The concept of tacit knowledge was first introduced by Polanyi (1966; 1967) as a distinction from explicit knowledge. Since then, it has been investigated in diverse fields, such as sociology, management, engineering, and education.

According to Collins (2010, 86), tacit knowledge is ”that [knowledge] which cannot be or has not been made explicit”. Explicit knowledge, on the other hand, is knowledge that can be transferred using strings of information in the right circumstances by elaboration, transformation, mechanization, or explanation. Explicit knowledge can be passed on through intermediaries or through direct contact, whereas tacit knowledge can be communicated only through ”hanging around” with persons who possess this kind of knowledge. However, tacit and explicit knowledge are not mutually exclusive categories. As Collins explains, modes of knowledge are intertwined in practice, and most importantly, the use of explicit knowledge relies on being tacitly understood and applied. Therefore, all knowledge either falls under the category of tacit knowledge or is grounded in it (Polanyi 1966), and it might be more relevant to speak of a distinction between tacit and explicit ”elements of knowledge” (Jensen et al. 2007, 681).

Collins identifies three main reasons for not having made knowledge explicit, resulting in three types of tacit knowledge. The first, strong/collective tacit knowledge is ”the irreducible heartland of the concept” since collective tacit knowledge is the only tacit knowledge that we do not know how to make explicit (Collins 2010, 87, 119). It is knowledge located in society that can be acquired only through socialization and cultural fluency. Collective tacit knowledge enables us to understand social rules and, e.g., negotiate traffic when driving on roads in the real world. This knowledge is acquired through humans being immersed in society.

Medium/somatic tacit knowledge is related to skills and is embodied in the human body and brain. An example often cited is the knowledge of how to ride a bike. While this kind of knowledge can be explained scientifically, it is more commonly learned through trial and error when individuals are in contact with those who possess the skill.

Weak/relational tacit knowledge is knowledge that is not explicit but could be made explicit. It is tacit due to the organization of society rather than the intrinsic nature of the social. For instance, a warehouse worker may know where to find each item without being able to create a list of all items and their positions. Weak/relational tacit knowledge also encompasses concealed knowledge, whether intentionally or unintentionally. This knowledge can remain hidden for various reasons. Blackman and Sadler-Smith (2009) explored the phenomenon of silence in organizational settings. They introduced a conceptual distinction between inability and unwillingness to articulate knowledge using the terms ”silent” and ”silenced”, respectively.

Tacit and explicit knowledge are used differently in practice. According to Collins and Kusch (1995), actions can be categorized as polimorphic or mimeomorphic, depending on the different modes of knowledge they rely on. Polimorphic actions are highly dependent on tacit knowledge. The prefix poly indicates the multitude of behaviours involved in these actions; however, being spelled poli also implies that the rules governing how these actions are performed right or wrong are found in society (polis). These actions are highly variable and depend on the specific context in which they are performed. They require social judgement and are an integral part of professional performance. Polimorphic actions cannot be automated or learned solely through information transmission; rather, they are learned through socialization.

In contrast, mimeomorphic actions build on explicit knowledge. These actions can be reproduced by simply observing or through the transmission of information about the activity, and the behaviour of these actions is not dependent on the context in which they are performed. That is, mimeomorphic actions can be formalized and standardized and are amenable to automation. In practice, however, the polimorphic and mimeomorphic actions are blended, just as the tacit and explicit knowledge on which they are based (Collins and Kusch 1995).

The first dimension used for the analysis of KBM and TBM concerns the modes of knowledge upon which governance concentrates and to what extent it is mostly explicit or tacit knowledge (Figure 1).

A watershed in theories of learning is between the individual and the collective locus of knowledge, i.e., whether the individual or the collective is the primary agent of learning. This is, to some extent, a philosophical debate of mereology, i.e., the study of parthood relations and the question of whether the whole results from its parts or the collective is independent of its parts (Felin and Hesterly 2007). However, the emphasis on learning based on either the individual or the collective as the locus of knowledge will have consequences for how one tries to achieve the positive learning environment that most actors agree they want to achieve (Bahlmann 2014).

Although they focus on different aspects of knowledge, classic epistemology and cognitive psychology both focus on knowledge as a property of individuals (Oeberst, Kimmerle, and Cress 2016). Learning is an inner mental process related to the acquisition and processing of knowledge. In philosophy, the most well-known definition of knowledge is ”justified true belief”, which provides three conditions for knowledge: knowledge is what a person believes, if it is true, and if the belief is justified. The condition of belief points to individuals’ innate cognitive abilities. The conditions of truth and justification point to knowledge as being a representation of the world. To be justified, some kind of justification process must occur.

Psychological theories focus on how knowledge or information can be transmitted between individuals through cognitive acquisition (Oeberst, Kimmerle, and Cress 2016). In educational research, the same focus on cognitive acquisition is commonly seen for traditional teacher-centred education. It relies on the metaphor of ”knowledge acquisition”, which in turn is based on the technical metaphor of transmission from a sender to a receiver (Kullenberg and Säljö 2022). In implementation science, the diffusion of innovation is seminal work that similarly aligns with the idea of knowledge transmission from a sender to a receiver, and an individual locus of knowledge (Rogers 2003).

The understanding of learning as individual cognitive acquisition has been questioned for its validity and universality, and numerous accounts of collective knowledge and learning have emerged. For instance, Hecker (2012) proposed a pluralistic epistemology consisting of three dimensions of collective knowledge: shared and complementary knowledge and knowledge embedded in collective artefacts. It is widely accepted today that groups contribute to the production of scientific knowledge, but whether groups themselves can be considered knowers is a more controversial question (Fagan 2012).

According to Collins (2013, 254) almost all kinds of expertise depend on tacit knowledge that ”can be acquired only by immersion in the society of those who already possess it”. He showed how the learning of polimorphic actions can be distinguished in contexts as diverse as experts exploring gravitational wave physics, car-driving, computer use and talking in natural language. This underscores a learning approach rooted in a collective locus of knowledge.

For approximately 30 years, researchers in organization theory have expressed concern that organizations tend to overlook tacit knowledge due to an overemphasis on individual explicit knowledge (Baumard 1999; Stover 2004; Johannessen, Olaisen, and Olsen 2001; Nonaka and Takeuchi 1995). In educational research, the participation metaphor is often used alongside the metaphor of acquisition to explain learning (Kullenberg and Säljö 2022). According to this view, knowledge is situated in the collective, and learning is a process of participation in communities of practice (Lave and Wenger 1991).

The second dimension for the analysis thus pertains to what kind of learning governance and management ask for – learning as cognitive acquisition based on an individual locus of knowledge or learning through socialization based on a collective locus of knowledge (Figure 2).

The third analytical dimension concerns the differentiation between learning through the creation of new knowledge and acquiring existing knowledge from others. Methods defined as ”scientific” have long held a unique position in understandings of how new knowledge is generated. Consequently, healthcare governance often emphasizes learning in the sense of reproducing this already established knowledge.

However, it has been highlighted for many years now that knowledge from this kind of research is not always used in healthcare practice for various reasons. Implementation science aims to address these challenges of moving knowledge from research to practice by focusing on barriers and facilitators to learning and reproducing existing knowledge. Improvement science is another related research field that recognizes the creation of knowledge based on ”human experience in the real world”, allowing open-ended learning in healthcare practice to progress through the Plan Do Study Act (PDSA) cycle to achieve process-oriented knowledge development (Perla, Provost, and Parry 2013, 179).

The distinction between open-ended learning and learning already existing knowledge has been extensively discussed in research on organizational learning, with ”exploration for innovation” in contrast to ”exploitation of existing technologies” being one example (Brix 2019). Engeström (2007) proposed a related distinction between ”possibility knowledge” and ”stabilization knowledge” to illustrate how learning can either open up for new ideas or reinforce existing ones. Lundvall and Johnson (1994) proposed two modes of innovation, in which Science, Technology and Innovation (STI) mode is based on the production and use of codified scientific and technical knowledge. The Doing, Using and Interacting (DUI) mode relies instead on informal processes of learning and experience-based know-how.

In education, learning is typically focused on reproducing knowledge. However, Kullenberg and Säljö (2022) recently proposed an alternative way of understanding learning even in this context as a process that includes the creation of knowledge. More specifically, they propose polyphony and authorship as two new metaphors of learning. These metaphors originate in assumptions of learning that are not encompassed with the former metaphors of acquisition and participation. They depict learning as dialogic and open-ended, building on the work of Bakhtin, (e.g. Bakhtin 1986) and Dewey, (e.g. Dewey 1938). In the Bakhtinian sense, Kullenberg and Säljö (2022, 547) described learning as a process ”in which learners struggle with understanding the intentions and ideas of others beyond the intent of reproducing them”. They follow Dewey’s distinction between inter-action and transaction, recognizing learning as a transactional activity in which the elements evolve and do not stay the same as in inter-action.

The metaphors of polyphony and authorship also build on the understanding of the modern risk society in which unpredictability and ontological insecurity are inherent aspects of existential conditions (Beck 1992; Giddens 1990). Ontological insecurity encompasses complexity and diversity, as well as uncertainty regarding agency and questions about whose knowledge and experience are legitimate (Kullenberg and Säljö 2022). When scientific research presupposes an ordered reality, it raises the question of how it can make sense of the complexities in real-world practice (Law 2004).

Based on the above, the third and final analytical dimension distinguishes between governance primarily focused on open-ended learning for innovation and governance focused on learning for the reproduction of established knowledge (Figure 3).

Both KBM and TBM aim for evidence-based practice built on three pillars: research evidence, clinical expertise, and patient preferences. As we will see in the following, the KBM and TBM nevertheless emphasize different parts of this triad.

The explicit aim of KBM is to always utilize the best available knowledge (Swedish regions in collaboration 2023), which addresses two related issues. The first issue concerns patients not receiving optimal healthcare based on current evidence, which is a concern for patient safety and can even be a matter of life and death in some cases. The second issue concerns unwarranted disparities in healthcare performance and outcomes, leading to an unequal distribution of healthcare. Although EBM, the early ancestor of KBM, initially emerged to address the first issue, the national model of KBM places even greater emphasis on addressing the latter:

The healthcare system in Sweden maintains a high level of quality. However, the differences in care, treatment, and outcomes are too significant. These differences mean that two people with the same medical condition, for example, risk receiving different investigations, treatments, and follow-ups depending on where they seek care (Swedish regions in collaboration 2023).

As previously described, the focus of KBM organization has been developing an infrastructure that can deliver knowledge supports and guidelines to healthcare professionals. In other words, the focus is on defining and utilizing already existing explicit knowledge, which is primarily built through research. The task for the KBM organization is to compile and weigh together the results from all available research and implement this knowledge in healthcare practices.

A point of departure for the introduction of the national model of KBM was that the utilization of the best available knowledge is important and even more critical when the scientific basis is inadequate. Transparency is an important principle for KBM, especially when scientific knowledge is limited. The scientific knowledge and transparency of other methods for reaching a consensus about the best available knowledge are linked to explicit knowledge. On the other hand, tacit knowledge, which is by definition not explicit, fails to meet the requirements of transparency.

However, despite the focus on explicit knowledge, KBM acknowledges that explicit knowledge is not always directly applicable to specific cases, as evidenced by the following quote:

We should strive for equality and, as part of this, follow guidelines as closely as possible, but it is important to emphasize that ”treating everyone equally means treating them differently”. Good care is about taking into account each patient’s specific situation and conditions. Good care sometimes means deviating from guidelines (SOU 2017:48).

Furthermore, each care pathway includes a section on person-centeredness and patient contract, which emphasizes the importance of adapting to local circumstances and meeting the needs and wishes of the patient. Knowledge is described as a tool for increasing patient participation (Swedish regions in collaboration 2023). However, since care pathways are designed to apply to all patients, they cannot provide guidance to healthcare professionals on how to tailor care to each patient’s specific situation.

The conceptualization of explicit and tacit knowledge explains why. Adaptation to a specific patient is a polimorphic action based on a combination of tacit and explicit knowledge. Tacit knowledge is either not possible to make explicit (strong tacit knowledge) or is not worth the effort to make explicit (weak and medium tacit knowledge). In the case of care pathways, it is easy to see how these already rather encompassing documents would grow substantially and with uncertain benefits if further guidance about the specific adaptations to different patient or context characteristics were included. This is not to say that explicit knowledge does not exist in the local context; only that it is not explicit in care pathways. Furthermore, there is always strong tacit knowledge that by definition cannot be made explicit, as strong tacit knowledge is embedded in every professional culture as well as all human cultures. In brief, KBM is focused on obtaining explicit knowledge for use in healthcare practices.

TBM shares the goal of KBM in that healthcare should be based on the best available knowledge. However, TBM is not focused mostly on how to define and make use of explicit medical knowledge. This is left to others, and mainly the organization of KBM. Instead, TBM highlights the importance of making use of knowledge that exists locally in practice. The supposed neglect of this knowledge was initially a motive for appointing a Trust Delegation, as was one of the messages from the delegation to demonstrate the inherent potential of benefiting from this knowledge:

One of the messages in this main report is that the public sector today does not fully utilize the commitment and expertise found in user-facing professions, nor does it utilize the engagement of users, patients, and students. There is untapped potential here, which has consequences for encounters between employees and citizens (SOU 2018:47, p 241).

Here, both professional knowledge/judgment and patient preferences are highlighted, i.e., the two other legs of the triad of EBM, in addition to the scientific evidence. It is worth noting that the equivalent of these two legs of EBM is consistently named ”knowledge” by the Trust Delegation. Patient preferences and experience are not always referred to as knowledge within EBM or even clinical expertise or judgement. This wider meaning of knowledge, including knowledge about individual values, is explained in the main report:

According to the new perspective, the experiences that an individual carries with them are considered an important source of knowledge for developing different activities and finding interventions that are helpful for them. Not taking into account the individual’s knowledge means losing valuable insights into reality. The same applies, at least to some extent, to the knowledge that user organizations convey (SOU 2018:47, 76).

In line with this understanding, one of the seven principles of TBM is: “Put the citizen’s experience and knowledge in focus and try to understand what the citizen values” (SOU 2018:47, 24).

This kind of knowledge is no use making explicit in knowledge supports as it is as local as it can be. The Trust Delegation focuses on knowledge that is defined locally. This kind of knowledge inevitably includes tacit knowledge, in the sense that it is knowledge that is not made explicit, at least not outside of the local collective. A governance focused on helping professionals listen more to what citizens need and want is focused on obtaining tacit knowledge to use rather than obtaining the already defined explicit knowledge to use. In summary, while the focus of KBM is on explicit knowledge elements, TBM acknowledges the importance of knowledge that has not been made explicit, i.e., tacit knowledge.

However, it is an oversimplification to assert that KBM solely prioritizes explicit knowledge, while TBM solely focuses on tacit knowledge. Notably, one of the primary characteristics of the Trust Delegation’s work is its foundation on scientific research, i.e., explicit knowledge. TBM builds on knowledge from organizational research, including the literature on the consequences of performance management and audit, with references to, for example, Michael Power (1999). This research highlights that the act of auditing itself affects organizations by shaping them to become auditable. Consequently, a significant aspect of TBM is providing professionals with adequate room for action to use their own judgement. The Trust Delegation concludes: “Instead of increasing transparency, openness, and accountability, the audit society risks decreasing the very transparency that drives it” (SOU 2018:48, 45).

While TBM focuses on explicit organizational knowledge, KBM focuses on explicit medical knowledge:

Knowledge-based management within healthcare means that choices and decisions in all parts of the system are based on recommendations developed from the best available medical knowledge (SOU 2017:48, 195).

It can then be concluded that both KBM and TBM rely on explicit knowledge, and these different forms of explicit knowledge can complement each other in certain instances. The organizational knowledge within TBM addresses how to govern healthcare, while medical knowledge focuses on the substantive aspects of healthcare (Dickinson and Pierre 2016).

Nevertheless, at times, a pivotal consideration in governance arises – whether to prioritize the implementation of explicit medical knowledge or the fostering of tacit knowledge. These instances notably present a potential clash between two types of explicit knowledge: organizational knowledge advocating for the nurturing of tacit knowledge and medical knowledge defining optimal medical care.

This trade-off between different modes of knowledge will have consequences for the knowledge processes that are the topic of the next section.

TBM and KBM agree on the significance of a learning environment, but the divergent emphasis on different modes of knowledge will have implications for the assumed and advocated learning processes.

A lack of compliance and adherence to “best available knowledge” has been one of the main challenges for EBM/P, resulting in the emergence of implementation science as a specific scientific field. Implementation science became part and parcel of EBM/P and later also KBM (Nilsen et al. 2022). In 2015, the Swedish government decided to convene a special investigator with the task of considering and submitting proposals for how increased compliance with national knowledge support in health and medical care can be achieved (Dir 2015:127). The investigation concluded, among other things, that there is a need for sharpened national governance and that the national model for KBM is a step in the right direction to more equitable healthcare:

After more than ten years of open comparisons of healthcare outcomes in Sweden, the general picture is that unwanted variations have not decreased to the extent required in light of the statutory goals of providing good care on equal terms throughout the country. The investigation’s assessment is that a sharpened national governance is required to achieve knowledge-based and equitable healthcare. By national, it means such measures or the like that occur throughout the country, i.e., a uniform application (SOU 2017:48, 422)

The problem that KBM is to solve is that healthcare practices do not work according to what is known from research, i.e., the gap between research and practice, which also presents as a problem of inequity. Whether the problem is defined as a lack of guidance or an implementation problem, the focus is on learning explicit knowledge. KBM’s primary focus is on developing knowledge supports, which involves defining knowledge, aims, and outcome measures. These measures enable the evaluation of compliance with the knowledge supports. Transparency and evaluation are considered essential tools for learning:

The processes and results of healthcare need to be followed up and fed back at various levels within the system. Transparency and open comparisons increase opportunities for learning (Sveriges kommuner och regioner 2022)

In healthcare practice, learning is understood primarily as transmission from the sender of the KBM organization to the receiver in healthcare practices. The vision of a national model of KBM is ”to create a learning and supportive system for healthcare” (Sveriges kommuner och regioner 2017). This supportive system has thus far come mostly in the form of documents that include explicit knowledge. How learning occurs in healthcare practice is largely left to the regional public authorities to define. However, with a focus on explicit knowledge and standardized actions according to care pathways with defined indicators for follow-up, the cognitive aspect of learning is implicitly emphasized.

TBM does not contradict the significance of implementing explicit knowledge. However, the focus has been on different approaches to improving practice. The focus is on professionalism and needed adaptation to the patient and the local circumstances. As the Trust Delegation writes in the main report:

Here, a different logic prevails than where problem solving is facilitated by standardization. Instead of standardizing a process and streamlining it by reducing the variation in effort, as is possible, for example, in knee replacement surgery, this is about varying the effort to achieve less variation in outcome (2018:47, 166).

The skills needed to adapt measures to diverse needs are widely acknowledged as not learnable through reading textbooks or care pathways alone, as they rely heavily on tacit knowledge. Therefore, professionalism and collegial learning are key principles in fostering a learning environment based on TBM (2018:47). Professionalism and collegial learning are based on learning through participation and socialization rather than just cognitive acquisition. The desired professional approach is achieved through exposure to patients and other professionals, i.e., the locus of knowledge is in the collective rather than the individual. The Trust Delegation recommends:

Managements and employees within the municipal sector should collaboratively establish an overview of the operations to identify how arenas for collective, collegial learning can be created within and beyond various operational and professional boundaries. The work should be organized to allow both regular and spontaneous opportunities to discuss issues and transfer knowledge and experiences, to thereby learn from each other. Establishing such working methods and a strong learning culture requires active and engaged leadership (SOU 2018:47, 339).

It is worth reiterating that there is no inherent contradiction between TBM and KBM in this context. However, they diverge in focus. KBM directs attention to the content to be acquired through learning, whereas TBM underscores learning as a collective process, emphasizing the necessity of collegial learning, professionalism, and cooperation.

KBM and TBM both highlight the significance of knowledge growth in terms of innovation and the creation of new knowledge, albeit with different focuses. In KBM, the innovative form of learning is primarily conducted through research and in specific groups of experts who are assigned a mission to reach consensus about the best available knowledge, including in cases where the scientific basis is scarce. These practices prioritize innovative and collective learning. However, with KBM’s focus on producing care pathways, including indicators for evaluation, the expectations of healthcare practices are consequently focused more on reproductive rather than innovative learning.

In contrast, TBM proposes that innovative learning is situated in work practices, organizations, and new forms of collaboration. Innovation is linked to the concept of trust, as it must be expected that new goods and services will be invented when actors meet in processes characterized by mutual respect and reciprocity: “The traditional monopoly of the scientific community on knowledge creation is being challenged, and more situationally adaptive knowledge production is emerging” (SOU 2018:38, 228).

The second report from the Trust Delegation is a research anthology with contributions from 23 researchers providing their experiences from 12 case studies (SOU 2018:38). These projects all share some features. These are not the kinds of studies that would be ranked as being of high scientific quality for a systematic review according to EBM. None of them used a control group; many of them did not have specified aims or outcome measures from the start, and they were not implementation projects either. The focus is rather on collaboration with the ambition to make the most for the citizen. The process includes creating new knowledge rather than implementing already defined knowledge. Innovation and continuous improvements share the feature that knowledge is not predetermined. Here, it is not just about knowledge being tacit; knowledge might not exist at all yet. The Trust Delegation states that there is a need for renewal and innovation to spring directly from healthcare practices, which involves the following:

[…] a clear ambition to include an employee perspective and to make the staff closest to the patients, as well as the patients themselves, experts on what changes and improvements need to be made to increase the quality of care (SOU 2018:38, 105-106).

Another aspect of a focus on innovative rather than reproductive knowledge is that established knowledge can be questioned. In one of the projects, a Scottish school is highlighted as a role model:

Principals and teachers willingly acknowledged that they did not follow the curriculum or the knowledge requirements strictly. It was considered much more important to follow the child and his or her needs and then integrate the curriculum based on the student’s strengths (SOU 2018:38, 288).

The role of civil servants in TBM explicitly encompasses the practice of re-evaluating prior knowledge. The Trust delegation clarifies that the term ”civil servant” encapsulates the ethical expectations governing the conduct of public employees. As stated in the quote:

Ultimately, this can mean that the civil servant must openly sound the alarm when malpractice occurs. Through this important role for employees in publicly funded operations, corruption can be prevented, and trust in the employee is therefore a central part of this process (2018:47, 248).

TBM focuses on granting greater autonomy to individuals – patients, professionals, and civil servants – enabling them to act according to their own judgment and find practical solutions. Consequently, TBM emphasizes open-ended learning rather than the reproduction of already established knowledge. This approach contracts with KBM’s primary focus on utilizing knowledge that has been established in other contexts, particularly research practices and by representatives of the national organization of KBM.