Implementation of clinical practice guidelines in the healthcare setting: A Concept Analysis

Clinical practice guidelines (CPGs) are developed through a rigorous systematic methodology synthesizing the ever-increasing amounts of published literature into a practical and digestible set of clinical recommendations to be used in a healthcare setting. 1,2 The Grading of Recommendations Assessment, Development and Evaluation (GRADE) collaboration has established a widely-accepted approach for developing guidelines by rating both the quality of the evidence and the strength of the recommendation. 3 The U.S. Preventive Services Task Force (USPSTF) also develops guidelines using a similarly rigorous and transparent methodology. 4 Both approaches recognize that guidelines need to be trustworthy and understandable from the perspective of key stakeholders, and the ultimate goal is to direct clinicians in providing the most up-to-date, evidence-based, and highest quality care for their patients.

Complexities in Implementing CPGs

Although our literature is replete with guidelines, the expected improvements in patient outcomes and reduction in healthcare-related costs have not been realized. 1,5,6 Research suggests that on average, it takes up to 17 years for only 14% of published evidence to translate into practice. 7,8 In response to this, research has shifted to focus on how guidelines can be effectively implemented in a clinical or community setting. 9,10 These strategies are extensive and comprehensive; as one example, the Expert Panel for Recommending Change (ERIC) project, 10 names 73 strategies and their clear definitions, highlighting the breadth and combinations of possible implementation strategies. These strategies are defined as “techniques used to enhance the adoption, implementation, and sustainability of a clinical program or practice 11 ” and often involve multi-faceted components. 12 Another nuance to implementation strategies is the measurement or outcomes which often differ from traditional clinical trial outcomes which focus on efficacy or effectiveness of an intervention. Implementation outcomes instead include acceptability, appropriateness, costs, and sustainability of the evidence-based intervention, and/or the implementation strategy. 13 Because of the notable time lag in implementing research into practice, the barriers to implementation have also been described and vary by the implementation strategy being utilized, setting characteristics, provider and patient cultural characteristics, and the clinical practice change itself. 5,14

The new evidence or practice change being introduced or implemented into a clinical or community setting is important to consider. Traditionally, the highest level of evidence has been the randomized-controlled trial, often used to inform robust systematic reviews and clinical practice guidelines. 15 These trials are designed to answer questions about efficacy of an intervention, and therefore have strict inclusion criteria to maximize internal validity. This is an important consideration when interpreting the results of a published trial because the results tend to be limited in generalizability and external validity. 16 As such, clinicians often have little understanding as to whether results from RCTs are applicable across settings, populations, and conditions. 16 More recently, focus has shifted towards effectiveness or pragmatic trials, as this type of trial can better evaluate the real-world effectiveness of an intervention. 17 These trials can help to address mistrust among clinicians and decision-makers of clinical practice guidelines that has been cited due to actual and perceived limitations of generalizability to general populations. 18,19

Implementation science has emerged over the past ten years, taking a transdisciplinary research approach and seeking to advance understanding of the methods and strategies used to promote the systematic uptake and integration of evidence-based practices and programs across diverse real-world clinical or community settings. 20 The field of implementation science continues to grow as the facilitators, barriers, frameworks, and strategies for uptake of the evidence into practice are developed and identified across contexts and populations. 5,10,20,21

Need for a clear definition of guideline implementation in healthcare settings

Despite the large volume of research in implementation science, the concept of implementation is relatively new to the clinical setting. It is instrumental to changing clinical practice and ultimately, providing the highest quality of care to patients. However, the theories and models that have defined this field may not be apparent or accessible to the clinicians providing care to patients and the community. It is therefore important to define and describe this concept to ensure nursing and other clinical communities understand its full implications. To provide clarification on this critical and timely concept, the intent of this paper is to conduct a concept analysis of implementation of clinical practice guidelines in healthcare settings.

METHODS

This concept analysis was developed using Walker and Avant’s eight-step methodology. 22 This process includes 1) choosing a concept; 2) determining the purpose of the concept analysis; 3) identifying all uses of the concept; 4) identifying the defining aspects or attributes; 5) describing a model case; 6) describing a borderline, related, contrary, invented, and illegitimate case; 7) identifying antecedents and consequences; and 8) defining empirical referents. This method is useful for defining and describing concepts that are not well-defined and can subsequently be used to inform future studies relating to the concept of interest.

We conducted a focused literature search to identify the existing definitions of implementation of clinical practice guidelines as outlined in step 3 of Walker and Avant’s methodology. We searched PubMed using a combination of keywords and MeSH terms relating to “implementation,” “evidence-based practice,” and “delivery of health care” after consultation with experts in the field and an information specialist (see Table 1 for our search strategy). Studies were included if they were systematic, scoping or integrative reviews of guideline implementation in a healthcare setting. We included conceptual frameworks and focused on broader implications of implementation processes. We excluded studies that focused on a specific disease or specific clinical settings. We also excluded text books from our formal literature review although we acknowledge one major text book in this area. 23 Finally, we searched the National Institutes of Health/National Cancer Institute’s Grid-Enabled Measures Database (GEM), an interactive website of scientific measures and organized by theoretical constructs. 24 We searched GEM for the construct “implementation” to ensure we did not miss any relevant theoretical frameworks or operational definitions.

Table 1.

Literature Search Strategy

We also searched Google.com (google scholar?) to identify the various settings that use this concept of “implementation” outside of the healthcare arena.

FINDINGS

Our search strategy retrieved 3226 articles from PubMed, and one additional study from reference searching. In addition, utilizing GEM helped to ensure we identified all relevant existing frameworks and measures for implementation as a construct; no additional studies were identified through this database. We excluded the majority of our search by title and abstract review; and ultimately identified 17 publications to include in this review (see Figure 1 for PRISMA flow chart).

PRISMA Flow Diagram for Search Strategy

Clarification on Definitions of Implementation

Implementation is derived from the verb “to implement” and is defined in the Merriam-Webster dictionary as “to give practical effect to and ensure of actual fulfillment by concrete measures.” Implementation is discussed frequently not only in healthcare settings, but also in relation to computer science, information technology, political science, and conservation studies. For example, due to the availability of increasing volumes of data and new information, many disciplines are finding that implementation of new or developing strategies or evidence has become fundamental yet challenging.

The literature search confirmed a similar response in the healthcare community to the overwhelming amount of published data that has resulted in a shift toward implementation science. Our results are summarized in Table 2 , and the publications included in the review described the relevant frameworks, high-level concepts and domains relating to implementation in a clinical setting, and a few select publications that related specifically to the nursing field. 25,26

Table 2.

Summary of Publications included in Literature Review

5-factors: Reach, Efficacy, Adoption, Implementation and Maintenance needed to determine impact of an intervention

Focused on three factors that influence uptake of evidence: attributes of evidence, barriers/facilitators to practice change, effectiveness of dissemination and implementation strategy

Highlights the need for communication between stakeholders (funders, practitioners, trainers, and researchers)

Defines 5 elements within framework: Program (intervention), External Environment, Implementation and Sustainability Infrastructure, and Recipients

Defined 5 domains: intervention characteristics, outer setting, inner setting, individual characteristics, and implementation process

CFIR can guide formative evaluations of implementation strategies as well as build implementation knowledge base

Describes implementation in phases: Exploration, Adoption decision/Preparation, Active Implementation, Sustainment

Each phase has an outer context, inner context, interconnections, innovation characteristics, system and organization fit

Responses from RNs were classified in 4 categories: internal facilitators, external facilitators, internal barriers, external barriers

Further categorized into attitude-based, knowledge-based, organizational characteristics, and patient characteristics

Sorted by three categories: construct flexibility (broad to operational), Dissemination and/or Implementation (D-only, D=1, I-only), and Socioecologic framework (Individual, organization, community, system)

12 models were D-only; 4 were I > D; highlights the substantial overlap between D&I and other theoretical foundations

Expands to frame implementation within social mechanisms and processes; highlights complexity of implementation

12 systematic reviews identified to develop comprehensive, integrated checklist of determinants of practice (TICD)

7 domains include: guideline factors, individual health professional factors, patient factors, professional interactions, incentives and resources, capacity for organizational change, social/political/legal factors

Demonstrated various levels of EBP implementation across medical, nursing, pharmacological, and allied health professionals

Identified internal factors (attitudes, perceptions, knowledge) and external factors (format/usability of CPG, resources, leadership, organizational culture) as potential barriers/facilitators of implementation and uptake of CPGs

1. Kitson A. Enabling the implementation of evidence based practice: a conceptual framework. Quality and Safety in Health Care. 1998;7(3):149–158.

2. Bero LA. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. The Cochrane Effective Practice and Organization of Care Review Group. BMJ British medical journal (Clinical research ed). 1998;317(7156):465.

3. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: the RE-AIM framework. American journal of public health. 1999;89(9):1322–1327.

4. Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients’ care. The Lancet. 2003;362(9391):1225–1230.

5. Wandersman A, Duffy J, Flaspohler P, et al. Bridging the gap between prevention research and practice: the interactive systems framework for dissemination and implementation. American journal of community psychology. 2008;41(3–4):171–181.

6. Feldstein AC, Glasgow RE. A practical, robust implementation and sustainability model (PRISM) for integrating research findings into practice. Jt Comm J Qual Patient Saf. 2008;34(4):228–243.

7. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implementation science : IS. 2009;4.

8. Brouwers MC, Garcia K, Makarski J, Daraz L. The landscape of knowledge translation interventions in cancer control: What do we know and where to next? A review of systematic reviews. Implementation science : IS. 2011;6:130.

9. Aarons GA, Hurlburt M, Horwitz SM. Advancing a conceptual model of evidence-based practice implementation in public service sectors. Administration and Policy in Mental Health. 2011;38(1):4–23.

10. Abrahamson KA, Fox RL, Doebbeling BN. Original Research: Facilitators and Barriers to Clinical Practice Guideline Use Among Nurses. AJN The American Journal of Nursing. 2012;112(7):26–35.

11. Tabak RG. Bridging research and practice: models for dissemination and implementation research. American journal of preventive medicine. 2012;43(3):337–350.

13. Flottorp SA, Oxman AD, Krause J, et al. A checklist for identifying determinants of practice: A systematic review and synthesis of frameworks and taxonomies of factors that prevent or enable improvements in healthcare professional practice. Implementation Science. 2013;8(1):35.

14. Weng Y-H, Kuo KN, Yang C-Y, Lo H-L, Chen C, Chiu Y-W. Implementation of evidence-based practice across medical, nursing, pharmacological and allied healthcare professionals: a questionnaire survey in nationwide hospital settings. Implementation Science. 2013;8(1):112.

15. Gagliardi AR, Brouwers MC. Do guidelines offer implementation advice to target users? A systematic review of guideline applicability. BMJ open. 2015;5(2).

16. Jun J, Kovner CT, Stimpfel AW. Barriers and facilitators of nurses’ use of clinical practice guidelines: An integrative review. International journal of nursing studies. 2016;60:54–68.

17. Birken SA, Rohweder CL, Powell BJ, et al. T-CaST: an implementation theory comparison and selection tool. Implementation science : IS. 2018;13(1):143.

The literature reviewed for this concept analysis identified that implementation is a term used in many healthcare settings, but particularly relates to clinical healthcare delivery systems, 27–29 quality improvement, 30–32 and patient safety initiatives. 33,34 The results described four important domains within the construct of implementation: 1) the practice change characteristics; 2) the guidelines or other forms of evidence which should inform the practice change; 3) strategies to implement an intervention or clinical practice guideline; and 4) the challenges and, less frequently, potential interventions to overcome these barriers. Table 2 provides further information on the domains discussed in each included publications. The literature supports the operational definition, which will be used for this concept analysis of implementation of clinical practice guidelines in the healthcare setting, as a process of changing practice in healthcare while utilizing the best level of evidence that is available in the published literature.

Defining Attributes

The defining attributes are the characteristics that are necessary for understanding and defining the concept. The attributes may be examined through the published literature, general consensus or expert opinion; a clear outline of attributes improves the ability to convey the conceptual meaning to individuals as well as across disciplines. 22 We drew the defining attributes from the literature search conducted for this concept analysis.

Attribute #1: Current Practice and Policy

In any clinical setting, there are common practices in the local health care delivery model. Current practice may be informed by previously-published evidence and may include adapted guidelines, clinical care models, standard procedures, or clinical pathways operationalized in inpatient hospitals, outpatient clinics, or private offices providing healthcare services. Current practice may also not be formally defined but instead can be viewed as current patient practice in a specific setting to address a defined diagnosis or symptom. In addition, current practice is often influenced by local, state, and federal policies. The World Health Organization (WHO) defines health policy as the “decisions, plans, and actions that are undertaken to achieve specific health care goals within a society 35 .” These decisions, plans, and actions guide the funding, development, accountability, and implementation of health-related topics. 23 Although some current practices may not be directly affected by a specific policy, because policies from the Centers for Medicare and Medicaid Services and other governmental organizations often dictate healthcare related coverage and reimbursement, it is important to understand and acknowledge the influence that policy may play in influencing clinical practice.

Attribute #2: New evidence/Innovation

When new evidence, which can also be thought of as the innovation, regarding the practice or care delivery mechanism is published or discovered in a literature review, it may either support current practice or demonstrate the superiority or efficiency of a different practice that should be adopted to provide the best care to patients. This evidence may be in the form of a clinical practice guideline developed from meta-analyses of randomized controlled trials, historically considered the strongest level of evidence, 15,36 or may be informed from the results of a single randomized-controlled trial comparing a newer treatment to the standard of care. It may also be evidence from a prospective cohort or observational study comparing the outcomes of multiple cases and controls or perhaps just one group of patients. Because the level of evidence may vary, its strength should be assessed using a standardized tool, such as The Appraisal of Guidelines for Research & Evaluation (AGREE) Instrument for guideline assessment, or Cochrane’s GRADE assessment for other research designs. 37,38 This process, however, is burdensome and often takes a trained, dedicated team to operationalize. In addition, the process of appraising the evidence should include consideration of the generalizability of the results. 18,16

Attribute #3 Introduction of new practice or new evidence

The new evidence or practice can be introduced into the healthcare setting through various entry points. Ultimately, each of these (XX seems like word is missing here?) is an important component of this attribute.

Institutional assessment of the new practice

Once the evidence or innovation is published or made publicly available, an assessment may be conducted by hospital administrators, a guideline-review committee, risk management, or other committees responsible for evaluating the current practices and policies of an institution, hospital, private practice, or other community-based healthcare system. This attribute might be a more feasible and appropriate phase to introduce the evidence-assessment tools discussed above. 3,37,38 In addition to the above stakeholders who contribute at this phase of implementation, whenever possible, the healthcare setting should consider including the health information technology team or, if possible, clinical informaticians. Implementation processes should include integration into the clinical workflow wherever possible, and this often translates into electronic health record integration or clinical decision support systems. 31,39,40 The literature cites integration into workflow as a potential barrier to implementation, 26,41,42 and having those who can help integrate a guideline in an automated approach may optimize the implementation strategy.

Individual clinician assessment of the new practice

The next attribute of implementation is the assessment of the practice change recommendation by an individual clinician or healthcare provider. Because this involves an individual’s personal assessment, this attribute is likely to cause greater variability in an implementation process. For example, although the entire process of implementation is built on an underlying foundation in evidence-based practice, the individual clinician’s assessment of the new practice will depend heavily on that individual’s perception and trust of the evidence (and the source?). 43 If the individual understands and values the methodology that informs the evidence-based practice and guidelines and the source is credible, it is likely that the clinician will determine the process to be credible and acceptable, allowing progress from this attribute toward subsequent practice change. 44,45

Patient Education/Involvement in Decision-Making about New Practice

Many changes in clinical practice will directly affect the patient, and consideration of the patient as a stakeholder is an additional requirement for effective implementation. Depending on the level of direct contact that a patient will have with a proposed change in practice, involvement in the decision-making and implementation plan may vary. For example, updating a cancer screening process or a vaccination schedule heavily relies on the patient to comply with the recommendation; on the contrary, an update to a method for measuring blood pressure or assessing for risk of stroke relies less heavily on patient compliance and buy-in. Regardless, the patient’s decision-making should be a consideration during the development of a clinical practice guideline implementation plan. 42

Attribute #4: Practice change

Practice change, the next phase and attribute, is often complex and multi-dimensional depending on the modification to the practice and the setting in which the practice occurs. 42 This phase involves a change in the healthcare system, healthcare providers and other staff who are involved provision of care, and patients. 41 This may also be referred to as the active implementation phase. 46 Because of the multi-dimensionality, this attribute is instrumental. Its presence confers an understanding that the previous attributes have been established and the decision is being made to change the current practice while considering the complexities that affect a successful process change.

As discussed in May’s general theory of implementation, the process of implementation as it relates to this attribute of practice change can be considered more of a process within a broader social context, and without integration into the already existing workflow, it is unlikely to be effectively implemented. 47 This attribute requires not only the prior assessments and acceptance of the proposed practice, but also a system that will support the adaptation. This attribute is where the majority of the described multi-level and multi-factorial facilitators and barriers will become evident. For example, a healthcare system that is culturally-adaptive to change and has a well-integrated clinical informatics program to support the technical requirements to integrate the change will likely implement the guideline more effectively than one that does not have these characteristics. 40

Attribute #5: Plan for Reassessment and Evaluation

Following an effective implementation of an updated or new clinical practice guideline, a clear plan for follow-up of the effectiveness should be established. This often takes place in clinical settings as a quality improvement/quality assurance protocol. It can also be more formalized through an established group of clinician champions who are tasked with evaluating the literature at regular time intervals. This reassessment phase will also help to identify challenges in the implementation plan and may provide discussions for adapting the current procedures to make them more feasible in a particular clinical setting. Similar to the published methodologies for developing and updating guidelines, an institution should establish similarly transparent practices for updating and adapting its procedures.

Deimplementation

Because implementation by definition requires a change in practice, we recommend consideration of deimplementation as an important component, often occurring during the reassessment and evaluation phase. Deimplementation occurs when a previously-accepted practice, studied under new, or refined, research methods, is found to be ineffective or even harmful. 48 A recent concept analysis defined deimplementation as “the process of identifying and removing harmful, non-cost-effective, or ineffective practices based on tradition and without adequate scientific support.” 49 Sometimes referred to as medical reversal, this attribute should be developed in parallel to implementation strategies of practice change. Although deimplementation describes a reversal from a previously-established practice, the plan to fill the gap created from a deimplemented practice needs to be clear. One cannot simply stop providing care in a certain way without a plan to continue to provide the care that is needed in an accepted and pre-determined approach. Deimplementation, in fact, has been cited as political and contentious. 50 Financial, political, and other special-interests may present an additional barrier for deimplementation of a specific practice. 51

Cases

Cases provide exemplars of the concept in a contextual format and helps to further clarify the concept of interest. 22 The ideal case provides an example of the concept in an optimal setting or use and can be thought of as a gold standard. The other cases are either suboptimal uses of the concept or not meeting conceptual criteria at all.

Ideal case

Westwell Hospital had been prescribing antiemetic medications for their pediatric oncology patients utilizing an algorithm that was developed by the providers caring for the patients based on a 2013 guideline published in a relevant oncology journal. The algorithm was developed following the guideline’s publication in early 2014 by the division’s Clinical Guidelines Committee and implemented into their electronic health record in collaboration with their clinical informatics department through a clinical decision support system that provides recommendations based on the algorithm. A nurse involved in the Division’s quality assurance research initiatives recently attended a conference and received education about a more current guideline from Pediatric Oncology Group of Ontario (POGO) published in 2018 on the prevention of acute chemotherapy-induced nausea and vomiting. The updated guideline presented data that no longer recommended a previously-used medication for antiemetic purposes and allowed another antiemetic to be used in younger patients than previously published. The nurse presented this updated guideline to the policy and procedure committee and suggested they revise their algorithm for prescribing antiemetics to remove the ineffective medication. They evaluated the guideline using the AGREE approach and it received scores that support approving implementation. The nurse subsequently presented this data to the nurses and physicians in the division to assess for their interest and agreement in modifying their prescribing process and remove the outdated medication from their algorithm as a first-line medication. They came to consensus and agreed to modify their practice, and the pharmacists, providers, clinical informaticians and health information technology (IT) team worked together to integrate the new guideline into the prior automated clinical decision support in conjunction with the computerized-provider-order-entry (CPOE) algorithm.

In addition to developing the algorithm, the implementation process included an educational initiative for the patients and their families to ensure they understood any changes in individual antiemetic regimens, and that any questions related to changes in their medication schedules were addressed. A team of nurses developed a research project to assess the success of the implementation strategy by examining the adherence to the guideline; they also explored the barriers and facilitators to adopting the guideline through qualitative study. The research project was approved through the local institutional review board, and ongoing assessments were conducted.

This is an ideal case because it includes all of the attributes previously defined as necessary in the implementation process. Specifically, all relevant stakeholders were involved in the process including nurses, physicians, informaticians, pharmacists, and in this case, patients. The key champion, a nurse, assessed the evidence and considered it rigorous enough to implement into practice. In addition, the ongoing assessment in the form of a research project strengthened the practice change process and helped to ensure that the relevant providers will continue to value optimal antiemetic prescribing practices.

Borderline case

Westwell Hospital had been prescribing antiemetics for their pediatric oncology patients utilizing an algorithm that was developed by the physicians caring for the patients. The algorithm was not developed from the literature even though extensive literature is available on this topic. The hospital had an administrative committee to develop policies and procedures, and this committee decided to update their guideline for antiemetic prescribing practices. An administrator searched online for a guideline and subsequently chose an outdated publication to use as a template for their updated algorithm. After implementing the new guideline into their algorithm, they posted the updated policy on the internal hospital website and sent an email to the providers in pediatric oncology to notify them of the implementation. They also notified the pharmacy department of the implemented practice change. The division of pediatric oncology provided ad hoc education to both physicians and nurses about the changes, and some of the nurses provided education to their patients and families.

Subsequently, the hospital committee received updated practice guidelines through a central emailing mechanism, and one of the committee members noted that there is an updated guideline on prevention of chemotherapy-induced nausea and vomiting in pediatric oncology patients which omitted one medication that had previously been widely used by the hospital. The committee reviewed the guideline and followed the same procedures to update their algorithm and reverse, or deimplement, the prior practice that included this medication. Emails were again sent to the physicians, nurses, and pharmacists to notify them of the deimplementation. The updates were also briefly discussed at a division meeting.

This case is borderline because while it does include some of the attributes required for implementation, the vital component of individual clinician assessment is missing from the process. Without this important step, it is not clear that the algorithm will actually be practiced; clinicians who do not understand or believe in a practice change are less likely to adopt the change. Furthermore, without the input of these important stakeholders, it is unlikely that nurses and physicians who are providing the care would extend themselves and provide additional education to their patients if they do accept the deimplementation of the practice or conduct an assessment of patient outcomes with the change in practice, as the nurses did in the ideal case.

Related case

Westwell Hospital identified a need to update their policies and procedures for a Joint Commission inspection. A committee was formed of pharmacists, nursing supervisors, nurse practitioners, physician assistants and physical therapists with a physician who was designated as the chair. They examined each policy in alphabetical order and described if they follow the policy as stated or if the usual practice in the hospital is a variation on the policy. If the policy is not followed as prescribed, they explained why they have adapted the policy. The usual practice was then assessed by the committee to determine if the variation was within standard acceptable procedure, and if so, they modified the policy to match the practice. Occasionally, a published guideline was introduced and, if everyone on the committee agreed, the guideline informed an amendment to the policy. When more than one guideline was available from the different disciplines, the committee voted on which one to incorporate.

This is a related case because policy and practice are being aligned, but it is not necessarily consistent with guidelines or best practice. There is no individual provider assessment or step to allow for education and ultimately, adoption by the clinicians. This case demonstrates a related but distinct concept from implementation.

Contrary case

Westwell Hospital had a new pediatric oncology division chief who did not agree with the current antiemetic algorithm for pediatric oncology patients. The chief determined that they change the algorithm to utilize an outdated medication not backed by evidence because it is the way that she always practiced. There was no plan in place to review the algorithm to identify new information or modify the current practice.

This is a contrary case because it does not contain any of the required attributes. One person makes a determination that is not based on any evidence nor does it have stakeholder input, and the practice is determined without a plan for any future assessment or modification.

Illegitimate case

Westwell Hospital advertised an interest in implementation science, and they delegated an administrative committee the task of reviewing the medications prescribed in the hospital. The committee focused on the most expensive and frequently prescribed medications and realized that a certain class of antiemetics were at the top of the list. Under the premise of implementation science, the committee decided to remove the class of drugs from the formulary and replaced the recommendation with an older class of drugs in the antiemetic algorithm. A notice was sent out to the nurses, physicians, and pharmacists in the pediatric oncology division to update them on the new algorithm recommended by a literature review.

This is an illegitimate case because the hospital cites this as part of their implementation process, but in fact, it is not implementation as the evidence that they are citing is illegitimate or non-existent. Similar to the related case, new evidence, the initiator of the process, is not present. This case would be deemed illegitimate because it hides under the ruse of implementation.

Antecedents, Consequences and Empirical Referents

The antecedents and consequences of a concept describe the preexisting requirements and subsequent value or potential risks of the concept if uniformly understood. Empirical referents provide the relationships to the defining attributes and therefore, benchmarks to measure the conceptual outcome of interest. 22 These components of the concept analysis can provide a framework to guide the development and evaluation of future related studies.

Antecedents, or requirements, for implementation of guidelines include a healthcare setting that provides care to individuals. Furthermore, the concept requires that providers of the care are established as well as the patients who are receiving the care. In addition to the setting, the concept of implementation requires the availability of evidence-based clinical practice guidelines. This is important to acknowledge as guidelines can only be developed when there is evidence to support them. 19 These factors must be in place to create the setting for this concept. It is important to acknowledge, however, that although the described cases involve a hospital setting, implementation of guidelines can occur in any setting where healthcare is provided, such as a public health clinic or through community health workers.

Although implementation of guidelines should lead to provision of high-quality, evidence-based care, there are potential consequences to consider. Because the process of implementation involves multiple steps and individuals, errors may occur along the way. These may include but are not limited to: practice change to a non-evidence-based practice, individual fatigue of the evidence-based process leading to distrust of the resulting evidence and recommendations, and increased tension between healthcare institutions and the individual clinicians due to discord in the assessment of the evidence. Each of these potential outcomes may lead to poorer quality care in the healthcare field.

The defining attributes of implementation can also provide benchmarks to measure progress or highlight incorrect or ineffective implementation strategies. Intervention or measurement at any of the five previously-described attributes will guide further understanding and improvement of this process. Prior frameworks, specifically those discussed in Tabak and colleagues’ evaluation of 61 models, provides a historical referent point to evaluate implementation of clinical practice guidelines. 27 These are consistent with our findings which highlight the multi-level process that is necessary for effective implementation.

IMPLICATIONS FOR CLINICAL PRACTICE

Clinical practice is constantly shifting and adapting to new evidence and published guidelines. Practice may also be mandated at the policy-level of a hospital, clinic, or other healthcare setting. Because implementation of these evidence-based guidelines directly affects clinicians in their work environment, it is necessary to clearly define and delineate the specific attributes of this concept. We have demonstrated that the healthcare provider is an integral part of an effective implementation process, and the uninformed or opposed clinician may delay or arrest even the most well-designed implementation strategy.

In addition, healthcare providers, as the frontline of the care delivery system, should be actively involved in the process of adaptation of guidelines to integrate them into the local workflow. 52 Without clinical input, it is unlikely that an implementation strategy will be effective and sustainable. 13 The informed clinician can and should be an integral part of sustaining new, effective clinical care and conducting ongoing evaluation to monitor its impact over time.

LIMITATIONS

This concept analysis describes implementation in the healthcare setting. The cases are focused on guideline implementation in a clinical setting, but the scope of implementation is broad and includes health policy or another system-based implications. The breadth of this concept may be simplified in a limited concept analysis such as this where only implementation of clinical practice guidelines in a specific setting, healthcare, is examined. We propose, however, that limiting the concept to this setting will aid clinicians in better understanding the concept and its importance to their local care delivery model.

In addition, the narrow focus of the theoretical cases may simplify the concept and underappreciate its extensive value. It is important, however, to provide concrete examples to clarify the processes involved in implementation. The cases were chosen specifically because of their concreteness in the hope that clinicians will find them relatable and tangible.

Another potential limitation of this concept analysis is the literature review. Although we included broad terms in the search, it is possible that we missed important publications that contributed to the implementation science literature. To reduce this limitation, the included publications were reviewed by an expert in implementation science as well as through the GEMS search. In addition, we prioritized systematic reviews and theoretical frameworks or reviews of multiple frameworks to ensure comprehensiveness.

CONCLUSION

This concept analysis provides a comprehensive definition of implementation of clinical practice guidelines in clinical settings. It is our hope that this analysis will provide a clear definition of the concept for clinicians and healthcare providers as they are exposed to implementation science studies at their institutions. Quality improvement projects, implementation strategies, and full-scale interventional studies are being conducted more often in a wide-variety of clinical settings, and it is imperative that clinicians and researchers are able to communicate clearly and understand the meanings of these terms, concepts, and projects.

Implementation and deimplementation fundamentally are both vital components that are distinct but related. In the case of deimplementation, the required practice change signifies that the previous practice is accepted as no longer effective. The healthcare system has determined, through the same process as implementation, to stop the ineffective practice in exchange either for a previously-established practice, or introduce a newer practice if there is evidence to move toward a newly-established practice.

We propose a repeating-process model ( Figure 2 ) that is continually resting on a belief and reliance on evidence-based practice. Each of these attributes is a potential area of focus for future interventions in implementation and deimplementation research. These are supported by previously-established frameworks of implementability that describes multiple domains necessary for a guideline to be implemented 53 . Although they are not identical, they share concepts and sufficient overlap. Our conceptual model, however, visualizes the process as iterative which is necessary to define and integrate implementation and deimplementation processes into a clinical practice.