Barriers and facilitators to implementing evidence-based guidelines in long-term care: a qualitative evidence synthesis

Evidence-based guidelines summarize the best available research on health care practices to enhance the provision of consistent and appropriate care [1]. However, bringing evidence into clinical practice is an ongoing challenge. Systematic reviews on guideline adherence and utilization found that a large percentage of available guidelines do not have sustained implementation where appropriate [2, 3]. For example, an organization may implement a new guideline into practice, but the behaviours associated with it do not continue after initial introduction. In contrast, if new evidence emerges, suggesting current practices are not effective, they must be de-adopted. Guideline implementation into routine healthcare can be unpredictable, and trial-and-error approaches have been costly and ineffective, producing variable results of guideline dissemination and implementation [4, 5]. Consequently, there has been increasing interest in employing theories, models, and frameworks to direct guideline implementation. Knowledge translation focuses on developing ways to efficiently and effectively translate evidence-based knowledge into clinical care. Theory-based guideline implementation is desirable as it ensures the implementation plan and processes consider complex factors that influence success of guideline uptake prior to implementation. In this way, implementers navigate around potential pitfalls to successful implementation by conscientiously accounting for previously identified factors which could hinder their success.

Many existing knowledge translation frameworks guide researchers to consider complex factors that influence the success of guideline uptake prior to the implementation process [6‐8]. The Behaviour Change Wheel is one framework that prompts users to select knowledge translation interventions based on physical, social, psychological, and environmental factors that influence the capability, opportunity, and motivation needed for behaviour change (COM-B) [7]. Central to the Behaviour Change Wheel, the COM-B system incorporates Capability, Opportunity, and Motivation as sources of Behaviour. Users can determine what needs to change for the desired behaviour (e.g., guideline implementation) to occur by identifying barriers and facilitators and mapping them onto the COM-B system. The Behaviour Change Wheel then guides users to select potential knowledge translation interventions based on their COM-B analysis [7]. Therefore, by studying barriers and facilitators in a context-specific environment, interventions can be designed in a theory-informed manner which increases the potential for sustainable practice change.

The need to effectively translate evidence-based guidelines into practice is especially pressing for older adults [9] as the proportion of the population aged 65 years and over is growing exponentially [10]. Older adults with complex needs and comorbidities often live in long-term care (LTC) homes, which are living spaces for adults who have significant health challenges to receive access to 24-h nursing and personal care [11]. Guidelines have been developed for various health conditions in LTC homes ranging from diabetes to pressure ulcer prevention [12]. However, most knowledge translation studies on guideline implementation for older adults do not include LTC homes [13]. Knowledge translation strategies from other settings are poorly transferable to LTC because of the skill mix of the staff, environment, complexity of the residents’ conditions, and availability of resources [14]. Knowledge translation strategies must be specifically designed for LTC given the unique context of health care provision in this setting. While barriers and facilitators to guideline implementation have been systematically reviewed in other healthcare settings [13, 15], no such analyses have been conducted for the LTC sector.

The findings of our study will synthesize barriers and facilitators to evidence-based guideline implementation across health conditions in LTC and mapped onto the COM-B components. Our identified barriers and facilitators and suggested knowledge translation strategies based on the COM-B mapping can be used to design theory-guided knowledge translation interventions in LTC. This will save time, effort, and resources in identifying barriers and facilitators so that planners can design interventions more quickly and efficiently. Further, our review will identify gaps in research related to evidence-based guideline implementation in LTC and make suggestions for future work.

The objectives of this qualitative evidence synthesis are to (1) synthesize barriers and facilitators that LTC staff experience during the implementation of evidence-based guidelines and (2) map the identified barriers and facilitators to the central component of the Behaviour Change Wheel framework to inform future theory-guided knowledge translation intervention development in the LTC setting. Our research question is “What are the barriers and facilitators to implementing evidence-based health care guidelines in LTC homes from the perspectives of staff (e.g., nurses, health care aides, physicians)?” The phenomena of interest is implementation of health care guidelines into practice and the factors that hinder or facilitate implementation. The context is LTC homes who provide 24-h nursing care for mostly frail, medically complex older adults across the world in the 21st century.

Two team members (CM and YB) independently extracted and charted the following data in duplicate using a pilot-tested data extraction form: study description (title, author, country, province/state/region, design, objectives, data collection methods, data analysis methods, name of guidelines examined, health topic of guideline examined, behaviour change framework, model, or theory used), individual participant description (profession(s), number, mean age, sex, sampling technique, response rate), LTC home description (number, size, ownership, rurality), and results/findings (identified barriers and facilitators). Data for the study results were extracted verbatim from the text under the heading “results” or “findings” where authors identified barriers and facilitators (or a synonym, e.g., challenges or supports for change) to implementation of the guidelines examined.

The validity, robustness, and applicability of each included study was appraised by two team members (CM and PH) independently and in duplicate using the Critical Appraisal Skills Programme (CASP) Checklist [18]. Consensus between the two reviewers was required, and any discrepancies were adjudicated by a third party. No studies were weighted or excluded based on the appraisal results.

Our synthesis follows the three-stage Thomas and Harden approach to inductive thematic synthesis [19]. We completed two steps of this process, as our primary aim was to produce descriptive themes of barriers and facilitators to guideline implementation across different health guidelines to then map on the COM-B components. After extracting the reported barriers and facilitators, two team members (CM and YB) created a codebook that was grouped into recurrent themes (e.g., resources, staffing issues). The two team members then independently and in duplicate coded each extracted barrier and facilitator with the themes from the code book. If new codes emerged, they were added iteratively to the code book and the barriers and facilitators were re-themed accordingly. The frequency of the themes was tallied as the number of times the theme was mentioned across the included articles. Finally, the themes were mapped onto the COM-B components of the Behaviour Change Wheel by the two team members independently and in duplicate. Based on a synthesis of 19 previously published behaviour change frameworks, the Behaviour Change Wheel provides tables that link the central COM-B components to potential knowledge translation intervention functions based on their expected effectiveness in relation to the barriers and facilitators. For example, if physical opportunity is a barrier, then training, restriction, environmental restructuring, and enablement are potential intervention functions. Potential knowledge translation intervention functions were listed with their associated barriers and facilitators and COM-B components. Any discrepancies between the two members were resolved by a third party. All data analysis and synthesis were performed in Microsoft Excel. Table 1 provide definitions for the COM-B components and knowledge translation intervention functions as outlined by the Behaviour Change Wheel.

Two review authors (CM and PH) assessed the level of confidence for each finding using the GRADE-CERQual [20]. GRADE-CERQual assesses confidence in the evidence based on four key components: methodological limitations of included studies, coherence of the review findings, adequacy of the data contributing to a review finding, and relevance of the included studies to the review question. After assessing each of the four components, we made a judgement about the overall confidence in the evidence supporting the review finding and report it as high, moderate, low, or very low. The final assessment was based on consensus among the two review authors. All findings started as high confidence and were graded down if there were important concerns regarding any of the GRADE-CERQual components.

We present summaries of the findings and our assessments of confidence in these findings in the Summary of qualitative findings table (Table 3). We present detailed descriptions of our confidence assessment in an Evidence Profile (Additional file 3).

The authors of this article are a multidisciplinary group of researchers and clinicians focused on geriatrics and improving care provision in LTC. They have engaged in several research studies in LTC including assessment of barriers and facilitators to implementation of practices, development of guidelines, knowledge translation, and randomized controlled trials. Since we have prior experience assessing barriers and facilitators in the LTC setting, some biases may exist as we may have preconceived ideas of what barriers and facilitators exist. Included studies that were conducted by one of the authors of the current paper were analyzed by two team members who were not authors of the included studies.

After screening 2680 articles, 33 that were published between 2004 and 2020 were included in the analyses (Fig. 1).

Most studies were conducted in Canada and Australia, with much fewer in the Netherlands, the USA, England, Sweden, Germany, South Korea, and Belgium (Table 2). A wide range of guidelines were examined, with the most frequent being oral health, medication reviews, and pain protocols. A variety of study designs were employed including qualitative studies, mixed method, multiple case studies, and process evaluations. Focus groups, interviews, and document analysis were the most frequent data collection methods, and thematic or content analysis was used to analyze data for 73% of included studies. Only six studies used a behaviour change framework, model, or theory to guide their work which included the framework developed by Greenhalgh et al. (Capability, Opportunity, and Motivation), Organizational Readiness for Change, Theoretical Domains Framework, Organization Learning Theory, Promoting Action in on Research Implementation in Health Services, and Normalization Process Theory.

Included studies recruited 12 to 500 LTC home staff from a variety of professions including nursing, medicine, management, rehabilitation (e.g., physical and occupational therapy), pharmacy, and food services (Table 3). Many studies did not report the age or sex of their participants. For those that did, the mean age of included staff ranged from 38 to 54 years, and the percentage of participants who were female ranged from 46% to 100%. Convenience and purposeful sampling were the most common methods of recruitment. At the LTC home level, the number of homes included ranged from 2 to 120, and the number of residents per home ranged from 40 to 251; though many studies did not report these values (11% did not report number of homes, 46% did not report number of residents per home). Similarly, more than half (58%) of the included studies did not report the ownership or rurality of the included homes.

Most studies had a clear research aim which was appropriately addressed through a qualitative research design. Likewise, most studies employed appropriate recruitment strategies and data were collected in a way that addressed the research question. In some studies, the description of data analysis techniques was limited. Overall, we found poor reporting of research reflexivity across most of the included studies. Details of the assessments of methodological limitations for individual studies are found in Additional file 4.

We had moderate or high confidence in all but one of our review findings. Confidence was most often downgraded due to concerns with methodological limitations including a lack of discussion about credibility of qualitative findings and a lack of reflexivity. The data was almost always relevant as most studies examined our phenomena and population of interest. The full CERQual evidence profile can be found in Additional file 3.

The line-by-line thematic analysis of barriers and facilitators is found in Additional file 5. Table 4 provides a summary of the identified barrier and facilitator themes, their definitions and frequency, the articles contributing to the theme, and the CERQual assessment and explanation of confidence in the findings. The most frequently identified barriers and facilitators were consistent across guideline topics, while others were more specific to the content of the guideline. For example, nearly all articles identified time constraints and inadequate staffing (high confidence), and cost and lack of resources (high confidence) as barriers. However, guideline impracticality (high confidence) and taking a reactive approach (moderate confidence) were only identified in articles that discussed physical activity, influenza immunization, pneumonia treatment, and heart failure. In some instances, barriers and facilitators were opposites of each other, with barriers being actual and facilitators being perceived. For example, if time and money were an identified barrier, the staff perceived they could more easily implement the guideline with more time and resources (facilitator). However, some facilitators were also actual. For example, champions to promote implementation of the guidelines within the home was an actual facilitator in several articles.

Physical and social opportunity were the COM-B components that the identified barriers and facilitators mapped onto most frequently (Table 5). Within physical and social opportunity, time constraints and inadequate staffing (high confidence), cost and lack of resources (high confidence), and lack of teamwork (high confidence) and organizational support (high confidence) were frequently reported barriers, while leadership and champions (high confidence), well designed strategies, protocols, and resources (high confidence), and adequate services, resources and time (high confidence) were frequent facilitators. Training, restriction, environmental restructuring, modelling, and enablement are knowledge translation intervention functions suggested by the Behaviour Change Wheel to overcome barriers associated with physical and social opportunity. The COM-B component of psychological capability represented knowledge gaps (high confidence) as a barrier and adequate knowledge and education (high confidence) as a facilitator. Education, training, environmental restructuring, modeling, and enablement are knowledge translation intervention functions suggested by the Behaviour Change Wheel to overcome barriers associated with psychological capability. Finally, reflective and automatic motivation had barriers relating to conflict with clinical autonomy (high confidence), beliefs against the guideline (high confidence), moral distress (moderate confidence), reluctance to change (high confidence), emotional responses to work and confidence in skills (moderate confidence), and change fatigue (moderate confidence). Facilitators with respect to reflective and automatic motivation were having noticeable outcomes occur from guidelines implementation (moderate confidence), a sense of conviction that the guidelines are evidence-based and will demonstrate improvement (low confidence), and a positive emotional response to work and the intervention (high confidence). The Behaviour Change Wheel suggests training, education, persuasion, modelling, enablement, incentivization, coercion, and environmental restructuring as potential knowledge translation interventions to overcome automatic and reflective motivation.

We previously described our initial positioning earlier (see review author reflexivity above). Throughout the review, our positioning remained the same. During analysis and writing of the discussion, we felt our findings confirmed our initial ideas about the most frequent barriers and facilitators.

We systematically identified barriers and facilitators to implementing evidence-based guidelines in LTC and used behaviour change theory to link them to candidate knowledge translation functions. Across several guideline topics, time constraints and inadequate staffing, cost and lack of resources, knowledge gaps, and lack of teamwork and organizational support were frequently identified barriers. In contrast, leadership and champions, well-designed strategies, protocols, and resources, and adequate services, resources and time were frequently identified as facilitators. Linking to the central components of the Behaviour Change Wheel suggests physical and social opportunities and psychological capability are common targets for change to overcome barriers and leverage facilitators. While the most frequently identified barriers and facilitators appear to be universal regardless of guideline topics (e.g., pain, mood, physical activity, heart failure), some guidelines may have nuanced actions that have unique barriers and facilitators. We suggest that future knowledge translation and implementation science researchers assume the most frequently identified barriers and facilitators in our review are present and that they design strategies targeted at physical and social opportunity and psychological capability. A further analysis of barriers and facilitators may be necessary if the actions outlined by the guideline have unique features that could create additional barriers and facilitators.

The reported barriers and facilitators in our qualitative systematic review most frequently mapped onto the central Behaviour Change Wheel components physical and social opportunity: the opportunities afforded by the environment (e.g., time, resources, locations, cues, physical affordances) and interpersonal influences (e.g., social cues and cultural norms that influence the way we think about things). The findings that environmental opportunities (e.g., changing the social and physical context of care provision) are significant barriers to implementing evidence-based guidelines echo recent concerns surrounding quality of care provided in LTC highlighted by the COVID-19 pandemic [21] and is consistent with previous literature. Indeed, there have been recurrent reports of lack of funding and subsequent personnel shortages leading to decreased time to provide services to increasingly complex residents in LTC [22, 23]. Limited teamwork has also previously been identified as a barrier in LTC [24]. Linkage within the Behaviour Change Wheel suggests that training, restriction, environmental restructuring, enablement, and modelling are candidate knowledge translation intervention functions to overcome the identified barriers and leverage the facilitators.

Given the recent international interest in improving LTC during and after the COVID-19 pandemic and the subsequent impetus to support significant changes to the sector [21, 25], several of the Behaviour Change Wheel identified intervention functions could be feasible. For example, environmental restructuring involves changing the physical or social context to support guideline implementation. Resident-centred care approaches restructure the environment of care provision around the resident and address several of the barriers and facilitators identified in our review. For example, one such evidence-based approach, Neighbourhood Team Development, focuses on modifying the physical LTC environment, reorganizing delivery of care services, and aligning team members (e.g., LTC staff, family, residents) to collaborate in providing care [26]. Several of the studies included in our review also identified involving residents and family members as a facilitator of implementing evidence-based guidelines, supporting a resident-centred care approach.

Knowledge gaps pertaining to the information within guidelines, change fatigue, and lack of interest in work were frequently identified barriers and facilitators in our systematic review, which mapped onto the COM-B domains of psychological capability and reflective and automatic motivation. In many countries, most direct care within LTC homes is provided by care aides (e.g., personal support workers, health care aides, continuing care assistants, resident assistants) [27, 28] who often have the lowest level of education, receive the lowest financial compensation, have the least autonomy, and experience work-related burnout and poor job satisfaction [27, 29]. Knowledge gaps also apply to other members of the LTC interprofessional teams including licensed nurses, physicians, pharmacists, and rehabilitation and recreation and leisure providers. Indeed, several of the studies included in our review revealed knowledge gaps for different members of the LTC team. Education and training are potential knowledge translation intervention functions to overcome barriers associated with psychological capability and reflective and automatic motivation. Training for care aides is variable within and between countries. For example, in Canada, there are currently no national education standards for care aides working in LTC, and training varies widely between provinces [30]. Training of other members of the interprofessional team (e.g., physicians, physical therapists) often does not include a focus on geriatrics or LTC, nor is it standardized. Indeed, the COVID-19 pandemic revealed a major gap in standardized training for all team members about proper personal protective equipment use and conservation [31]. Consistent education and training with monitored national standards for all LTC staff may be one targeted knowledge translation strategy. However, for continuing education to be effective in LTC, it must be supported by the organization, and ongoing expert support is needed to enable and reinforce learning [32] which further bolsters the argument for a team-based, resident-centred approach.

This is the first study to synthesize barriers and facilitators to guideline implementation in LTC from the perspectives of staff across healthcare conditions. Barriers and facilitators to guideline implementation have been systematically reviewed in other healthcare settings, but until now, no syntheses have been developed for the LTC context. Further, we not only identified the barriers and facilitators but also mapped them onto the central constructs of the Behaviour Change Wheel. This helps us explore the reasons why the factors identified are barriers and facilitators and the findings can be used to inform the development of future theory-guided knowledge translation intervention development.

From a methodological point of view, the studies included in our review had several limitations. First, studies often did not report important information about the LTC home(s) which provides context from which the results were derived, such as the size, ownership, and rurality of the LTC home. The context of the LTC home including number of residents in a home, funding structure, and access to resources has been previously shown to affect implementation of best practice guidelines in LTC [14]. Future authors of LTC research are encouraged to fully describe the setting so that readers can adequately assess the generalizability of the results to their context, or reasons why they may experience different outcomes. Further, authors should include a fulsome description of the context including care philosophy of the home, staffing levels, and health system influences (e.g., public or private funding). Second, most authors did not critically examine their own role, potential bias, and influence during analysis and presentation of results. Reflexivity, or the acknowledgement of underlying beliefs and values held by researcher in selecting and justifying their methodological approach [33], is essential in assessing the authenticity of qualitative results [34]. Authors of qualitative research are encouraged to include a reflexive statement when reporting their results that describes their role in data collection, analysis and interpretation, and potential resulting biases that may arise.

A strength of our study is that we synthesized information across different health conditions within the LTC sector. Given that there are likely many similarities among barriers and facilitators across guidelines for different conditions in the LTC setting, the findings of this qualitative evidence synthesis can help inform the implementation of any evidence-based guideline in LTC homes. However, a limitation of our study is that we did not assess the strength of the barriers and facilitators identified in this review. A frequently identified barrier may not hinder implementation as much as one that is less frequently reported. We argue that frequently reported barriers across several guideline topics are nonetheless important to identify as they can inform design of knowledge translation strategies regardless of topic. Future work should examine the strength of barriers and facilitators in LTC for implementing evidence-based guidelines and determine which barriers significantly limit implementation to add to our work. Another limitation is that we did not complete the third stage of the Thomas and Harden approach to thematic synthesis [19] to develop analytical themes that enable the development of new theoretical insights and findings not seen at individual primary study level. However, we saw mapping the barriers and facilitator themes onto the COM-B components as a way to take our analysis to the next step and provide recommendations for theory-guided knowledge translation strategies and understand why barriers and facilitators may exist. Additionally, as per the Thomas and Harden approach, we did not code directly onto any part of the manuscripts and focused our extraction on the results and findings sections, meaning key evidence may have been missed. We only included studies published in English which limits the generalizability of our findings to English-speaking countries or those that can pay for translation services. There is subjectivity in mapping of barriers and facilitators onto the COM-B components; some barriers and facilitators could map onto different components depending on the readers’ interpretations. Though we identified candidate intervention functions for implementing guidelines in LTC, we did not assess which ones are feasible and realistic to implement. Our next steps are to use the APEASE criteria [35] in consultation with stakeholders to determine the most appropriate intervention functions for the LTC sector.

We suggest that people designing LTC interventions to support guideline implementation assume the most frequently identified barriers (time constraints and inadequate staffing, cost and lack of resources, knowledge gaps, and lack of teamwork and organizational support) and facilitators (leadership and champions, well-designed strategies, protocols, and resources, and adequate services, resources and time) in our review are present and design strategies targeted at physical and social opportunity and psychological capability. Further analysis of barriers and facilitators specific to the guideline they are implementing may be necessary if the actions outlined by the guideline have unique features that could cause additional barriers and facilitators.

Implications for research have been developed based on the findings of our study and our GRADE-CERQual assessment of findings. Future qualitative work in this area should transparently report researcher reflexivity including a reflection of the researchers’ roles and the influence this may have on the findings of the study. Additionally, researchers must fully describe the context of their LTC setting to ensure readers can determine whether the findings apply to their local LTC context. A full description of context would include the care philosophy of the home, staffing levels, and health system influences (e.g., public or private funding) among other factors.