Background

Recent forecasts estimate 152.8 million global cases of dementia by 2050, which will increasingly strain health systems that already struggle to meet current elderly care demands [1]. Recent studies suggest that home- and community-based services (HCBS) for people with dementia (PwD), facilitated with primary support from informal caregivers, present a cost-effective and patient-preferred alternative to institutionalization [2, 3]. Informal caregivers are identified as family members, friends, and neighbors of PwD, and their roles consist of facilitating instrumental activities of daily living, care management, and care continuity [4]. In 2019, the World Health Organization reported an estimate of 133 billion hours of global unpaid informal dementia care [5]. Additionally, Rabarison and colleagues [6] estimated that the 3.2 million informal dementia caregivers, based in North America, included in their review provided unpaid care valued at US $41.5 billion, highlighting the social and economic value of informal care.

To succeed in their role, informal caregivers also require support to reduce personal experiences of stress, anxiety, burnout, and depression, commonly exacerbated by their caregiving demands [7, 8]. Cheng and Zhang [9] produced a meta-review, synthesizing over 500 individual studies on the effectiveness of non-pharmacological evidence-based interventions (EBI) that support informal caregivers of PwD, which revealed EBIs can effectively reduce caregivers’ psychological distress and strengthen dyadic communication and coping skills, improving their overall quality of life [9,10,11,12]. Types of caregiver-focused interventions include psychoeducation, eHealth, support group interventions, case management and care coordination, respite care, and exercise [9]. However, despite the multitude of EBIs that effectively support informal caregivers, the pertinent details surrounding the implementation of these interventions remain unclear.

The effectiveness of EBIs is merely one component that cannot be studied in isolation but must be considered among other contextual variables across multiple levels within the local health system and implementation setting, including clients, providers, organizations, and communities [13, 14]. EBIs must be systematically implemented within HCBS to strengthen caregiver resilience, improve quality of life, and delay institutionalization of PwD [15, 16]. This goal can be actualized by applying implementation science knowledge to steer dementia care research and practice.

Application of implementation theories, models, and frameworks

Implementation theories, models, and frameworks, hereby referred to as frameworks, allow researchers to structurally examine the implementation and sustainment processes and the contextual determinants (i.e., barriers and facilitators) to implementation [17]. The Consolidated Framework for Implementation Science Research (CFIR) is a comprehensive determinant framework that uses a multilevel, multidimensional approach to identify “what works, where, and why”, and the breadth of constructs provides the most coverage to accurately reflect the complex nature of real-world implementation [18,19,20]. The CFIR has been widely applied in both empirical research [21] and in a systematic review [22] to structurally assess the barriers and facilitators to implementation.

In addition, the process of implementation can be systematically studied using the refined Expert Recommendations for Implementing Change (ERIC) taxonomy, which consists of 73 discrete implementation strategies that provide a structured set of “building blocks” used to homogenize implementation reporting and tailor a multicomponent implementation strategy [23]. Waltz and colleagues [24] grouped these strategies into nine clusters and rated each discrete strategy based on its perceived feasibility and importance. Implementation strategies act via mechanisms, which explain how the implementation strategy has an effect by describing the set of strategic actions that occur [25].

The Implementation Outcomes Framework (IOF) can be used to evaluate the degree of implementation success and the effectiveness of selected implementation strategies and to provide important distinction between intervention failure and implementation failure. The IOF explores the acceptability, adoption, appropriateness, feasibility, fidelity, implementation cost, penetration, and sustainability of the EBI [26]. The ERIC taxonomy and the IOF have both been applied to specify and compare implementation strategies and outcomes in empirical implementation research [27, 28] and in recent literature reviews [29,30,31]. The combination of the ERIC taxonomy, IOF, and CFIR allows researchers to comprehensively examine the multiple levels and stages of implementation.

Study aims

Lourida and colleagues [32], and Bennet and colleagues [33], synthesized the implementation literature of EBIs for PwD and, indirectly, their caregivers, and each study determined an urgent need for additional synthesized literature, guided by implementation science frameworks, on the implementation of home- and community-based EBIs that support informal caregivers of PwD. This scoping review combines three implementation science frameworks to create a detailed and systematic synthesis of implementation science literature, to construct a comprehensive understanding of implementation, reflective of multifaceted, real-world complexities. This facilitates the understanding of implementation strategies employed, outcomes reported, and the contextual barriers and facilitators to implementation. Accordingly, this scoping review aims to accomplish the following objectives:

  1. 1)

    Guided by CFIR, map, describe, and synthesize the contextual barriers and facilitators to implementation of EBIs.

  2. 2)

    Guided by the ERIC taxonomy, map, describe, and synthesize the implementation strategies employed to deliver home- and community-based EBI that support informal caregivers of PwD.

  3. 3)

    Guided by the IOF, map, describe, and synthesize the implementation outcomes that have been used to report and measure the success (or failure) of implementation of these EBIs.

Methods

Arksey and O’Malley’s scoping review framework [34] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) reporting recommendations were used to guide this review [35] (see Fig. 1 in Additional file 1. Method Overview). The scoping review protocol for this article [36], published in January 2022, provides a detailed overview of this review’s methodological steps and justifications at each stage; therefore, the methods are summarized in the sections that follow.

Study eligibility criteria

The review included studies that focused on home- and community-based EBIs that support informal caregivers of PwD, which a) explicitly reported the implementation strategies used and implementation outcomes examined and/or b) explicitly reported the barriers and facilitators to implementation of EBIs. Studies were excluded if they examined EBIs that primarily focused on supporting the PwD or were delivered outside of the HCBS settings (e.g., institutionalized care, acute care).

Information source and search strategy

The research team, with support from a specialized medical librarian, developed a full search strategy surrounding four key words: “dementia,” “informal caregivers,” “intervention,” and “implementation and dissemination” (see Additional file 2. Search strategy). Following, literature search was conducted across Embase, MEDLINE (Ovid), Web of Science, and Cochrane Central Register of Controlled trials (Wiley) to include all peer-reviewed studies, written in English, published from inception to 08 March 2021. Critical appraisal of included texts was performed by two reviewers (E. M. Z. and M. B.) using the Mixed-Methods Assessment Tool-version 2018 (MMAT), which is used to appraise the quality of empirical research designs and the comprehensiveness of data reporting [37].

Study selection

In title and abstract screening stage, all relevant publications identified were imported into ASReview (https://asreview.nl/), an artificial-intelligence-aided tool that sequentially presented all imported publications to the reviewer from most to least relevant [38]. Previous studies indicated that ASReview’s algorithm could detect 95% of the final included publications in their study within the first 20% of publications presented, which significantly reduced time spent screening titles and abstracts while effectively maintaining result quality and integrity [39].

The first author (E. M. Z.) programmed the tool by screening 10 randomized (trial) publications and manually screened all imported titles and abstracts to completion. Following, the second author (M. B. S.) only screened the titles and abstracts of studies excluded by the first author to avoid false negatives. Given the tool’s capabilities, the second author stopped screening after 50 successively excluded studies, which was the team’s predetermined terminal point [36]. Following, the full texts of all included publications were assessed by both the first and second reviewers to exclude false positives. Any disagreements between the two authors were resolved by the third (K. A.) and fifth author (R. H.). Lastly, the reference lists of final included studies were checked to detect additional publications.

Data extraction

Data extraction, summarizing, and collating process were conducted by the first and second author using a consensus approach, with regular discussion with all co-authors. A first table, guided by the domains and (sub)constructs of the CFIR, was used to extract and chart the identified barriers and facilitators. A second table was constructed based on the ERIC taxonomy and the nine clusters of implementation strategies reported in the literature. The first author identified detailed actions and mechanisms reported within each study and then “translated” and “matched” each with its corresponding discrete implementation strategies and respective clusters within the ERIC taxonomy. For example, a reported mechanism, such as “provide alternative mode of service delivery,” would “match” the discrete strategy “promote adaptability (ERIC 51)” found in “adapt and tailor to context (Cluster 3).” A third table, guided by the IOF descriptions, was also developed to systematically extract and chart the data for implementation outcomes reported. Prior to data extraction, the first author trialed the three unique data extraction tables on 10 random studies and made iterative refinements to each table after discussion with the research team.

Upon team consensus, the implementation strategies, outcomes, and barriers and facilitators to implementation from included studies were extracted by the first author (E. M. Z.). Categorization and “matching” of extracted data were reviewed for accuracy and confirmed by the second author (M. B. S.); any disagreements between reviewers at this stage were resolved by discussion until consensus was achieved. Additionally, study characteristics, including country of study origin, research design, type of intervention, target population, outcomes reported, and frameworks applied, were also extracted and synthesized. Further details on the methodology can be found in Fig. 2 of Additional file 1.

Results

The full search yielded 2667 de-duplicated publications, 175 full-text publications were assessed for eligibility, and the reference lists of 62 publications were searched for additional relevant literature, which identified five additional publications. Sixty-seven publications were included in the final qualitative synthesis. Using the MMAT-version 2018, 56 of 67 studies were rated 100%, and 11 studies were rated 80%. The study exclusion process can be found in Fig. 1, and details of study characteristics and findings can be found in Table 1, found below, and Table 1 in Additional file 3.

Fig. 1
figure 1

PRISMA diagram illustrates the process used to identify eligible studies

Full size image
Table 1 Overview of results from included studies
Full size table

Study characteristics

The 67 included studies were published between 1996 and 2021; more than half were published between 2016 and 2021 (40/67; 59.7%). These studies reported 58 unique interventions, which were classified into one of eight types of interventions for informal caregivers of PwD based on the most prominent intervention components. This stratification was performed to examine the implementation characteristics of EBIs with clear commonalities to enhance the review’s usability. Multicomponent interventions (e.g., the combined use of case management, support groups, and eHealth tools) (18/67; 26.9%) [84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101] were most common, followed by eHealth (15/67; 22.3%) [40,41,42,43,44,45,46,47,48,49,50,51,52,53,54], psychoeducation (12/67; 17.9%) [60,61,62,63,64,65,66,67,68,69,70,71], care coordination and case management (6/67; 8.9%) [75,76,77,78,79,80], support interventions (5/67; 7.4%) [102,103,104,105,106], respite care (5/67; 7.4%) [55,56,57,58,59] exercise (3/67; 4.4%) [72,73,74], and occupational therapy (3/67; 4.4%) [81,82,83]. Studies originated mostly from the USA (36/67; 53.7%), followed by The Netherlands (11/67; 16.4%), the UK (9/67; 13.4%), Australia (4/67; 5.9%), Portugal (2/67; 2.9%), and India, Israel, Poland, Germany, Canada (each n = 1). The most common study designs were pre-posttest studies (38/67; 56.7%), followed by descriptive qualitative studies (20/67; 29.9%) and parallel convergent mixed-methods design (9/67; 13.4%).

Use of implementation theories, models, and frameworks

Twenty-one articles were explicitly guided by an implementation framework (21/67; 31.34%). Ten unique frameworks were used, including adaptive implementation model [90, 102, 103, 105, 106], multimethod assessment process (MAP)/reflective adaptive process (RAP) [46], reach, efficacy, adoption, implementation, and maintenance (RE-AIM) [83, 98,99,100], Medical Research Council Framework [44, 45, 89], Fixsen and Blasé Implementation Process Model [67, 95], Consolidated Framework for Implementation Research [48], Leontjevas process evaluation model [45, 53], process evaluation model by Reelick and colleagues [74], Lichstein’s treatment implementation model [84], and normalization process theory [88].

Several constructs were frequently included within these frameworks. Intervention characteristics, including quality and validity of evidence, were prevalent considerations made prior to implementation [44, 45, 48, 53, 83, 88,89,90, 98, 100, 102, 103]. All ten frameworks included constructs relating to implementation setting factors, including both internal (e.g., resources) and external (e.g., government policy) to the implementing organization, and the implementation process, including planning, program adoption, implementation execution, and sustainment. Iterative and reflexive monitoring and (re-)evaluating implementation strategies and outcomes were also components of all included frameworks (see Table 2 in Additional file 3 for details).

Table 2 Barriers and facilitators to implementation of EBIs for caregivers of people with dementia, mapped onto the Consolidated Framework for Implementation Research constructs
Full size table

Barriers and facilitators to implementation (CFIR)

The barriers and facilitators to implementation were mapped based on the domains (and constructs) of the CFIR, including intervention characteristics, outer setting and inner setting of the implementing organization (e.g., nursing home), characteristics of individuals, and process of implementation, which allowed for systematic examination of the contextual variables.

Barriers to implementation

Intervention characteristics domain presented barriers to implementation, including lack of relative advantage (4/67; 6%), poor adaptability (12/67; 17.9%), and unsuitable design quality and packaging (25/67; 37.3%). New interventions are hindered by high market saturation and are less likely to penetrate organizations due to the presence of similar “usual care” programs [75, 98, 100, 105]. The EBI user’s poor digital literacy hindered use, as did the interventions’ complicated user interface designs, fragmented information, complex language, and unsuitable components that fit poorly with users’ capabilities [40, 47, 53, 54, 75, 98, 100, 105].

The outer setting domain presented barriers to implementation, including patient needs and resources (24/67; 35.8%), such as implementing agencies’ lack of awareness surrounding influential cultural nuances that deter caregivers from seeking external support (e.g., filial piety) [92, 105], and caregivers’ personal circumstances, including insufficient personal finances, time constraints, poor digital literacy, and adequate information to confidently participate [41, 55, 59, 74, 89, 92, 106]. Additionally, an intervention is less likely to be positively received if introduced to caregivers at an inappropriate stage. For instance, introducing occupational therapy to caregivers immediately following a PwD’s dementia diagnosis creates confusion; alternatively, engaging caregivers in a support program at a later stage in the care trajectory will be less effective since they need communication training and decision-making guidance beginning in early stages [61, 62].

Barriers to implementation under external policy and incentives (15/67; 22.4%) include lack of care coordination and continuity within less developed health systems [77, 79, 103, 106], top-down policies that established unsuitable or limiting funding mechanisms to implement and sustain community-based programs [66], and fragmented care financing that requires caregivers to (re)apply for assistance covered under different legislations [83, 94, 102, 103, 105, 106]. Cosmopolitanism (14/67; 20.9%) also contained barriers to implementation, including the complexities of vast networks that foster misalignments between partnering agencies and obscure respective actors’ roles and responsibilities [95, 99, 102, 105]. Consequently, poorly networked EBI initiators face distrust with implementing agencies, limited regional partnerships, and impeded service referrals and dissemination [77, 79, 102, 103, 105, 106].

Inner setting barriers to implementation are found within implementation agencies (e.g., community nursing homes). Barriers classified under structural characteristics (2/67; 3.0%) and internal network and communications (2/67; 3%) constructs included rigid hierarchal organization structures, inflexible operating budgets, and lack of role clarity and fragmented information transfers between staff members [102, 105, 106]. Tension for change (5/67; 7.5%), compatibility (7/67; 10.45%), and relative priority (2/67; 2.99%) presented barriers, including staff reluctancy toward adopting externally developed interventions and implementing agency’s lack of capacity for and commitment toward promoting new innovations [68, 95, 103, 105]. Leadership engagement (4/67; 6.0%), available resources (15/67; 22.4%), and access to knowledge and information (5/67; 7.5%) presented barriers, including ambiguity surrounding leadership roles [102], inadequate physical and human resources [55, 78, 100], and the absence of implementation guidance and staff training resources [55, 79, 96].

Characteristics of individuals, including caregivers’ and implementors’ knowledge and beliefs about the intervention (5/67; 7.46%), also impeded implementation if they are skeptical about the intervention’s privacy and safety [45, 50, 72, 98]. Caregivers’ and implementors’ self-efficacy (3/67; 4.48%) and individual identification with organization (2/67; 2.99%) impeded implementation if the actors lacked confidence in their roles or if they perceived a misalignment between the organization’s mission and the intervention’s intended outcome [72, 73]. Caregivers’ and implementors’ other personal attributes (15/67; 22.39%), such as a deficit in caregivers’ personal capacity (e.g., financial, and physical capacity, digital literacy) to participate in the intervention [73, 74, 82, 84] or staff members’ lack of social and cultural awareness [59, 92, 98], impeded implementation.

The process of implementation also presented barriers to implementation. Planning (13/67; 19.4%) was hindered by the absence of implementation manuals and fidelity monitoring mechanisms [84, 96], inconsistent and fragmented communication between partnering agencies [43, 78, 103], and poor familiarity with the implementation sites’ contextual nuances [105]. Engaging (13/67; 19.4%) was hindered by ineffective recruitment strategies employed exclusively at the local intervention sites and unanticipated difficulties promoting the intervention and gaining caregivers’ and implementation partners’ acceptance due to a fragmented regional network [48, 68, 74, 90, 98, 103]. Formally appointed implementation leaders (2/67; 3.0%), champions (3/67; 4.5%), and external change agents (2/67; 3%) presented fewer barriers to implementation, but the absence of clear leadership, high staff turnover, and fragmented information across partnering agencies created tension that disrupted all stages of implementation [98, 99, 102]. Executing (7/67; 10.5%) was hindered by high caregiver attrition rate [52, 96] and unexpected organizational changes and diminished capacity [78, 106]. Reflecting and evaluating (3/67; 4.5%) revealed discrepancies between clinical and real-world results, which caused unanticipated implementation barriers that required iterative responses from implementers [95, 98, 106].

Facilitators to implementation

Intervention characteristics that facilitated implementation include the EBI’s relative advantage (10/67; 14.9%), adaptability (19/67; 28.4%), design quality and packaging of intervention components (42/67; 62.7%), and cost (4/67; 6.0%). Advantageous interventions possessed flexible, patient-centered, and culturally adapted programming, and they promoted service continuity through a comprehensive range of integrated services. Adaptable EBIs ensured homogenous participant groups and provided multimodal delivery of intervention components [51, 53, 75, 92, 101, 103]. EBIs were more successfully adopted by end users, if moderated by a human facilitator (e.g., therapist, IT specialist, coach), and by organizations, if implementation is guided by a protocolized implementation guide [42, 43, 46, 51, 52, 61, 66, 68, 71,72,73,74, 82, 92, 93, 96, 101]. Interventions with costs covered through sustainable funding sources (e.g., private foundation or government grants) were more likely to survive [59, 67].

Outer setting domain contained the most reported facilitators to implementation. Patient needs and resources (22/67; 32.8%) included convenient service location equipped with appropriate physical infrastructure and scheduling flexibility [55, 65], sufficient user awareness and preparedness [69, 75, 82], and suitable fit between intervention and users’ levels of digital literacy and needs [40, 42, 43, 52]. Cosmopolitanism (29/67; 43.3%) facilitators included establishing and harnessing strong, active local collaborative networks with dedicated implementation and dissemination partners, including intersectoral organizations (i.e., intermediary organizations) with influence spanning across sectors, whose insights and contributions are valuable across all stages of implementation [47, 57, 66, 67, 75, 85,86,87,88, 91, 102, 105,106,107]. External policy and incentives (20/67; 19.9%) facilitate implementation through the successful funding and reimbursement of intervention costs, delivered through mechanisms established by existing national legislations [59, 67, 76, 90, 94, 101, 102, 106, 107].

Inner setting constructs, including structural characteristics (1/67; 1.5%), network and communications (3/67; 4.5%), and culture (3/67; 4.5%), facilitated implementation through continuous structural financing, regular staff communication and training, and staff enthusiasm about the intervention [90, 98,99,100,101, 105]. Facilitators associated with tension for change (2/67; 3.0%), compatibility (15/67; 22.4%), and learning culture (1/67; 1.5%) included the alignment of the intervention’s intended outcome and implementing agency’s mission, the agency’s willingness and administrative capacity to routinize the intervention as part of usual care (e.g., utilizing existing billing/work codes to receive compensation, integrate EBI into clinical workflow), and the modification of existing staff members’ roles to adopt new interventions [46, 68, 69, 90, 91, 95, 98, 100, 106]. Facilitators under leadership engagement (7/67; 10.5%) included engaging managers that possessed a clear agenda, a creative mindset, and a proactive approach of continuous improvement [48, 67, 78, 95, 102, 106]. Facilitators under available resources (13/67; 19.4%) included motivated, well-trained staff members, accessible and convenient implementation location, and supplemental financial and collaborative support from regional government agencies [43, 48, 55, 59, 67, 98, 100, 105, 106]. Access to knowledge and information (11/67; 16.42%) was facilitated by using a cascade model of training, hiring external training agencies, and requiring protocolized licensure and certification for intervention staff to ensure fidelity and program validity [66, 67, 87, 90, 93, 94, 96, 97, 99, 101].

Characteristics of individuals, including caregivers’ and implementors’ knowledge and beliefs about the intervention (2/67; 3.0%), facilitated implementation if the intervention was developed locally or within the implementing organization [48, 92]. Caregivers’ and implementors’ self-efficacy (8/67; 11.9%) and individual state of change (2/67; 3.0%) facilitated implementation if they possess competencies required to succeed in their roles and are well-equipped with communication and coping skills [40, 45, 61, 62, 67, 81, 95, 98]. Individual identification with organization (3/67; 4.48%) facilitated implementation if the implementation agents identified with the intervention initiators and were enthusiastic about its success [48, 67, 90]. Other personal attributes (10/67; 14.9%), such as staff members’ ability to adapt and cater to caregivers’ iterative needs (e.g., bilingual and technical competencies) and caregivers’ positive attitudes toward participation, also facilitated implementation [40, 57, 66, 82, 89, 90, 92, 98, 102].

The process of implementation was also facilitated by unique contextual factors. Planning (13/67; 19.4%) was facilitated by adapting and translating interventions to fit local implementation setting and co-creating implementation and marketing plans that considered influential contextual nuances [57, 78, 83, 84, 88, 96, 99, 100, 102, 105, 106]. Engaging (21/67; 31.3%) facilitators included the active dissemination of intervention information, by applying marketing strategies to reach specific audiences and disseminating recruitment materials through partners’ networks [40, 47, 51, 53, 57, 66, 72, 76, 78, 87, 90, 92, 94, 99, 100, 102, 103, 105, 106] and the engagement of caregivers through referrals from general practitioners and members of local care organizations [51, 75, 80, 98, 99]. Additionally, opinion leaders (2/67; 3.0%), formally appointed internal implementation leaders (8/67; 11.9%), champions (7/67; 10.5%), and external change agents (11/67; 16.4%) facilitated implementation by engaging local influential religious leaders to support normalizing the use of new interventions [78, 92], by leveraging individual strengths from external agencies to establish a multidisciplinary advisory team [47, 87, 98, 99, 106], and by appointing a leader to guide implementation and sustainment [58, 75, 76, 78, 102, 103, 105, 106]. For example, faith-based organizations may influence public perception and approval of interventions; academic partners support recruitment and registration of new participants [92], and intermediary organizations (e.g., Alzheimer’s Association) inform regional partners and support in facilitating knowledge transfer. Executing (14/67; 20.9%) and reflecting and evaluating (8/67; 11.9%) facilitated implementation through regular monitoring and evaluation, securing partnerships through formal agreements (e.g., Memorandum of understanding), and iteratively adapting operational processes to meet real-world demands and unanticipated complications. Table 2, found below, and Tables 3 and 4 in Additional file 3, provide further details found surrounding barriers and facilitators to implementation.

Table 3 Implementation strategies and mechanisms reported
Full size table

Implementation and dissemination strategies (ERIC taxonomy)

Of the 67 included studies, 61 studies reported details on the implementation strategies employed to support the delivery of the chosen EBI for caregivers of PwD. Sixty-eight of the 73 ERIC taxonomy’s discrete strategies, across all nine clusters, were identified (see Table 5 in Additional file 3 for details); six discrete strategies (ERIC 45, 50, 68, 3, 28, 10) were not reported by any included study. Multicomponent interventions employed the widest range of discrete strategies (58/73; 79.5%), followed by psychoeducation interventions (48/73; 65.8%), and care coordination and case management (40/73; 54.8%). The most frequently identified discrete strategies were found in the “Train and educate stakeholders” cluster. Mechanisms found within this cluster included training through multimodal delivery, including delivering education and information through an Internet platform equipped with real-time feedback from trainers via a toll-free telephone line [40, 47, 53, 73, 88, 91, 95, 98]. The “Provide interactive assistance” cluster also contained frequently employed discrete strategies; mechanisms identified included providing tailored, individualized feedback to end users [54, 66, 80], facilitating flexible scheduling for end users [57, 65, 72, 80, 98], and enhancing the connectivity and reflexivity between referrers and services [47, 66, 67, 75, 76, 87, 88]. Further implementation strategies and mechanisms are included in Table 3 found below, and more detailed mechanisms and actions can be found in Table 6 of Additional file 3.

Several discrete strategies within the same cluster were also frequently employed together. In the “Develop stakeholder interrelationship” cluster, “Build a coalition” and “Obtain formal commitments” (9/67; 13.4%) were employed together across six studies [66, 78, 85, 100, 102, 105]. In the “Train and educate stakeholders” cluster, “Develop educational materials” (27/67; 40.3%), “Make training dynamic” (34/67; 50.7%), and “Distribute educational materials” (31/67; 46.3%) were employed together in 15 studies [47, 48, 51, 52, 58, 63, 64, 68, 69, 80, 84, 86, 88, 93, 95]. In the “Adapt and tailor to context” cluster, “Tailor strategies” (26/67; 38.8%) and “Promote adaptability” (27/67; 40.3%) were employed together in 18 studies [40, 43, 47, 51, 53, 61, 67, 72, 74, 80, 82, 84, 85, 90, 91, 93, 100, 104].

Eighteen of 67 studies [58, 67, 74, 83,84,85,86, 88, 91, 95, 98,99,100,101,102,103, 105, 106] conducted initial assessments of contextual determinants and, based on these, adapted the implementation strategies to target the barriers and improve the translation of the EBI into local practice. Adaptations made to enhance feasibility due to local constraints (i.e. available financial resources, compliance with local insurance reimbursement regulations) include reducing the frequency of intervention delivery [74, 83, 85, 98] and adapting the professional profile of the EBI provider to fit the available local human resources [91, 99, 101, 102]. Other challenges included the need to adapt the language used to suit users’ capabilities [84, 101] and the location, medium, and format used to deliver the EBI [85, 100, 105]. However, none of the studies was explicit about the mechanism of each adaptation nor did they report a formal evaluation of the impact the adaptation had on the effect of the selected strategies on implementation outcomes, which may indicate a lower degree of maturity of implementation science application in this area.

Implementation outcomes (Implementation Outcomes Framework)

The IOF presents an implementation outcome taxonomy, including acceptability, adoption, appropriateness, costs, feasibility, fidelity, penetration, and sustainability [26]. Appropriateness (49/67; 73.1%) was reported as the intervention’s “suitability,” “usability,” and “helpfulness” for users, and it is “fit into existing workflow” within implementation agencies [48]; evaluative indicators included respondents’ rating of perceived “helpfulness” and their “intention to use.” Acceptability (55/67; 82.1%) was reported as the end users’ and implementing agencies’ “satisfaction” with intervention effectiveness and components, including delivery modality, timing of intervention, duration of program, and quality of interventionist [44, 45, 49].

Penetration (52/67; 77.6%) was only reported in relation to the wider implementation setting; studies mainly descriptively reported how users were recruited, including marketing strategies, and leveraging financial resources and interpersonal relationships from cross-sector partners [47, 51, 63, 68, 70, 75, 77, 82, 86, 87, 92]. Sustainability (40/67; 59.7%) was described as users’ and organizations’ “demand for program continuation” and “routinization of care.” Studies mainly focused on describing the existing internal and external financing mechanisms and the role of collaborators and external agencies in training and scaling up [44, 59, 66, 76, 83, 86, 87, 100, 103].

Implementation fidelity (14/67; 20.9%) was characterized as the facilitators’ degree of “adherence” to the implementation protocol and was explicitly reported through fidelity enhancing, measuring, and monitoring mechanisms. Implementation fidelity enhancing strategies included protocolizing implementation [58, 63, 93, 97], training certification programs with initiators [58, 63, 68, 88, 90, 93, 97,98,99,100], and using fidelity checklists and guiding scripts [68, 95, 99]. Fidelity measuring and monitoring strategies included the use of delivery assessment forms and checklists [58, 83, 88, 99] and ongoing coaching and consultation with initiators [58, 65, 68, 88, 97,98,99].

Adoption (18/67; 26.9%) was reported as how administrations are motivated to “buy into” the intervention and how the engagement of local “influencers” promotes user uptake [92, 95, 101, 105]. Feasibility (18/67; 26.9%) was reported as the degree to which intervention components fit within the organization; for instance, components tested in the RCTs (e.g., fidelity monitoring mechanisms [i.e., surveillance records]) were not pragmatic, or practices could not be easily streamlined into existing workflow [54, 84]. Implementation cost (9/67; 13.4%) was mainly reported as how operational and staffing costs were covered, mainly though government-regulated financing programs (e.g., Medicare, Social Support Act, Older Americans Act) [58, 59, 67, 76, 83, 87]. Implementation outcome details can be found in Table 7 of Additional file 3.

Studies did not evaluate the relationship between implementation strategies and implementation outcomes, but several descriptive trends were identified across included studies. Facilitation (ERIC 33) was employed in 23 of 55 studies that reported on acceptability. Using train-the-trainer strategies (ERIC 71) influenced implementation fidelity in 11 of the 14 studies that reported on fidelity and 23 of 40 studies that reported on sustainability. Mass media (ERIC 69) were employed in all studies that reported on penetration (see Table 8 of Additional file 3 for details).

Discussion

To our knowledge, this is the first review to be guided by three unique implementation science frameworks to study barriers and facilitators to implementation, implementation strategies, and implementation outcomes found in literature relating to EBIs for informal caregivers of PwD.

Applying multiple frameworks allows researchers to examine the various components across implementation processes to potentially establish links between contextual determinants, implementation strategies, and implementation outcomes [108]. Through this methodological approach, our findings illuminate the achievements and gaps in theory-informed implementation thinking in modern dementia care, and they highlight contextual factors that influence successful implementation of EBIs of importance to informal caregivers of PwD.

The MMAT rating results indicated that included studies were of high quality overall, but the appraisal criteria did not assess the quality of implementation reporting nor the rigor of evaluative implementation research designs, suggesting that more suitable appraisal tools are essential to ensure high-quality implementation research [109]. Only 21 out of 67 included studies were guided by an implementation science framework, indicating a need to reinforce the application of implementation science in dementia care research. Furthermore, this review also found that the mean importance and feasibility ratings for discrete strategies, as determined by Waltz and colleagues [24], did not reflect the frequency of implementation strategies used in the real-world implementation of EBIs in home- and community-based services (HCBS). For example, the discrete strategy “use mass media,” employed by 12 of 67 studies, and “use train-the-trainer strategies,” employed by 26 of 67 studies, were both labeled in the original study as low feasibility and low importance, revealing the potential lack of suitability and relevance of existing ratings in HCBS contexts. These results call for an extension of the ERIC taxonomy, or the development of an entirely new framework, with insights from real-world community practitioners with implementation experience, as proposed by Balis and associates [110].

Included studies were also not explicit about implementation strategy mechanisms and did not evaluate implementation strategy effectiveness, nor the degree of influence on implementation outcomes, potentially due to shortage of funding for types II and III implementation-effectiveness hybrid study design prior to 2020 [111, 112]. Only one study in this review reported the rationale for the use of an implementation-effectiveness hybrid design [88] — overall, a direct link (statistical or otherwise) between the implementation strategy selected and implementation outcomes assessed could not be established or evaluated formally in this review. Furthermore, 18 included studies seemed to have adapted their implementation strategies to target barriers and enhance the translation of EBIs to fit their context, but these studies did not directly evaluate the degree of alignment between the barriers and adapted strategies, nor did they propose evaluative methods, which may suggest low maturity of implementation science application in dementia care research.

Similar to the challenges mentioned by Lengnick-Hall and colleagues [113], implementation outcomes were also inconsistently reported, and authors were not explicit about the level of analysis (i.e., individual or organizational level). Delineation is critical to determine casual mechanisms and evaluate implementation strategy effectiveness, particularly when reporting fidelity as an outcome, as authors often referred to both end-user adherence to intervention protocol and facilitator adherence to implementation protocol. The Outcomes Addendum to the CFIR can be used to support researchers in delineating the level of measurement to improve the reporting and synthesizing of contextual determinants [114].

Relating to the barriers and facilitators to implementation, the modifiable intervention characteristics, primarily design quality and packaging, should be strategically and iteratively adapted through feedback from end users to fit the implementation context. In accordance with Lundmark and colleagues [115], this review concluded that consideration of inner and outer setting determinants is also central to ensure alignment between the intervention, the implementing agency’s mission and structural capacity, and sociocultural needs and preferences in the local community [51, 53, 75, 92, 101, 103]. In the outer setting domain, cosmopolitanism included the relationship dynamics between the implementing agency, cross-sector stakeholders, and researchers in academic institutions (e.g., community-academic partnerships [116] and public–private partnerships [83]). The findings suggest for the description of cosmopolitanism to distinguish between multi-level, cross-sector partnerships to focus resources and expertise more effectively, which aligns with the recommendation of Proctor and colleagues [117] to leverage the individual strengths of each partner and co-develop toolkits to facilitate evidence dissemination and EBI implementation. These complex networks facilitate multiple stages of implementation, but further implementation research supported by experiential knowledge from implementation support practitioners is required to systematically examine processes of collaboration, including each partner’s role in knowledge translation, knowledge brokering, and EBI sustainment and scale-up, to advance implementation theory [118,119,120].

Recent developments

To ensure the relevance of the results, an updated search was conducted in August 2023 using the original search terms. Only ten of the 1186 results published after March 2021 fitted the inclusion criteria, and these studies primarily focused on the early-stage adaptation and implementation of three EBIs, iSupport [121,122,123,124,125,126], Reducing Disability in Alzheimer’s Disease (RDAD) program [127, 128], and STrAtegies for RelaTives (START) [129, 130], which have been previously included in the results (see Table 1). The new articles indicated progress in enhancing real-world applicability but did not yield any new barriers or facilitators (as summarized in Table 2). Implementation and adaptation processes were guided by the i-PARIHS framework [129], ecological validity framework [123], WHO iSupport Adaptation and Implementation Guidelines [121, 122, 124,125,126], and EBI adaptation guide by Escoffery and colleagues [128, 131]. Trends in recent publications suggest that implementation science in dementia care research is slowly progressing, mainly with implementation and adaptation guidance from the World Health Organization and through international collaboration. Overall, there has been little significant progress made in recent years, and the results from this review remain representative of current literature.

Limitations

This review has several limitations. First, the synthesized results did not include studies published after March 2021, which may have excluded implementation details from recent publications. Next, the ERIC taxonomy has limitations since it was developed exclusively through insights from hospital-based clinicians, and implementation strategies employed at the community setting may not be clearly presented in the taxonomy, which potentially limited the reviewer’s ability to optimally extract and match reported strategies from the literature. The review proposes a call to action for the implementation science community to systematically develop a new taxonomy more appropriate for use in the community setting. Additionally, since the search strategy was also developed with guidance from existing implementation science research largely conducted outside of the community setting, more suitable terminology may have been missed, which may exclude relevant articles. Next, although the validity of ASReview tool has been studied [39], there is currently no evidence-based terminal point for article screening by the second reviewer using ASReview, potentially (although unlikely) excluding relevant records. Lastly, due to the poor utilization of suitable implementation reporting guidelines by included studies, the review results were unable to present clear connections between implementation determinants, strategies, and outcomes.

Future directions and recommendations

The main findings from this scoping review indicate a growing demand for systematic implementation and dissemination of EBI for caregivers of PwD. Further research to develop implementation frameworks that systematically guide implementation processes and address contextual barriers involved in community-based implementation of non-pharmacological EBI is needed. For example, the Community-Academic Aging Research Network’s pipeline for dissemination [116] provides a framework, inclusive of community, academic, and intermediary stakeholder perspectives, to create a contextually suitable implementation plan and to leverage cross-sectoral partnerships that facilitate EBI implementation and continuation.

Future research in this area would benefit from employing more rigorous evaluative methodology, and future reviews may perform meta-analyses to further evaluate the impact of implementation strategies on implementation outcomes. Lastly, scoping reviews focused on implementation literature often report limitations due to heterogenous implementation reporting [132, 133]. Therefore, promoting the use of standardized implementation reporting guidelines (e.g., STaRI [134]) in future studies will enable reviewers produce more clear, consistent, and reliable results.

Conclusion

The novel combination of three implementation frameworks in the context of evidenced interventions to support informal caregivers of PwD has offered a first analysis of the implementation strategies and mechanisms applied to actualize implementation and the multi-level implementation barriers and facilitators that directly impact implementation success (or otherwise) of these interventions. This review provides a systematic overview that can be used as a foundation to inform and guide implementation researchers to structurally examine outer setting facilitators and implementation strategies, at multiple levels and across sectors, and can guide implementation agents to strategically leverage existing resources and regional networks to streamline local implementation. Mapping local evidence ecosystems will facilitate more structured implementation planning and support for HCBS interventions, and new evidence will also contribute to strengthening implementation science theory and application in dementia care.