The search in January 2019 yielded a total of 5652 references after removal of duplicates. Of these, we included 98 publications for full text screening. We excluded 72 publications for the following reasons: wrong language (n=1), wrong study design (n=3), wrong publication type (n=11), wrong population (n=8), not focussing on barriers and/or facilitators (n=26), barriers and facilitators not related to a concrete intervention (n=12), no nurse-led intervention (n=9), data collection before implementing the intervention and impossibility to describe actually experienced barriers and facilitators (n=2). Finally, we included 26 studies in our review. Figure 1 shows the search and selection process in detail.
Characteristics of included studies
Table 1 displays characteristics of included studies. Most of the studies were from the Netherlands [18–23] and the UK [24–29] (each n=6), followed by Australia [30–32], Canada [33–35], Norway [36–38] (each n=3), Germany [39, 40], and Belgium [41, 42] (each n=2). One multinational study took place in Italy and the Netherlands [43]. 80% of the studies were published since 2017 [18, 19, 21, 22, 26–39, 41–43] (n=21). Most of the studies were conducted in the long-term care setting [18, 19, 21–28, 30, 31, 34–43] (n=22), two were performed in an acute hospital [29, 32] and one in the outpatient setting [33]. One study had a mixed setting (outpatient and long-term care) [20]. Four studies used the “Promoting Action on Research Implementation in Health Services” (PARiHS) framework as a theoretical framework for implementation [33, 36–38], one study was based on the normalisation process theory [27]. The remaining publications used a conceptual framework for implementation of advance care planning [42] (n=1) and the “COM-B (capability, opportunity and motivation-behaviour) system” [32] (n=1) as a framework. Nineteen studies did not refer to a specific implementation framework.
Data concerning barriers and facilitators were collected through interviews [10, 19–22, 24–40, 42, 43] (n=24), by means of questionnaires [18, 24, 30, 37, 40] (n=6), field notes or process data notes [22, 27, 36, 40] (n=4), observation [25, 29, 39] (n=3), workshops [38] (n=1), written evaluations by trainers/instructors [22] (n=1), residents’ records [40] (n=1) and/or by asking open-ended questions [41] (n=1). Qualitative data were analysed using thematic or content analysis [18–20, 22–25, 27–33, 33–41, 43] (n=23) or framework analysis [26, 27] (n=2). Quantitative data analysis was based on multilevel regression analysis [37] (n=1) or descriptive statistics [40] (n=1). Participants were mostly health professionals on different hierarchical levels (e.g. registered nurses, healthcare assistants), from different disciplines (e.g. nurses, physicians, psychologists) and working on management or clinical level. All were part of the intervention or the implementation (n=23 studies). Family caregivers were asked in three studies [23, 26, 34] and people with dementia in two studies [20, 26]. Three studies included persons facilitating the intervention [21, 29, 30, 38] and one study included volunteers [20]. The number of participants ranged between six and 90. All studies, except for one [32], reported barriers and facilitators.
Barriers and facilitators
We identified five domains of barriers and facilitators: policy, organisation, intervention/implementation, staff and person with dementia/family. Within these domains, we created one to twelve categories describing influencing factors (barriers or facilitators) (table 2). For full information about extracted data, see additional file 2: Summary of included studies.
Policy
The policy domain describes enabling or hindering factors on the governmental or municipal level.
Barriers: Financing issues, e.g. no clear reimbursement for the delivery of an intervention [23] or governmental regulations concerning task-oriented practices [35] are considered as barriers in the implementation process.
Facilitators: The authors of one study [23] mention the organization of health insurance promoting collaboration of dementia care networks and allowing reimbursement of intervention delivery as a facilitating factor.
Organisation
We categorised factors relating to characteristics, structures and processes of an organisation (e.g. nursing home, hospital) in the organisation domain.
Barriers: The category organisational culture and vision summarizes influencing factors. Distinctive hierarchical structures [26, 35, 38–41], inadequate regulations within the organisation [19, 43] and a task-focused, functional culture of care [31, 39, 40] are examples for hindering factors within this category. Furthermore, authors of several studies report a lack of management and leader support and engagement as a hindering factor [19, 22, 26–29, 37, 38, 40]. This was the case when leaders were passive or authoritative or when their role was not clear. Limited resources, e.g. lack of financial support, staff, time, space and material, are described as barriers [18–29, 31–36, 41–43]
High rates of staff turnover and fluctuation [18, 22, 23, 27, 29, 31, 34, 40] as well as demands competing with the intervention [18, 19, 22, 25, 26, 28, 29, 31, 34, 36, 39] are mentioned as further barriers. Competing demands comprise other innovations or organisational changes taking place at the same time as well as multiple and contradictory requirements.
Facilitators: Some authors describe a facilitating organisational culture and vision if a person-centred culture was already established [39, 40] and the hierarchies were flat [30, 33, 39, 40].
Committed and supportive leaders and managers are mentioned as facilitators [18, 19, 22, 26, 27, 29–31, 33–42]. This also applies to adequate resources (staff, time, budget, space) [18–21, 23, 26, 27, 31, 35, 41, 43].
Intervention and implementation
In the intervention and implementation domain, we identified facilitating and hindering factors concerning characteristics of intervention content and delivery as well as features of the implementation process.
Barriers: Within the category perceived value of the intervention a barrier is described if the impact or effect of the intervention is not obvious to staff [22, 23, 27, 36, 38]. Another category mentioned in the studies is sufficiency of intervention training delivery. This is considered a barrier if training was not sufficient, e.g. with regard to staff participation or due to scheduling outside regular working hours [19, 29]. In the category degree of intervention clarity, uncertainties concerning the intervention or the implementation are mentioned as hindering factors [19, 26, 27, 40]. Furthermore, authors describe it as hindering if the intervention was not suitable for current practice, e.g. if it overlapped with current working methods [18, 22, 27]. Hendriks et al. [20] mention environmental conditions and describe bad weather as a barrier to outdoor intervention. The category support from a defined person proves to be hindering if a defined and sensitive person is missing [26, 40]. Thus, lacking qualification and enthusiasm of the supplying person [26–29] are considered as barriers. This is also the case with regard to conditions for the supplying person, e.g. concerning the use of personal resources, overload due to organisational requirements or missing support from staff [26, 28, 38].
Collaboration with stakeholders is described as a barrier if relevant stakeholders did not want to be involved in the intervention [23, 28]. Concerning the category implementation methods, unfamiliar methods, e.g. teleconference supervisions or online communication can hinder the implementation process [24, 33, 38].
We also identified the complexity of the intervention as a barrier. Authors of integrated studies describe time-consuming, complex and expensive interventions as hindering successful implementation [18–20, 23, 26, 27, 30, 33, 43]. Barriers summarized in the category issues concerning trial procedure comprise the organisation of the trial, follow-up or supervision periods and communication problems between the study team and staff in clinical practice [24, 27, 30, 38].
Facilitators: We recognized experience of the value of an intervention as a facilitator. Seeing the positive results of an intervention, e.g. enhanced quality of care and quality of life of the person with dementia, is motivating for staff [19, 22, 25, 26, 31, 35]. Furthermore, sufficiency of intervention training delivery can influence the implementation processes. Authors of various studies describe this category as facilitating if the training followed a practical or interactive approach [19, 29, 43]. Additionally, some authors mention the degree of clarity of the intervention as an influencing factor. They report clear and structured interventions as facilitators [19, 22, 26, 27, 30, 31, 34, 40, 41].
The category suitability for current practice covers enabling factors like applicability of an intervention to daily practice [19] as well as alignment with current organisational structures and procedures [22, 33]. Furthermore, several authors describe support from a defined person as an enabling factor [18, 24, 26–28, 30, 33, 35, 37]. If the supplying person is qualified and enthusiastic [27–29] and the conditions for the supplying person [24, 26, 28] are good, this proved to be facilitating. Examples for good conditions are support from clinical staff and from another supplying person, sufficient time and respect on the part of the organisation. Involvement of staff in intervention development and delivery [33, 38] as well as involvement of multiple disciplines and hierarchical levels are described as facilitators in various studies [18, 19, 21–23, 26, 28, 33, 39–41, 43].
Staff
Barriers or facilitators within the staff domain refer to factors directly related to staff characteristics, qualities and attitudes.
Barriers: It became obvious that certain characteristics of team cultures, e.g. different cultures within a team, inefficient communication, inflexible team members or conflicts within the team can impede the implementation [19, 22, 23, 34, 40]. Moreover, staff knowledge, experience and skills can influence implementation processes. Insufficient dementia-specific or intervention-specific knowledge or missing confidence regarding dementia care are described as barriers [19, 20, 23, 25, 26, 29, 38, 40, 41]. Various authors mention a lack of staff motivation and energy as a barrier [19, 20, 29, 36, 38, 40]. If staff is not committed or displays passive behaviour during the implementation process, this is described as hindering. Limited or unclear responsibilities, e.g. non-transparent intervention-related responsibilities can negatively influence staff motivation and thus adversely affect the implementation process [19, 23, 40, 41]. Familiarity of the intervention also seems to be an influencing factor. Some authors describe it as hindering, if staff is not familiar with the intervention or feels uncomfortable with it [19, 20, 23, 34, 43]. In this context, staff attitude towards the intervention is relevant as well. Negative attitudes, e.g. scepticism, resistance or lack of acceptance regarding the intervention are identified as barriers [19, 24, 26, 27, 29–31, 34, 40]. Furthermore, the focus of care on tasks or high efficiency hinders the implementation of nurse-led interventions in dementia care [19, 36].
Facilitators: Positive team cultures ensuring mutual support and well-functioning collaboration and communication are described as facilitators [19, 20, 23–25, 29–31, 33, 35, 38, 40, 41]. Furthermore, staff knowledge, skills and experience can positively influence the implementation if staff is highly qualified and confident [20, 26, 38, 41, 43]. Staff motivation and openness are also mentioned as influencing factors. Various authors describe motivated and enthusiastic staff with high commitment to change as a beneficial factor [8, 10, 18, 20, 26, 27, 29, 34, 38, 39, 43] .
Person with dementia and family
In this domain, enabling and hindering factors refer to characteristics of the recipients of an intervention, i.e. people with dementia and their relatives.
Barriers: Lacking engagement of families can complicate the implementation of a nurse-led intervention in dementia care. If relatives are not available or if they are not willing to engage in the intervention process, this is mentioned as hindering [19, 29, 34, 43]. Moreover, negative attitudes towards the intervention on the part of family caregivers or other patients are described as barriers [29, 30].
The category nature and stage of dementia includes hindering factors directly relating to characteristics of the person with dementia. Residents’ or patients’ cognition, particularly quickly changing needs and fluctuating behaviours of people with dementia are challenging in the implementation process [19–21, 33, 36, 42, 43]. Furthermore, in one study, lack of background information about the person with dementia is mentioned as a barrier [19].
Facilitators: The engagement of families is described as an important factor. Several authors mention strong relationships with relatives and involvement of relatives as supportive [28, 34, 35, 43]. Positive response of people with dementia and their families, e.g. positive feedback or cheerful reactions are also described as facilitating factors [19, 31, 35]. Moreover, education, knowledge and experience of the person with dementia and the family are mentioned as enabling factors as well [20, 35, 43].