remaining studies were read in full text (n = 35) and studies which did not meet the eligibility criteria were removed (n = 8). All 27 original studies, identified in stage 1, were in stage 2 reference checked resulting in 55 related studies being identified and included.
In all, a total of 82 studies were included for analysis, 27 original studies and 55 related studies. The search process is presented in a PRISMA flowchart in Fig. 1.

Overview of original studies
The total number of participants in the included studies (extracted primarily from the original studies, if available) was 34,238. One municipality-based study included a very large sample (n = 24,365) [40]. Considering all studies except the one by Poulstrup and Jeune [40], sample sizes varied from 30 participants [16] to 4030 participants [53]. The duration of interventions varied from a one-session discussion group [8] to three weekly group exercise sessions over a period of one year [54]. Of the 27 original studies, 12 focused specifically on fall prevention (looking primarily at fall-related parameters and fall risk factors, e.g. falls, fear of falling, balance performance, bone mineral density) [17, 32-35, 37-41, 55]. Eight fall prevention interventions were single component and included only exercise [10, 32-36, 38, 39], while five combined an exercise component with one or more different components, e.g. preventive home visits (PHV), discussion groups, nutrition, medication review [17,37,40,41,55]. Eleven studies had a health promoting approach. Five of these studies focused on promoting general health (interventions which in addition to focusing on functional status also focused on health-related quality of life and/or social support aspects) [8,18,36,43,44], four promoted exercising [48][49][50]56], and two focused on promoting mental wellbeing [16,45]. The four remaining studies focused on preventing disability [46,[51][52][53]. Findings on intervention type, intervention aim, context, and population are presented below in Table 1. These findings are also described in the text, separately for each factor, in the sections below.

Overview of related studies
There were no related studies identified for 12 of the 27 original studies, so all 55 related studies found were linked to only 15 of the 27 original studies. Of the 15 original studies: one study reported results in nine related studies [43], two reported in seven related studies [41,44], and one reported in six related studies [18]. The 11 remaining interventions reported results in one to five related studies. For further details, see Table 2 below.
Among the 55 related studies, 38 included evaluations of effects, eight were qualitative studies analysing experiences of participants, four were health economic evaluations, three were study protocols, and two were pilot studies. Findings on intervention content, effects and feasibility aspects are also described separately in the sections below, while detailed information on these factors is presented in Table 2.

Population
The population targeted in the included studies varied regarding age and health-related conditions. In six studies, the target population was defined in relation to age and location of residence [16,33], four of these studies were municipality-based and targeted a broad population of older people from several municipalities [40,45,53,107]. The remaining studies defined the target population in relation to age and location of residence/municipality in     [48], whereas two applied a narrow age span 77-82 [8], 75-79 [45]. One study reported only the mean age of the participants [39]. Five studies had samples consisting only of female participants [32,      Table 2 Detailed results concerning intervention content, effects on health outcomes, and feasibility aspects of included studies in the field of health-promoting and preventive interventions for community dwelling older people in the Nordic countries from inception to 2019 (Continued) Original study

Related studies
Intervention content Effects (significant between-group differences) Feasibility aspects month after the intervention period.
The number of sessions included in the interventions varied, as did the duration. For individually-based interventions, the number and duration of sessions ranged from one single home visit [43,44] or one personal counselling session on nutrition [46] to daily independently performed exercise sessions (5-15 repetitions) over a period of 16 months [52]. Group-based components ranged from one single discussion group [8] to three 50 min exercise session a week for over one year [10], while the education programme for home visitors included regular education over a period of three years [53].
Studies combining group and individually-based components ranged from one single home visit and four discussion groups [43,44] to two weekly exercise sessions over one year in combination with monthly lectures on various themes and psychosocial activities combined with a single individual geriatric assessment and counselling on fall prevention [41].

Feasibility aspects
Feasibility aspects were reported sporadically across studies. All interventions reported on methodological aspects Table 2 Detailed results concerning intervention content, effects on health outcomes, and feasibility aspects of included studies in the field of health-promoting and preventive interventions for community dwelling older people in the Nordic countries from inception to 2019 (Continued) Original study

Related studies
Intervention content Effects (significant between-group differences)

Feasibility aspects
Duration: Regular education (17 municipal meetings) for home visitors during 3 years and one education programme (2 h) for GPs in the first year Control: No intervention (education program) for home visitors in another 17 control municipalities.
Notes: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. a Maximum score for FES-1 = 28, higher score implies higher concern for falling, lower score implies lower concern for falling, b Maximum score for BBS = 56, higher score implies higher degree of functional balance and vice versa, c Maximum score for 13-item of feasibility such as recruitment and retention/dropout numbers. With recruitment numbers, we refer to the total number of eligible participants (meeting inclusion criteria) who agreed to participate in the study. The mean recruitment rate (eligible participating population/total eligible population) in all the studies included in this review was 63%, varying from 9% [49] to 100% [33,35,41,46,52]. However, there was some inconsistency regarding how the eligible population was defined. For instance, in one study the total eligible population consisted of only those who volunteered [33], or of the population receiving an invitation [53] or the whole population in a specific community [40]. Thus, participation rates are not consistent among included interventions and this inconsistency should be taken into consideration when interpreting the mean recruitment rate. Mean retention rate in the total number of original studies included in this literature search was 85%. Retention rate varied from 37% [51] to 99% [53]. Beside the information related to recruitment and retention rates, only two feasibility/pilot studies were identified [83,92]. Kristensson et al., investigated the feasibility of a case management intervention by specifically assessing sampling and sample characteristics as well as possible effects on perceived health [92]. Lood et al., (2016) investigated the feasibility of evaluating senior meetings in the "Elderly in the risk zone" intervention [43] among a specific group of older people (foreign-born) by specifically assessing recruitment and retention rates, questionnaire administration, and variability of data [83].

Participants' experiences
In relation to five of the original studies, eight related studies explored the experiences of participants [75,76,82,85] or both the experiences of participants and professionals delivering the intervention [61,62,87,91]. Based on qualitative methods and interviews, participants' experiences were described related to i) a single preventive home visit (PHV) [75], ii) senior meetings [76,82,85,87], iii) multidisciplinary fall prevention programmes [61,62], and iv) case management intervention [91]. Findings from interviews on PHVs showed that home visits contributed to empowerment and increased selfesteem by making participants feel in control over their health. However, for some, it did not come at the right time, either because they felt too healthy to benefit from it or because they felt too ill to be able to participate [75]. Findings on senior meetings revealed that although independent older people may find it difficult to accept or act upon health-promoting information, the discussion groups, provided in a multi-dimensional approach, could motivate acting upon such information, and thus, senior meetings were perceived as a "key to action" [76]. These findings were in line with experiences of foreign-born older people who felt empowered by the opportunities gained, such as the possibility to meet other people, discuss experiences, as well as become acquainted with possibilities to make everyday life better and safer [82]. However, their capabilities to adhere and act upon knowledge in the long-term (six months to one year after their participation in the programme) was dependent on personal and environmental resources [85]. Furthermore, professionals delivering the interventions, revealed that for a senior meeting intervention to succeed in reaching out to the target group, it is necessary to recognise the person's resources and empower their capabilities in maintaining health [87].
Empowerment and raised awareness were also emphasized in a group-based multidisciplinary fall prevention program delivered through a client-centered approach. The involved professionals observed that building trust and a safe atmosphere within the group increased participants' engagement in discussions which contributed to the success of the intervention. A contributing factor for creating this sort of atmosphere was the role-shifting negotiated by the group leaders from being the expert to being a facilitator of the discussion [61,62]. However, it was noticed that for a group format to be successful, group composition should be taken into consideration for the participants to feel fellowship [61,82]. Furthermore, in a home-based case-management intervention, participants experienced case managers as a helping hand in navigating within the health system, and thus, contributed to feelings of control and safety [91].
Additionally, experiences of participants were explored as secondary outcomes through a survey related to a Nintendo Wii training fall prevention intervention [38], or through a single open-ended question related to a telephone-based health-promoting intervention [16]. Findings from the survey showed that training with a digital device (Wii) was experienced positively and did not lead to any adverse effect [38]. A self-care telephone intervention influenced participant's attitudes positively, e.g. towards self-care [16].

Effects
For several interventions, effects were evaluated in relation to a wide range of outcomes, and all, besides one intervention on nutritional counselling [46], reported a positive effect on at least one health outcome evaluated in comparison to a control group. However, the magnitude of effects and follow-ups at which interventions were evaluated, varied substantially and therefore, should be taken into consideration when evaluating effects. To summarise intervention effects, we classified health outcomes in broader categories (Table 3). For example, Balance confidence, Balance performance, Dynamic balance, Impaired balance, Postural balance, Postural sway, Velocity moment in standing balance, are categorised under "Balance". Details on effects are found in Table 2.

Cost-effectiveness
Four studies presented a health-economic evaluation. Three studies adopted a cost-effectiveness analysis method [13,72,102] and one a cost-utility analysis method [90]. Two studies provided an economic evaluation of single interventions; a case-management intervention [90] and an education programme for home visitors [102]. The other two studies compared different interventions focused on health promotion [13], and falls prevention [72]. In these four studies, a societal perspective was chosen including cost from different sectors e.g., health care and social care. The time horizon used varied from three months [13], one year [13,90], two years [72] and up to three years [102]. All studies based their estimates of costs on intervention costs, healthcare costs and municipality costs. In addition, the value of informal care was included in one study [90]. Costeffectiveness was evaluated in relation to active life-years gained [102], quality-adjusted life-years (QALYs) [13,90] and number of injurious falls prevented [72].
Findings from the economic analysis showed that two interventions were considered cost effective [13,72] whilst two were not [90,102]. A one-session discussion group was found to be more cost-effective when compared to an individual intervention or an activity group in an intervention comparing three different occupation-focused healthpromoting interventions to a control group [13]. The discussion group showed significant effects on QALYs gained at 3 and 12 month follow up's and lower total costs [13]. Furthermore, an exercise intervention showed high probability to be cost-effective in preventing falls in relation to a threshold of 3000 euro per injurious fall prevented when compared to three other fall preventive interventions focusing on exercise and vitamin D supplements [72]. In contrast, no significant difference was observed in total costs or QALYs gained when comparing a case management intervention to no intervention in a cost-utility analysis. Nevertheless, the case management intervention led to lower levels of informal care and need for help with instrumental ADLs [90]. Neither did a training programme for home visitors result in significant differences in total cost or active life-years gained in comparison with usual practice of performing preventive home visits [102].

Discussion
This scoping review provides a comprehensive overview of health-promoting and preventive interventions for community-dwelling older people in the Nordic countries that to some extent, can guide decision-making in a Swedish municipality context. However, while all included studies report some positive effects, not all potentially effective interventions can be implemented since resources are limited. Thus, the evidence on effects needs to be critically reflected upon, but several other factors need to be considered as well. Our study exposes gaps in knowledge regarding cost-effectiveness, experiences of participants and feasibility of the interventions, knowledge that could broaden the understanding of which interventions seem most promising and feasible to implement from a decision-makers´perspective. While the scope of this review includes interventions with different foci, the summary of findings on the seven evaluated factors, show that some interventions such as senior meetings, preventive home visits (PHV) and exercise interventions alone or combined with other components, seem to be strong candidates for implementation, e.g. [10,43,50]. In all, the total evidence for these interventions included positive effects on a range of outcomes, in some cases confirmed by evaluations at different follow-ups, with established cost-effectiveness, and supported by qualitative findings based on the experiences of participants.
In the section below we provide a deeper discussion about the previously mentioned intervention examples and argument how the findings from this review could guide decision making and how additional knowledge, generally missing across the different interventions, is needed to better guide decisions on which interventions to implement.
Senior meetings, one type of intervention investigated in four different studies, seems potentially effective in promoting general health and wellbeing among communitydwelling older people [8,37,43,44]. The study which provides the broadest evidence base is the "Elderly Persons in the risk zone"-study conducted in Gothenburg [43], which evaluated a four-sessions senior meeting intervention combined with a home visit. Several related studies support the implementation of senior meetings given the positive results on a range of health outcomes, e.g., physical function [77] and ADLs [81], outcomes for which effects were established at different follow-ups (3 months to 2-year follow-ups). Qualitative findings on the experiences of participants also provide an understanding of why the intervention was effective by concluding that senior meetings were experienced as a "key to action" in empowering participants to engage in preventive approaches to improve health [76]. The benefits of senior meetings, albeit with other content, were also verified in the studies by Zingmark et al., [8] and Johansson et al., [37]. In the study by Zingmark et al., [8] two group-based formats of interventions (a discussion group and an activity group) were implemented by occupational therapists which both resulted in positive effects. In our results, evidence on costeffectiveness regarding senior meetings was limited to the study by Zingmark et al., who found a one-session discussion group to be the most cost-effective intervention format [13]. Recently, however, a publication based on data from the "Elderly Persons in the risk zone" supports the cost-effectiveness of senior meetings as well, even in the long term (over four years) [109]. Thus, senior meetings seem to be a strong candidate for implementation in a Swedish municipality context. Yet, the exact format can be further discussed given the variation in the number of sessions and the specific content, e.g. one session discussion [8], four sessions combined with a home visit [43], twelve sessions combined with two home visits [37]. In addition, feasibility aspects related to recruitment during implementation in a municipality context seem to be a critical feature to improve reach in the intended population, thus requiring specific contextual knowledge [110].
Our results show that PHVs have the potential to improve general health by preventing deterioration in health in community dwelling older people. However, PHVs have varied regarding the specific format e.g. from one visit [43] to twelve visits [18] and have shown positive effects on several outcomes e.g. limiting progression in morbidity [78], reducing the number of emergency department visits [18], maintaining ADL ability [8] reducing lower extremity fractures [40]. Positive effects were also reported for an education programme for the home visitors conducting the PHVs, in terms of lower admission rates to nursing homes for those receiving two home visits per year [103]. The most promising results on PHVs were established in the "Elderly Persons in the risk zone" study where a single home visit was evaluated and showed positive effects ADLs [81], frailty and fear of falling [77], life satisfaction and morbidity [78]. This study was the only one, among PHV interventions, to conduct a 2-year follow up at which some effects persisted and thus validates post-intervention effects [81]. The positive effects of PHVs in the "Elderly Persons in the risk zone" study are partly explained by the experiences of participants, who felt empowered and in control as a result of the information given and having the opportunity to discuss health-related matters with a qualified professional [75]. However, these findings on long-term effects are in contrast to a previous PHV trial that indicated that intervention effects remained only for as long as the home visits were ongoing [111], and thus, highlights the importance of long term follow-ups over. Conflicting results regarding specific effects of PVHs and their health-economic effects have been reported also in a recent report from SBU Enquiry Service (Swedish Agency for Health Technology Assessment and Assessment of Social Services) about preventive home visits, also referred to from the Swedish National Board on Health and Welfare [112]. In some studies, though, PHVs have shown to be cost-effective while annual follow-up visits can be potentially even more costeffective. Such findings have been established when conducting health economic analysis based on data from the Elderly Persons in the risk zone [109] as well as in a previous Swedish study including twice-annual home visits over a period for two years [111]. Despite the conflicting results on some outcome effects of PHVs [113], they still can be considered a good alternative to group-based interventions, e.g. senior meetings, since not all potential participants can or like to engage in a group format.
Interventions including exercise or combining exercise with other components (e.g. medication review, guidance on nutrition, cessation of alcohol and smoking, home hazard assessment and modifications) showed to be promising for preventing falls. Findings on these interventions showed improvements in different factors related to falls risk and physical functioning, e.g. muscle strength, mobility, balance or self-rated health [34,35,38,59] which could indirectly lead to fall reduction [114]. Positive effects were observed for both home-based [34] and group-based interventions [39], regardless of whether they were shorter (3 months) [35] or longer (1 year) in duration [33]. Furthermore, interventions including more frequent group sessions reported additional effects, such as improvement in motivation to continue with physical activity [36,47], and perhaps consequently a reduction in injurious falls and fractures, as reported in two fall prevention interventions [10,17]. Both interventions included balance exercise in combination with resistance/strength exercise provided over one year or longer, but varied in terms of content, number of sessions, and delivery approaches used e.g. multifactorial [115] and multiple components [17]. In line with evidence from a recent systematic review and metaanalysis, exercise-based interventions, aiming to improve balance and strength, are one of the most feasible and cost-effective approaches to prevent falls among older people living in the community [114]. This approach has also been integrated into some current Swedish guidance, on physical training, balance and more, issued from the National Board on Health and Welfare in the form of training for professionals working with older people and fall prevention [116]. However, effectiveness of exercisebased interventions is dependent on the uptake and longterm adherence [117]. Groups sessions led by professionals over a longer period (1 year or more) seems to affect this aspect positively but can be costly, foremost in terms of human resources needed if provided to a large population of older people. Since group training might not be the solution for all, other effective alternatives such as multifactorial interventions could work in these cases. Also, multifactorial interventions have shown positive effects on preventing falls [118] and could be considered an alternative to exercise-based interventions. Nonetheless, no health-economic evaluation was identified for these interventions, and thus, still makes them less robust in terms of cost-effectiveness.
While our results, indicate that there are several healthpromoting and preventive interventions that could improve health and well-being among community-dwelling older people, implementation needs to be considered, not only in relation to effects but also concerning the resources available, i.e. how limited resources can be used in a way that yields the largest health benefits [20,21] and other feasibility aspects such as reach in the population; a key factor for successful implemtation of research in practice.
Health economic evaluations, including evaluation of both costs and effects, can provide such important information. However, in this scoping review, only four health economic evaluations were identified, indicating a general lack of information to guide decision making. However, information regarding intervention content, e.g. duration and intensity of interventions, can at least provide some information about the resources required. Regarding individual interventions, the study by Dahlin-Ivanoff et al. included one single preventive home visit requiring one and a half to two hours of a professional's time [43] in contrast to the study by Möller et al. in which a case management intervention, required at least one hour per month during a 12-month intervention of professional's time [18]. Similarly, for group-based interventions, the span for the time required was two hours for a one session discussion group [13], to two and a half hours per week over the course of one year [10]. While these examples all include interventions with some positive effect, the time for which staff need to be allocated differs substantially. Even though these examples lack information on other types of costs that can be affected by interventions (e.g. social care consumption), they provide some guidance on which resources are needed and the magnitude of staffing which is a central cost of a healthpromoting or preventive intervention [13].
Despite a growing literature of health-promoting and preventive interventions that have shown positive effects in well-controlled trials, the translation of such trials to practice has proven to be challenging [20]. Evidence has shown that feasibility or pilot studies are important to ensure effective practical implementation and to decrease threats to validity of health outcomes [119]. However, in our literature search, there was a lack of piloting and feasibility studies. In the absence of feasibility or pilot studies, other reported aspects such as information on study participation rates and adherence could indicate the degree to which an intervention reaches out to the target population, and thus, increase chances of a successful translation of research evidence into clinical practice [119]. Reaching older people with health promotion is crucial for achieving a health impact for the whole population, but has also been shown to be challenging [110,120]. Findings from all 27 original studies, in this review, showed that approximately a third of the persons eligible declined to participate due to different reasons, i.e. being too sick or too healthy [75]. Qualitative data on experiences of participants could to some extent reveal why an intervention is or is not appealing to larger groups of older people, however, only a few studies on experiences of participants were identified in this review.
While this review provides some guidance on which interventions have shown positive health effects in a Nordic context, future research is needed on how to translate evidence into practice, e.g. through exploring alternative ways of reaching out to a larger population and incorporating support for behaviour change and adherence in the long-term. Some examples of new promising approaches explored in this review were Wii training [38] and physical activity counselling [50]. The digital approaches used through video training or self-care telephone calls are potentially feasible to be implemented considering the more limited resources required to implement them, e.g. the smaller number of direct personal contacts needed with providers of health care for older people while still resulting in positive effects. In light of the ongoing coronavirus pandemic and related measures of social distancing, the importance of addressing loneliness and isolation among older people is accentuated. Digital approaches to delivering effective interventions could complement the challenge of isolation and the need to reach out to a higher number of older people. For example, using smartphones and tablets may be a potentially cost-effective way to increase reach in the population. At present, there is a big supply of smartphone applications for exercise, however, most lack evidence regarding their scientific and implementation validity in the older population. Research in the area is, however, developing and one example is an ongoing large clinical trial on digital fall prevention in Sweden [121].
Finally, in discussing the results of this study, it is notable that some important aspects of healthy ageing, were less frequently evaluated. Only two studies focussed on mental wellbeing and social participation, one showed some effects in reducing loneliness [45] and the other in improving general mental health [16]. This gap in research has also been supported in other reviews, where promoting wellbeing and mental health have shown to be both effective and potentially cost-effective [122,123], and should, therefore, be further researched.

Strength and limitations
The scope of this review was broad. It included information on several factors extracted from all identified original and their related studies, and therefore provides an overview of the knowledge base in the field of healthpromoting and preventive interventions in the Nordic countries. Given the broad scope of this review, we choose to not include some information, e.g. data concerning when studies were performed or adverse events, which could be seen as a limitation of the study.
Data concerning when studies were performed would enrich information on the context and content of the interventions. However, the description of the study period, e.g. the period for the recruitment of participants, have not been reported consistently among all studies, therefore might not have produced many data. Although a wide range of outcome effects was extracted, important information on adverse events was not extracted and beyond the scope of this study, guided primarily by the MRC guidelines. Additionally, recent systematic reviews show that adverse events, for example, concerning fall prevention programmes seem to be rather poorly reported hence, would probably not make a significant difference in our conclusions, if included in the analysis [12,118]. Another important factor to consider, which may lead to better developed and evaluated interventions, is if the studies have a theoretical foundation that may explain the causal link between intervention and outcomes [28]. However, considering the already broad focus of this review, we choose to limit the presentation of results and not include data on the theoretical foundations for each intervention. Furthermore, the quality of the included studies has not been evaluated the same way it would be assessed in a systematic review, meaning that the quality can differ between the studies. It is, however, in line with PRISMA guidelines on scoping reviews considering this step optional [27]. Yet a quality assessment of the included studies or grading of evidence might have led to stronger conclusions as a result of a reduction in uncertainty related to outcome effects.
Finally, this review did not include studies from the rest of the world, albeit such studies could have provided relevant information. The choice to do so was due to the importance of contextual factors concerning complex interventions [124]. Limiting the inclusion of interventions deriving from countries with similar welfare models and cultural context might increase chances of effective implementations of promising interventions. Furthermore, research shows that there is is often a lack of information regarding the influence of the context when conducting and evaluating complex interventions [124]. Thus, more research on the influence of contextual factors in the effectiveness of certain interventions would add to the knowledgebase important for decision-makers.

Conclusions
This scoping review, following the MRC guidelines, provides an overview of the evidence and evidence gaps of health-promoting and preventive intervention studies for community-dwelling older people in Nordic countries hence, of importance for decision-makers, research councils and researchers.
All interventions, besides one, showed positive effects on at least one health outcome, although the magnitude of effects and number of follow-ups differed substantially. Given that evidence on effects alone are not enough information for decision-makers, information on other factors is needed. Overall, there was a general lack of studies related to costeffectiveness, experiences of participants and feasibility. Therefore, such studies are strongly warranted. In all, based on the evidence presented, senior meetings, preventive home visits and exercise interventions alone or combined with other components seem to be strong candidates for implementation in a Swedish municipality context.
Additional file 1. Search strategies and numbers of records identified in each database.
Additional file 2. Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist.