Article Text

Download PDFPDF

Enacting medical leadership to address wicked problems
  1. Wouter Keijser1,2,
  2. Jo-Louise Huq3,
  3. Trish Reay4
  1. 1 Faculty of Behavioral, Management and Social Sciences (BMS) Change Management and Organizational Behaviour (CMOB), Universiteit Twente, Utrecht, The Netherlands
  2. 2 DIRMI Institute, Utrecht, The Netherlands
  3. 3 Haskayne School of Business University of Calgary, Calgary, Alberta, Canada
  4. 4 Strategic Management and Organization, University of Alberta, Alberta School of Business, Edmonton, Alberta, Canada
  1. Correspondence to Wouter Keijser, Faculty of Behavioral, Management and Social Sciences (BMS) Change Management and Organizational Behaviour (CMOB), Universiteit Twente, Utrecht, The Netherlands; wouter{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


This paper argues that although physicians’ established ways of working make them excellent clinicians, their education and experience make it difficult for them to address wicked problems. After defining wicked problems and illustrating them within healthcare contexts, we explain why physicians’ institutionalised and professionalised ways of working leave them underprepared to address wicked problems. We then describe the emerging concept and frameworks of medical leadership (ML) and show how ML training could provide physicians with the skills and abilities they need to enact leadership in collaborative environments. We also explain how ML could help physicians become more comfortable in ‘grey-zone decision-making’ that is needed to address current challenging problems. Finally, we draw attention to changes that are warranted outside of ML training to bolster its potential: incentivise training and reregulate top–down, empower the professional pipeline and create bottom–up opportunities.

Physicians and leadership

Physicians face many challenges as part of their day-to-day professional work. Their clinical skills serve them well in many ways. But when faced with wicked problems they are generally ill-prepared. Moreover, physicians’ training and established ways of working can result in physicians unwittingly becoming part of the problem when addressing wicked problems.

Wicked problems are those fundamental, challenging problems that exist within and between social sectors, are not solvable through linear planning or the application of causal models and tools, have no definitive problem formulations or solutions, and are impacted and changed when their intended solutions are implemented.1–3 Wicked problems are ‘complex, intractable, open-ended and unpredictable’ such as ‘global warming, drug abuse, child protection or natural disasters, … (the) safety of nanotechnology or growing numbers of refugees’ (see Head and Alford, p397).3 In healthcare, problems related to mental illness, palliative care, healthy ageing, sexual health and cancer care have all been classified as wicked.4–7

Because of their nature, addressing wicked problems requires the engagement of many stakeholders, each of whom can hold different, value-laden understandings and favoured solutions.1 8 9 As a result, healthcare reform and transformation and change management literatures point to the importance of bringing different stakeholders together to cocreatively reorganise healthcare services and service delivery so that wicked problems are governed and addressed more effectively.5 10 11 Collaborative approaches that change organising structures and processes can certainly improve our ability to address wicked problems; however, more lateral-type leadership from front-line professionals, such as physicians, is needed to change how day-to-day work is accomplished to better address wicked problems.12

In this article, we examine how physicians’ education, position, practice and approach to decision-making makes them excellent clinicians, but at the same time can inhibit their ability to lead initiatives designed to address wicked problems. We draw attention to the growing medical leadership (ML) movement, which encourages physicians to take training in areas such as human-factors design, organisation, innovation and change, which are not traditionally part of physicians’ practice. We then ask and explore the question: How can ML help to equip physicians to address healthcare’s wicked problems and what else is needed?

Wicked problems in healthcare

The concept of wicked problems originated in the policy planning literature1 13 to describe social issues that cross sectors and which cannot be understood and addressed in isolation. They are problems whose formulations are often ‘grounded in value perspectives’, meaning that stakeholders hold value-laden understandings of the issues and propose solutions based on these values; consensus across stakeholder groups is unlikely to be achieved by gathering more (eg, scientific) information [p3].14 Because wicked problems are inherently ‘ill-defined’ and imbued with political and value-laden judgements rather than ‘scientific certitudes’, they are ‘resistant to a clear definition and an agreed solution’ [p3].14 Furthermore, wicked problems are such that solutions proposed are often associated with better-or-worse options, and the introduction of these ‘solutions’ leads to outcomes that emerge over time, resulting in new challenges that can impact the problem itself.1

Jointly, these characteristics make some problems ‘wicked’ as compared with those that are more clearly definable or ‘tame’, and to which ‘linear’ and verifiable solutions can be applied [p160].1 Given the characteristics of wicked problems, solutions need to be designed around robust actions that support sustained engagement in ways that are non-committal and that keep future lines of action open.15

Ferlie et al, for example, described how networked arrangements could be helpful in addressing wicked problems because networks allow professionals to engage in shared leadership, interprofessional problem-solving and continuous collaborative change.5 16 However, research has revealed that attempts to create networks can foster problems arising from perceptions of lost (or diminished) professional autonomy and resistance to shared forms of governance.5 This resonates with the notion that addressing wicked problems requires deep knowledge and understanding of embedded social and cultural legacies that impact people’s daily interactions. In table 1 (column 2), we summarise approaches to addressing wicked problems. Throughout this paper, we refer to table 1 to explain related concepts.

Table 1

Four-dimensional overview of approaches to address wicked problems, traditional ways of working, ML competencies and selected challenges

Health issues that could be considered wicked are those that arise not only from physiological (and possibly medically curable) factors, but also from continuously changing environmental, social and organisational factors and political turbulence from the involvement of multiple parties (eg, public, voluntary, private, charity, social and medical care) with conflicting values and interests.3

For example, in healthcare, cancer care, sexual health and healthy ageing have all been classified as wicked problems.5 The wickedness of cancer care is visible in multifaceted efforts to improve treatment and services. These efforts involve encouraging behavioural change among citizens; corporations (eg, tobacco, alcohol, food and other commerce); preventive health services (eg, screening to ensure early diagnosis); and the biomedical research industry, each of which brings different values and belief about cancer and cancer prevention to the fore. Organisations promoting sexual health must deal with stigmatisation or isolation of their treatment clinics and potential resource scarcity because their clients’ behaviours (eg, unintended pregnancies, sexual transmitted infections) are deemed improper or inappropriate relative to broader value and belief systems. The ageing population challenges health and social systems in many countries to organise more pluriform and aligned care for older citizens. Determining appropriate strategies for elder care can be classified as a wicked problem because living independently in communities stands in contradiction to commonly held negative attitudes toward elderly people, ageist stereotyping and healthcare systems’ traditional focus on ‘cure and rehabilitation’ when various other foci and goals may be more relevant for aged persons [p119-120].5

Addiction to drugs or alcohol, is another wicked problem shaped by multiple dimensions3: the addiction itself (a condition characterised by ineffectiveness of any rational interventions); drug trade (including production and availability); social determinants of health; trauma; and even, some suggest, capitalist society itself.17 18 Additionally, from the healthcare perspective, people who are addicted to substances present with a wide array of physical and behavioural concerns. Approaches in advancing care for people coping with addiction, thus, requires attention to social, physical, psychoemotional, judicial and other aspects of their lives, and necessitates involvement of diverse professional and social supports.3 In addition, as is common with wicked problems, some treatment efforts may have negative and unanticipated effects on other efforts.

Finally, Periyakoil classifies palliative care as a wicked problem; she describes a study designed to improve ‘end-of-life decision making’ and ‘reduce the frequency of mechanically supported, painful and … prolonged process of dying’[p658].4 During the study, communication problems among different professions arose and resulted in an increased frequency of aggressive treatments. To (re)solve this problem, nurses were trained in communication to facilitate patient–physician discussion around advanced care planning.4 This solution then exposed aspects of the approach to palliative care that were ‘deeply and insidiously rooted in the culture of modern biomedicine’ [p658].4 Periyakoil’s description shows different ideas of what palliative care should look like, how it should be provided, and how standard, linear (biomedical) approaches can fail to address and incorporate patients’ ideas and values, even though the study’s approach to the provision of palliative care was specifically designed to involve both patients and medical professionals.

Established ways of working

The education, training and enculturation of physicians and other healthcare professionals produces a lifelong imprint on their ways of working, which are often resistant to change, especially in later career stages.19 20 Extensive clinical education and training contribute to processes of professional socialisation and identity formation that create consequences for professionals’ organisation of work and leadership approach, problem-solving and reasoning methods, accountability orientation and governance paradigm.19 Considering each of these four domains (table 1, column 1), we describe how they are reflected in physicians’ established ways of working (see also: table 1, column 3).

Organisation of work and leadership approach

Despite a variety of ongoing changes and attempts to reform healthcare, ever since Hippocrates of Kos, founder of contemporary Western medicine, healthcare has predominantly been organised in a physician-centric way. Physicians’ current ways of working reflect their historical professional status and stance towards others. Physicians, to a high degree, function independently as sovereign experts located at the top of a professional hierarchy vis-a-vis other healthcare providers to apply their distinct medical knowledge to diagnose and solve health problems.21 22 With their sovereignty in the art of medicine (ie, performing medical diagnosis and treatment within their patient-physician relationship) and exclusive knowledge,23 physicians bear the heavy-weight, end-responsibility of the majority of patient-related decisions and processes.14 As a result, daily medical routines are often tightly connected to a ‘doctors’ orders’ paradigm, which is illustrated in the way physicians delegate work to other professions.5 21 24 25

Problem-solving and reasoning approach

Medical preclinical training and compulsory continuous education produces physicians who are highly skilled in scientifically oriented problem reasoning and solving approaches. As Abbott21 showed, physicians and other professions approach their day-to-day problem-solving work through a process of diagnosing, treating and inferring, and through decision-making informed by verifiable scientific methods (eg, blood tests in well-calibrated laboratories).21 Generally speaking, the practice of medicine relies on these complex but mostly linear processes of collecting information about patients’ healthcare concerns, applying clinical evaluation and highly systematised scientific reasoning, investigating treatment options, applying treatment protocols and monitoring progress. These are all based on standardised, mostly quantitative, clinical parameters.21 26 27 Typically, the ‘medical model’ suffices to navigate ‘critical problems’, which demand instant action without time for pondering or procedures (eg, acute myocardial infarction) and ‘tame problems’, which can be challenging but are likely to be resolved (eg, open-heart surgical procedures).28 This well-established evidence-based way of doing clinical work informs and creates physicians’ professional identity, which is strongly grounded in ‘this is how we do things here’.29 30

Accountability orientation and governance paradigm

As experts, physicians are taught and trained to make independent judgements about what problems they solve and how. Although physicians are increasingly obligated to respond to bureaucratic rules and regulations, in their professional practice, physicians turn to peers for recognition and evaluation of their work.31 32 Importantly, peer assessment and evaluation in the medical profession relies on medical colleagues’ assessment of appropriate technical expertise while avoiding (quasi-)normative judgements about other’s work.33 From a governance perspective, the medical profession is highly represented and informed by professional associations. At (inter)national levels, these bodies are powerfully involved in establishing legislation and regulation that describe and delineate physicians’ professional duties and occupational scope. Moreover, medical associations facilitate the standardisation of work among specialisations and help physicians to organise and control their work and their professional position.31

A medical antagonism?

We argue that physicians’ (and other professionals’) established ways of working create challenges in their ability to engage with healthcare’s wicked problems. We debate that these challenges largely reside in the intertwined trias of physicians’ knowledge, position and, ultimately, power.

Physicians’ sophisticated, linear and biomedically evidence-informed approaches to problem solving and their scientific, clinical methods of diagnosis, reasoning and treatment21 lies in juxtaposition with the more experimental, emergent and action-oriented approaches advocated for navigating wicked problems.1 15 34 35 For example, approaches to engaging with wicked problems have been described as ‘distributed experimentation’ that incorporates robust action and leaves open the option for other approaches and action, which together can result in clumsy types of solutions.15 36 37 In contrast to applying clinical and scientifically derived end solutions, addressing wicked problems involves steering and coping with solutions that often create new problems.4 As Kyratsis et al argue, physicians’ deep allegiance to evidence-based medicine might prevent other potentially valuable, credible and relevant evidence and viewpoints from being considered, including the ‘experience, personal knowledge and expertise, perspectives and preferences of stakeholders, policy mandates and endorsement, and evidence from the local context’ [pXXIV].38 Unlike conventional medical problem solving, wicked problems are not the best addressed by applying a ‘technology of guidelines’ or through a common frame of reference.5 Instead, their ‘clumsy’ solutions often lie in a collective approach and a stance of ‘let’s just start, try and see’, which in many cases contradicts with medicine’s most prominent paradigm of ‘primum non nocere (‘first do no harm’).

Physicians’ clinical work orientation, problem solving and sovereignty are deeply engrained, even taken-for-granted and healthcare’s widespread physician-centric practices and routines can be relatively impermeable to change.25 Societal-level norms and belief systems, as well, affirm physicians’ unique position, placing them as unimpeachable, all-knowing healers in a ‘doctor-knows-best’ certitude. Additionally, the predominantly medically controlled, evidence-based paradigm that dominates much of healthcare’s contemporary practices, guidelines and regulations reinforces the ‘medical model’, which contributes even more to constituting its owners’ (ie, physicians’) powerful position, also mirrored in their exclusive ‘licence to treat’ by law: only physicians are authorised to ‘perform’ medicine. Consequently, in contrast with, for example, the nursing profession, physicians typically have more access to authoritative discourse on diagnosis and treatment.39 This knowledge-position-power nexus governs many of healthcare’s social relationships and work processes and has positioned professions that ‘own’ certain knowledge areas as ruling over other areas, for example, psychiatry over mental care; public health over preventive health.5 Although more inclusive and collective ways of working, alternating between leading and following, and at times leading from ‘back seat’ positions,30 could improve physicians’ engagement with wicked problems, they are very likely to challenge physicians’ unique and powerful professional position, autonomy and self-regulation.40

ML and wicked problems

Recently, the medical profession has started to incorporate leadership competencies aimed at enabling physicians’ to become leaders in health system transformation and change.41 Several countries now have national-level ML competency frameworks that describe new non-clinical skill sets recommended for physicians including interpersonal and teamwork skills; organisation and management skills; quality and innovation skills; and skills that encourage cocreation and entrepreneurship.42 ML training programme endeavour to bolster physicians to become more proficient professionals, beyond their roles as healers,40 enacting leadership qualities that foster engagement between multidisciplinary professionals in collaborative practice and transformation.42–45 The swelling number of ML training opportunities (in preclinical and postgraduate education) indicates a growing interest among physicians and their associations, as well as among administrators, policy-makers and educators.44 46 47 Meanwhile, a steadily increasing body of evidence denotes ML’s beneficial effects on, for example, quality and safety, sustainability of innovations and employee well-being.42 48

Scrutinising its discourse, we suggest that ML’s emergence holds relevance for physicians’ (potential) leadership role in addressing healthcare’s wicked problems. Moreover, it signifies physicians’ intentions in ‘opening-up’ and exposing themselves more to others, in particular in their roles as boundary spanners[35 ,49p116, 50]. Building on the content of five national ML competency frameworks (table 1, column 4), we illustrate how various ML competencies are supportive of approaches needed in navigating wicked problems. For example, physicians enacting ML through ‘leading professionals’, ‘engaging others’ and ‘developing coalitions’ should contribute to high(er) levels of trust and relational coordination and the emergence of networks among pluriform groups, which should help navigate wicked problems through collaboration and reciprocity in collective decision-making5 51 52 (see table 1, column 2). Enhanced interconnectedness between physicians and others is also likely to be further advanced through ML competencies, such as, ‘self-reflectiveness and self-development’; ‘personal leadership’; ‘demonstrating personal qualities’ and ‘personal development’ (table 1, column 4).

However, we argue that despite such promising notions, ML’s potential may be challenged because of the difficulty associated with changing some of physicians’ established ways of working, as we described earlier. Although ML training is a positive and encouraging step, it alone is unlikely to easily facilitate change in physicians’ behaviours and daily practices.35 In table 1, column 5, we distinguish the challenges that may continue to limit physicians’ ability to engage with wicked problems, even as they build competencies in ML.

Bolstering ML’S potential

We now draw attention to additional supports that hold promise to bolster ML’s potential. We recognise that these supports may not be needed to enable some physicians’ enactment of ML competencies in addressing wicked problems. However, we believe that these supports will encourage broader attention to the importance of ML competencies and will help establish new, beneficial practices in contemporary clinical work in order to effectively address wicked problems.

Incentivise and reregulate top–down

Existing professional jurisdictions and payment schemes, rooted in system-level directives and legislation, tend to incentivise the status quo of hierarchical and fragmented care, thus disenfranchising the potential effects of ML in addressing wicked problems. We argue that a rethinking and redesigning of legislation and regulation on financing (ie, professional compensation), professional jurisdictions and accountabilities could help reduce perverse triggers that sustain established ways of working and which hamper the creation of new ways to address wicked problems.35 53 Altered payment and auditing schemes, for example, could open avenues to invest in ML training and certification. However, reworking of incentives and regulations must be conducted with deliberation of the values and beliefs that underpin physicians’ and other professions’ practice in particular contexts (eg, countries and healthcare systems).54

Empower the professional pipeline

In healthcare professional norms and behaviours are almost endlessly iterated, starting in the early education of physicians and other health and social care professionals. Two alterations in traditional medical education, we argue, would provide opportunities to foster ML in better addressing wicked problems.

First, physicians’ primarily unidiscipline educational structures and mentor-apprentice approach inculcate normative traditions that govern physicians’ professional identity and behaviours in daily work and interactions.55 Early and ongoing exposure to multidisciplinary oriented education (including collective problem and solution formulation) holds potential to restructure roles and responsibilities and instil a more open-minded and collaborative approach to problem analysis and decision-making.52 56–58

Second, physicians are principally and continuously taught to avoid creative or multidirectional solutions for problems: often wisely, to prevent inaccuracies and errors in clinical work. Inevitably, this challenges their ability to engage with wicked problems. Deliberately expanding physicians’ repertoire of reasoning strategies beyond biomedical rational could invite new ways of thinking and acting.35 49 52 Therefore, we suggest that physicians should be educated more in navigating problems that ‘require adaptive solutions that are tailored to work in the local setting and need to be implemented by a group of local stakeholders and champions who are well acculturated in their organisational culture’[p658].4 Incorporating curricular content entailing, for example, plan-do-study-act principles may be one way to help physicians’ shift to more iterative and emergent decision-making and treatment approaches.59

Create bottom–up opportunities

Despite their often limited authority over physicians’ work,22 25 30 60–62 managers and administrators may also be able to facilitate and encourage physicians’ enactment of ML in the context of wicked problems. In particular, managers could create ‘spaces’ (eg, improvement projects) to provide opportunities for physicians to employ principles of ML and engaging with other professionals to experiment with collaborative approaches based on different problem formulations.24

Managers and clinical professionals could find ways to protect such spaces from reiterating norms that govern status quo and to work through the conflicts and tensions that can arise when different professionals collaborate to develop creative solutions.22 35 61 In these spaces, management could consider applying a ‘hands-off’ approach and avoid intervening quickly with controlling measures which have the potential to reinforce interprofessional differentiation even as they facilitate the emergence of different ideas and problem formulations [p221].4 5 7 Finally, managers and leaders could also encourage interorganisational learning and networking, and could link to the efforts of educational institutions, professional associations and legislators/regulators to scaffold local efforts to macro forces, ultimately to help prevent professionals from falling back into old habits.


Finding new ways to approach wicked problems in healthcare is important and requires physicians’ involvement. In this paper, we drew attention to physicians’ established ways of working and showed how these create significant challenges for physicians’ meaningful engagement with wicked problems. We also showed that emerging ML frameworks emphasise physicians’ potential to develop competencies to help them collaborate and engage with others in the areas of coaching, continuous improving and leading for innovation and change. National ML frameworks point to physicians’ beneficial contributions to addressing wicked problems.

Additionally, however, enabling, incentivising and supporting physicians through different governance models, new educational formats and organisational support would bolster the promise of ML and could reinforce physicians’ ability to become effective partners in the multidisciplinary and cross-sectoral problem solving needed to address wicked problems.

We recognise that such changes require shifts in not only physicians’ but other healthcare professionals’ also deeply institutionalised ways of working. Nursing and allied health professionals’ similarly enshrined ways of working, which include following ‘doctors’ orders’, contribute to the perpetuation of physician centredeness in healthcare.63 With adequate supports, wicked problems such as those mentioned in this paper, as well as those that may arise on the horizon, such as robotics and artificial intelligence,64 must be engaged with collaboratively and flexibly by all. Ultimately, significant change relies on all actors, including professionals, managers, legislators, patients and, eventually, society-at-large, redesigning their views and values regarding how our healthcare services are created and delivered.


We thank the Academy of Management for hosting the 2018 subconference on which this paper was initiated, as well as to its cocontributors, Graeme Martin, Peter Lees and Jamie Stoller and the subconference’s attendees. We thank the reviewers and our editor, Amit Nigam, for their valuable and constructive support and suggestions during the development of this paper.



  • Twitter @WouterKeijser

  • Contributors WK provided the concept and idea of the article. WK and J-LH contributed equally to consecutive drafts, which were reviewed and edited by TR for important intellectual content, with continued input from WK and J-LH. All authors approved the final version.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.