The scope of this review was to explore the features and critiques of the different models of EIPM in healthcare. We performed a systematic search of PubMed, Web of Science and Scopus databases from February 2023 to June 2023 using three categories of words; research dissemination or utilization, health policy or decision making, and framework or model. Only the models that satisfied the following criteria and provide practical guidance and support for evidence-informed policymaking were included in this review:

For systematic search and review, quality assessment may or may not be included (50). We acknowledge that quality appraisal can be useful but we decided not to conduct a quality appraisal of the models that we included in our review because quality appraisal involves subjective judgments and interpretations that may vary depending on the perspective or context of the appraiser (54); and we wanted to avoid imposing our own views or values on the models. Moreover, we followed the examples of some authors who have conducted similar reviews of normative or theoretical literature, and who have justified their choice not to perform a quality appraisal (19, 54). Using title, abstract and full text, two reviewers (SHJ and SHY) screened the search results independently. Any disagreements were discussed and resolved by consensus between them. We recorded the number and reasons of studies included and excluded at each stage of the screening process.

Through this process we identified the relevant sources of information for our review. We also checked all articles that cited the included models to find other scholars’ perspectives about the models. In the following sections, we will describe each model in detail, along with its original description and scholarly perspectives, and then we will discuss our own critique about different models of EIPM in healthcare. Figure 1 shows a PRISMA flow diagram of our selection process. Table 1 shows the reasons for including each model in our review.

According to Logan and Graham, the developers of the OMRU, this six-step framework provides a valuable approach for implementing and evaluating healthcare innovations in the context of evidence-informed policy making. The OMRU offers a comprehensive and dynamic approach to knowledge transfer, incorporating elements such as evidence-based innovation, potential adopters, and implementation interventions. The model emphasizes the importance of assessing, monitoring, and evaluating each element to overcome barriers and enhance supports for successful knowledge transfer. The six steps involve identifying individuals with the authority to implement changes, clarifying the details of the innovation, assessing barriers and supports to adoption, planning and tailoring implementation interventions, measuring and monitoring adoption, and evaluating the impact of the innovation on health, practitioner, and system outcomes (55, 56, 99). Analyzing the strengths and limitations of the OMRU provides insights into its utility as a model for evidence-informed policymaking, which aligns with the central aim of this review in assessing models that facilitate evidence-informed policymaking.

The OMRU and its strengths and limitations have been discussed by various authors. The OMRU is praised for its operational focus, which outlines the actions for effective implementation of behavior change interventions (100), its dynamic and interactive approach to research and knowledge translation, which involves stakeholders and reflects complex systems thinking (101, 102), its assistance to administrators to control factors influencing organizational-level changes (103), and its usefulness to facilitate knowledge translation at the local context, especially for continuity of care interventions that bridge sectors, settings and provider groups (104). The OMRU also follows the principle that knowledge translation strategies should be tailored to the specific barriers and supports of each setting, making it efficient and adaptable (105), and it is clear and easy to use (106). Moreover, the OMRU is one of the few frameworks that address implementation at multiple socio-ecological levels (107). However, the OMRU also has some limitations, such as the high demand for time and resources, the lack of specific guidance on how to choose and apply knowledge translation strategies, and the potential challenges for its applicability in low-resource setting (103). The OMRU also focuses on the clinical practice setting rather than the health system as a whole, needs further development and validation of tools and instruments to support its implementation, and does not cover research production or knowledge creation as part of knowledge translation (106). Additionally, the OMRU may limit the scope of the investigation and overlook some factors that are not included in the framework (108). According to the original authors of the model, some of the strengths of the OMRU are that it is flexible, interactive, and practical, while some of the weaknesses are that it is complex, time-consuming, and challenging (99). However, these perceptions may vary depending on the users and the contexts of the OMRU’s application. Some users of the OMRU have reported that it is clear and easy to use (106), while others have found it difficult and demanding (103). Therefore, the OMRU’s usability may depend on the specific situation and the needs of the users.

The CHSRF model was first published in 2000 by the Canadian Health Services Research Foundation (CHSRF), which is now called the Canadian Foundation for Healthcare Improvement (CFHI), to address the gap between research and practice in the health sector (109). It is also known as the Lomas model, after Jonathan Lomas, the first chief executive officer of the CHSRF, who led the development of the model (109). The model suggests creating communication channels among four key groups, namely researchers, decision makers, research funders and knowledge purveyor. It also emphasizes the need to enhance the ability and willingness of decision makers and their organizations to access and apply research evidence. It acknowledges that different kinds of decision makers face various uncertainties about facts and values, and that research findings are often not useful or applicable for their needs. It aims to promote health-related research, especially in the areas of pharmaceuticals, technology and medicine-related innovations. It shows the diversity of actors in each category and how they provide information and resources to each other (62).

The CHSRF model has been challenged by various authors who identified its advantages and disadvantages. Evaluating the pros and cons of the CHSRF model contributes to the overall goal of this review in assessing models that aim to facilitate the use of research evidence in policy decisions. The model has some advantages, such as its focus on linkage and exchange, which fosters trust and mutual exchange of knowledge and influence between researchers and decision makers (110), its consistency with the approach of Weiss (8) and Caplan (9), who conceptualized evidence-based policy making as an interplay between a supply of and demand for evidence (86), its attention to improving relationships and communication across the four groups in the health sector (62), and its relevance for understanding and influencing policy communities, funding and brokering aspects of research-policy nexus (111), and its promotion of interaction between the producers and users of research, which is a necessary strategy for health researchers in Canada (112). However, the model also has some challenges, such as its reliance on complex and variable partnerships between academics and non-academic partners, which could cause conflicts or misunderstandings (113), and its lack of development in the research literature, which may limit its theoretical and empirical foundations (112).

One of the comprehensive models that aim to support policymakers in using research evidence in their decision-making is the FRDU, which was developed by Dobbins et al. (65). This framework draws on the diffusion of innovations theory and integrates the concepts of research dissemination, evidence-based decision-making and research utilization. It suggests that four sets of characteristics influence the innovation adoption process for policymakers who want to adopt research evidence into their policies. These characteristics are innovation-related, individual, organizational and environmental. The framework also outlines the five stages of innovation adoption, which are knowledge, persuasion, decision, implementation and confirmation. The framework synthesizes relevant literature from the management and health-care fields, and is supported by empirical data from various studies. The original authors of the framework evaluated its strengths and weaknesses, as well as its challenges and facilitators for implementation. Analyzing the strengths and weaknesses of the FRDU framework provides insights into its utility for evidence-informed policymaking, connecting to the main aim of this review. They found that the framework was clear, applicable and useful, but also simple, linear and context-free. They also identified the scarcity of resources as the main challenge to implementing the framework, and having a champion or an external factor as the main facilitator. The authors of the framework intend to continue testing and refining it based on new knowledge and feedback.

FRDU has both strengths and weaknesses, as various authors have reported. Some of the strengths are that it considers constructs and stakeholders at different levels of the health care system (114), it helps to clarify the stages, areas, complexities, and variables of research dissemination and utilization process and evidence-informed decision-making in health care field and policy practices (115), and it has a strong theoretical and conceptual basis that captures the complexity and diversity of evidence-informed decision making and supports the role of communities of practice in facilitating the use of research evidence (116). However, the framework has some limitations and gaps. It suggests a one-way and limited relationship between researchers and knowledge users, which may not capture the complexity and reciprocity of evidence-informed decision-making (114). The framework’s reliability and usefulness are questionable due to the lack of empirical support or validation in real-world settings, which calls for more research to provide more evidence for the framework (117). Another drawback is that the framework does not have a specific or comprehensive theoretical foundation for specific knowledge translation mechanisms, such as evidence briefs (118). Moreover, the framework ignores the political and dynamic context of policy making, especially in low-income countries, which may affect the use of research evidence (119).

This model, developed by Bowen and Zwi (27), proposes a three-stage process for how evidence can inform policy and practice in different contexts. The first stage is sourcing the evidence, which involves finding and selecting various types of evidence that are relevant and responsive to the policy and practice issue. The second stage is using the evidence, which involves applying and interpreting the evidence in the policy and practice context. The third stage is implementing the evidence, which involves enacting and evaluating the evidence-based innovation. The model also considers the policy context, the decision-making factors, and the capacity for implementation as key influences on each stage of the process. The model draws on literature from health, public policy, social sciences, and capacity building to provide a comprehensive and dynamic framework for evidence-informed policy and practice.

The EIPPP has much to offer, but it also has some limitations that need to be considered. The model has several advantages, such as suggesting that policy is largely shaped by evidence and can respond to the community needs (120), connecting evidence-based policy-making with capacity building and enabling context-appropriate policies (121), focusing on three socio-ecological levels in the decision-making process, emphasizing the policy context and its influence on the interaction between research, evidence, and the policy process (122), identifying the key points of intervention in the pathways to evidence-informed policy and practice, and taking into account the contextual and decision-making factors and the policy influences (123), being applied to inform an intervention design and explaining the significance of policy processes and evidence use in local government planning (124), assisting researchers and policy actors in navigating the use of evidence (125), complementing other theories of knowledge utilization by highlighting how external factors affect the use of evidence in decision making (126), and showing how evidence may be used in the decision-making process (123). However, the model also has some limitations that need to be considered, such as not linking capacity building and interactive ‘knowledge to action’ in a consistent and theory-based way (121), focusing on evidence, how it is spread and applied with little attention to the political and changing policy-making context in low income countries (119), not accounting for knowledge generation beyond a policy idea (123), and being limited in the detail given to the influences on decision making (123). Evaluating the advantages and drawbacks of the EIPPP contributes to the central objective of this article in assessing different models for their utility in facilitating evidence-informed policymaking.

The JBI is a model that guides how evidence is generated, synthesized, transferred and implemented in healthcare settings. It was developed by the Joanna Briggs Institute (JBI), an international research organization and collaborative network based in the University of Adelaide, South Australia (66). The JBI Model was updated in 2019 (127) to reflect the changing international discourse relating to evidence and its translation into health policy and practice. The JBI Model proposes overarching principles related to culture, capacity, communication, and collaboration (the 4 C’S) and considers different types of evidence that address the feasibility, appropriateness, meaningfulness and effectiveness (FAME) of healthcare practices. The JBI Model consists of five components that are evidence generation, evidence synthesis, evidence transfer, evidence implementation and global health. Evidence generation is about asking relevant and meaningful questions that reflect the needs and concerns of health professionals and patients. Evidence synthesis is the process of combining and summarizing different types of evidence to inform healthcare decision making. Evidence transfer is the process of sharing and integrating evidence into healthcare practice and policy. Evidence implementation is the process of applying evidence into healthcare practice and policy. Global health is the component that represents both the global and local aspects of evidence-based healthcare. The JBI Model emphasizes the critical relationship between research and practice and policy and the importance of stakeholder engagement, the localization of knowledge, responsiveness to local knowledge needs, shared decision-making and sustainability (67).

The JBI model is praised for being a comprehensive and inclusive model that integrates and accommodates different components and kinds of evidence in healthcare (128). It provides insights into how to apply and use knowledge in diverse and complex healthcare settings around the world (128). Moreover, the model incorporates various aspects of healthcare, such as needs, research production, synthesis, transfer and implementation, and takes into account the variety of contexts and evidence types that are involved in evidence-based healthcare (129). Analyzing the strengths of the JBI model aligns with the overall goal of this review in evaluating the utility of different frameworks for evidence-informed policymaking.

Ashford et al. (68) developed TKPA to guide how knowledge can be translated into health policy-making. The framework consists of three key activities that can create a window of opportunity for policy change that are agenda-setting, coalition building, and policy learning. Agenda-setting is the activity of defining and prioritizing the problems and solutions that need policy attention. Coalition building is the activity of forming and maintaining alliances and networks among actors who have a common interest or goal in influencing policy. Policy learning is the activity of acquiring and applying knowledge and skills to inform policy decisions. The framework was also applied in Kenya to incorporate evidence from a national assessment of health services into decentralized planning at the district level. The application of the model revealed some lessons and challenges, such as the importance of engaging stakeholders at all levels, the need for capacity building and technical support for data collection, analysis and reporting, the challenge of ensuring data quality, availability and accessibility, and the need for a clear policy, legal and institutional framework to guide public participation and inter-governmental coordination.

Robertson (130), praises TKPA for adding a new element of policy learning, which emphasizes the importance of using data and information to influence policy decisions. He also gives an example of how Ashford showed that scientific evidence and research results helped create health plans that were accepted by the Kenyan government. He suggests that Ashford’s model is better than Kingdon’s policy window, because it has a new component of policy learning. Similarly, Votruba et al. (19) describe the TKPA as an effective and appealing approach to create opportunities for policy change in Kenya, as it involves action through three key elements: setting the agenda, building coalitions and learning from data and information. Reviewing the TKPA model contributes to the central aim of this study in assessing frameworks that facilitate the use of research evidence in policy decisions.

One of the models that aims to explain and guide the process of moving knowledge into action is the knowledge to action framework, developed by Graham et al. (25). The framework is based on a concept analysis of 31 planned action theories and covers both the aspects of knowledge creation and action. Knowledge creation involves the generation, synthesis, and adaptation of knowledge to different contexts and audiences. It also involves three types of knowledge, which are knowledge inquiry, knowledge synthesis, and knowledge tools/products. Knowledge inquiry refers to primary research that produces new evidence. Knowledge synthesis refers to systematic reviews that summarize and integrate existing evidence. Knowledge tools/products refer to guidelines, decision aids, or other tools that present knowledge in a clear and usable way. Action involves the identification of a problem or gap, the selection and tailoring of knowledge, the implementation and evaluation of interventions, and the monitoring and sustainability of outcomes. It also involves seven phases of action, which are identify problem, identify, review, and select knowledge, adapt knowledge to local context, assess barriers and facilitators to knowledge use, select, tailor, and implement interventions, monitor knowledge use, and evaluate outcomes. The framework emphasizes the importance of collaboration, communication, and context in the process of knowledge translation. It also recognizes that knowledge creation and action are dynamic and iterative processes that require feedback loops and ongoing evaluation. The framework aims to bridge the gap between research and practice by providing a comprehensive and practical guide for knowledge translation.

The knowledge to action process model is widely used and recognized for its many advantages in evidence-informed policy making (131–133). It provides a clear and comprehensive framework for understanding and facilitating knowledge translation and integrated knowledge translation (134), with a unique theoretical foundation grounded in social constructivism (133). The model offers direction on key elements to consider in the knowledge translation process, such as barriers, facilitators, stakeholders, and context (135), as well as a general and flexible guide for implementation that allows for adaptation to different situations (134). With a strong evidence base and validity (131), the model can support the process of translating research evidence into practice by providing a systematic framework and various methods to overcome barriers, measure outcomes, and ensure sustainability (24). The model is a dynamic and interactive process that involves creating and improving knowledge to make it more relevant and applicable for different knowledge users, such as health professionals and policy makers (136). Additionally, the model provides a detailed picture of the actions and strategies that can be undertaken for research use in practice (137). However, despite its strengths, the model does have some limitations. For example, it does not specify how to measure the success of integrating research evidence into policy and practice (138) or offer detailed guidance on how to perform each step of the action cycle. Its suitability as an account of the knowledge transfer process is also largely unknown (77) and it may not be applicable or relevant in low-income countries where the policy making process is influenced by various factors beyond evidence (119). Additionally, the model lacks details on the knowledge synthesis phase (137) and does not link the knowledge creation and action phases well (137). Analyzing the advantages and disadvantages of the KTA model connects back to the overall objective of this review in evaluating different frameworks for their utility in evidence-informed policymaking.

An additional model of evidence informed policy making is MLRA, which was developed by Lavis et al. (69) to assess country-level efforts to use research for health policy and systems decisions. The model consists of four elements and four clusters of activities. The four elements are (1) the general climate for research use, which reflects the extent to which research funders, universities, researchers and users value and support the linkage between research and action; (2) the production of research, which involves identifying users’ needs, conducting scoping reviews, systematic reviews and single studies to address these needs, and synthesizing the research findings for users; (3) the evaluation of efforts to link research to action, which involves measuring the processes and outcomes of the linkage activities and using the results for improvement; and (4) the mix of clusters of activities used to link research to action. The four clusters of activities are (A) push efforts by producers or purveyors, which are strategies to disseminate research messages to potential users; (B) user pull efforts, which are strategies to enhance users’ capacity and motivation to use research evidence; (C) exchange efforts, which are strategies to foster partnerships and collaborations between researchers and users; and (D) integrated efforts, which are strategies that combine elements of push, pull and exchange efforts. The model provides a comprehensive framework for understanding and enhancing the use of research evidence in policy making.

While the model for linking research to action has been widely adopted and accepted in the health policy arena as a useful tool for understanding and improving the use of research evidence in policy making (139), it is not without its limitations. Despite its many strengths, such as providing detailed guidance on research dissemination activities (137) and accounting for various factors that affect research-policy interactions (137), the model has some shortcomings. For example, it does not specify the most influential factor for research dissemination and use (137) and neglects the design stage of research and other actors besides policy makers (140). Additionally, the model does not make clear how health information is valued or utilized by different policy actors or stakeholders and does not consider contextual challenges or opportunities that may affect the adoption or implementation of research evidence in different settings or situations (141). The model also does not address how funding agencies set their research priorities or respond to the needs and demands of policy makers and other stakeholders (142). Furthermore, the model is partially based on research evidence that has not been evaluated or refined, which may affect its validity, reliability, and relevance for different contexts and situations (143). Critically reviewing the MLRA model contributes to the main goal of assessing frameworks that aim to facilitate the use of research evidence in policy decisions.

A recent model of evidence informed policy making that guides the selection and testing of strategies to increase the use of research in health policy is the SPIRIT Action Framework (43). The authors claim that it guides the selection and testing of strategies to increase the use of research in health policy. The model has four main elements, catalysts, capacity, research engagement actions, and research use. Catalysts are events or situations that trigger the need or opportunity for using research in policy. Capacity is the ability of the policy organization and its staff to value, access, appraise, generate and interact with research. Research engagement actions are the activities that policy makers do to engage with research and apply it to their work. Research use is the outcome of applying research to inform policy making. Research use can be classified into four types: instrumental, conceptual, tactical, and imposed. Research use can occur at different stages of policy making: agenda setting, policy development, policy implementation, and policy evaluation. The model also recognizes that policy influences affect all elements of the framework. Policy influences are external factors that shape the policy context and the demand for research. They include public opinion, media, economic climate, legislative policy infrastructure, political ideology and priorities, expert advice, stakeholder interests, resources, and research.

The SPIRIT Action Framework has several notable strengths, but it also faces some challenges and limitations. One of the strengths of the framework is that it provides a clear and practical guidance for studying the causal pathways between research and policy and testing the effectiveness of different strategies to increase research use in policy (144). Another strength of the framework is that it has a clear aim, a sound knowledge base, a practical guidance, and a knowledge-building potential for improving research use in policy and practice (106). Some of the other strengths of the framework are that it covers the agenda-setting stage, recognizes the diverse roles of research in policy, and provides practical strategies and tools for implementation, such as SEER, ORACLE and SAGE (19). Moreover, it is based on a rigorous and systematic trial of different interventions and measures, which has refined and validated the framework and its tools based on empirical evidence and feedback (145). Furthermore, it comprehensively and realistically depicts the various influences on policy decisions, such as the actors, context, and structure, and how they affect the use of research in policy (146). On the other hand, one of the challenges and limitations of the framework is that it is still under evaluation and lacks published evidence to support its validity and effectiveness. The framework may need further improvement and may face difficulties in being widely used by policy makers and researchers who may prefer more proven frameworks (147). Another limitation of the framework is that it ignores the role of stakeholders in EIPM and only focuses on the use of research evidence in policy (148). Evaluating the strengths and limitations of the SPIRIT Action Framework provides insights into its utility for evidence-informed policymaking, connecting back to the central thesis of analyzing EIPM models in this review.

In light of the detailed description of the all nine models of evidence informed policy making provided above, we will critique all nine models of evidence informed policy making, as well as their implications for policy making practice in this section, from our own perspective. We concur with the existing critiques of these models, but we also recognize their value in providing a framework for using research evidence in policy decisions. Therefore, we also propose that these models need further improvement and adaptation to address their shortcomings.

One of the main limitations of most models, except for FRDU, MLRA and SPIRIT action framework, is that they assume that policy making is a rational process that is informed by research evidence as a priori. However, this assumption may not reflect the reality of policy making in many contexts, especially in low-and middle-income countries (LMICs) where decisions are often made in a centralized and non-transparent manner. Therefore, these models may not be suitable or applicable for such contexts.

Another limitation is that only two models (FRDU, and MLRA) explicitly mention the importance of being aware of the general atmosphere of the policy context, but they do not explain how to act accordingly. Similarly, SPIRIT action framework introduces the concept of “catalyst” as events or situations that trigger the need or opportunity for using research in policy, but it does not provide guidance on how to identify or respond to such catalysts. These models do not address the challenges of time-sensitive or crisis situations, such as the COVID-19 pandemic or the emergence of Ebola, where policy makers may face uncertainty, urgency and pressure to act quickly.

A third limitation is that none of the models consider the complexity of the problem and the approach that is required when dealing with wicked issues. Wicked issues are problems that are difficult to define, have multiple causes and effects, involve conflicting values and interests, and have no clear or definitive solutions (149). Examples of wicked issues in healthcare include obesity, mental health and aging. These issues require a different type of EIPM that is more adaptive, participatory and systemic (150).

A fourth limitation is that most models overlook the role of cognitive shortcuts and social norms in policy making. Cognitive shortcuts are mental strategies that people use to simplify complex information and make decisions quickly and efficiently (151). Social norms are unwritten rules that govern the behavior and expectations of individuals within a group or society (152). Both cognitive shortcuts and social norms can influence how policy makers perceive, interpret and use research evidence in their decisions. For example, policy makers may rely on heuristics (such as availability, representativeness or anchoring) to judge the relevance, quality or credibility of research evidence. They may also conform to the opinions or preferences of their peers, superiors or constituents, regardless of what the evidence says. Only one model (OMRU) implicitly considers these factors in one of its elements (“concern”), but it does not explain them explicitly or provide strategies to address them.

A fifth limitation is that none of the models provide an explicit explanation about the proportion of external factors such as values in the provided solution or policy options, or how much adaptation of research evidence is acceptable or desirable. All models acknowledge that research evidence is not the only factor that influences policy decisions, and that other factors such as values, interests, resources and feasibility also play a role. Some models use different terms to refer to the combination of research evidence and external factors, such as innovation (OMRU and FRDU), policy idea (CHSRF and EIPPP) or adapted version of research evidence (The JBI model, TKPA, MLRA and the SPIRIT action framework). However, none of them specify how much weight should be given to each factor, or how much modification of research evidence is allowed or recommended to make it more adoptable.

A sixth limitation is that all models emphasize the importance of collaboration between researchers and policy makers or engagement of stakeholders in EIPM, but none of them consider the heterogeneity of these groups or provide clear guidance on how to facilitate this collaboration. Researchers and policy makers have different backgrounds, knowledge, experiences and responsibilities, which may create challenges for communication, trust and mutual understanding. Stakeholders may also have diverse and conflicting interests, values and perspectives, which may affect their willingness and ability to participate in EIPM. Therefore, it is not enough to simply state that collaboration or engagement is necessary, but also to explain how it should be done, what strategies should be used, and what outcomes should be expected.

Despite these limitations, the models also have some strengths that are worth noting. For example, OMRU accounts for the user’s perception of innovation or policy idea, which is an important factor for adoption and implementation. The CHSRF Model recognizes the role of research funders as key actors in EIPM, who can influence the supply and demand of research evidence. Three models (TKPA; SPIRIT and EIPPP) incorporate a dynamic view of change theory in their models, which acknowledges that EIPM is influenced change theory. Also, KTA recommends that the knowledge creation and dissemination should be adjusted to the user needs, which is consistent with the principles of knowledge translation. A summary of the strengths and drawbacks of the models, based on both previous and our own critiques, is presented in Table 2.