“It’s okay to not know …” a qualitative exploration of faculty approaches to working with uncertainty

Background Whilst it is recognised that a capacity to manage uncertainty is an essential aspect of working as a healthcare professional, there is little clear guidance on how to facilitate student learning in this domain. A lack of faculty development opportunities also suggests that health professions’ educators may feel ill-equipped to assist students in developing effective approaches to uncertainty. The purpose of this study was to explore a faculty development intervention designed to help educators unpack students’ experiences of uncertainty, and identify attributes which may help students to manage uncertain situations. Methods This qualitative study was informed by a constructivist methodological approach, where participants were encouraged to share meaning around the nature of uncertainty in health professions’ education. Two 90-min faculty development sessions were held. These sessions invited participants to apply Han et al.’s taxonomy of uncertainty to role-played scenarios of student uncertainty within a focus group setting. Focus group data were collected, and examined using a two-stage, hybrid approach of deductive and inductive thematic analysis. Results Han et al.’s taxonomy helped participants to identify multiple sources and issues of uncertainty in the role played scenarios, thus unveiling the extent of uncertainties encountered by health professions’ learners. Data analysis revealed four themes overall: “Sources of uncertainty”, “Issues of uncertainty”, “Uncertainty attributes”, and “Learning environment.” Participants also contributed to a list of attributes which they considered helpful to undergraduate health professions’ students in managing uncertain situations. These included an awareness of the nature of uncertainty within healthcare practice, an ability to recognise uncertainty, and adopting attitudes of adaptability, positivity, and resilience. Conclusions This study highlights the successful use of Han et al.’s taxonomy of uncertainty within a faculty development setting. Our findings suggest that the taxonomy is a practical and versatile tool that health professions’ educators can use in shared reflections and conversations around uncertainty with students or colleagues.

However, whilst a capacity to manage uncertainty has been recognised as a core dimension of professional competence [9], and is now a regular addition to professional competency-based frameworks [10][11][12][13], there is no consensus on how to support effective learning around uncertainty. A recent scoping review reveals that although health professions' students meet uncertainty regularly within the context of their undergraduate training, they appear to receive little, if any, formal training on how to manage this [14].
Uncertainty can be understood as a "subjective perception of not knowing what to think or what to do" [15]. This is a frequent experience for health professionals which can influence clinical decision-making and professional practice. Yet, there are few published examples of teaching interventions which specifically address uncertainty management. One explanation may be the lack of clear evidence which links training around uncertainty to explicit, measurable and positive outcomes. White and Williams [16] state that although there is a "substantial body of evidence in support of the implementation of formal teaching regarding uncertainty... There have been no trials on which to base judgements about the longterm effectiveness, outputs, value for money and beneficial effects on practitioner resilience and performance [of this teaching]. " This situation may be about to change. Although research into uncertainty in health professions' education has existed for more than half a century [17], more recent work has built a compelling case for using educational interventions to influence learners' uncertainty management [18][19][20]. Stephens et al. [21] state that "education may be a formidable moderator of tolerance of uncertainty, with multiple aspects of the learning environment impacting student tolerance of uncertainty. Therefore, educators should feel confident in trying to incorporate tolerance of uncertainty paradigms into existing curricula, even traditionally content-heavy science courses. " Returning to our opening question around teaching interventions, perhaps the word "should" holds a clue? Although educators should feel confident to offer teaching around uncertainty, it is distinctly possible that they do not.
A recent increase in research around uncertainty in health professions' education has also revealed subtle but important shifts in thinking [14]. Whilst older studies have focused on how to raise learners' tolerance for ambiguity, newer work hints at a more nuanced balance between tolerance and intolerance [22]. In other words, educators face ambiguity in planning how to help their students to face ambiguity. Further to this, educators may also view uncertainty, overall, with some trepidation. As Hillen et al. [23] explain: "Uncertainty can be aversive; large bodies of research from multiple disciplines, both in and outside of the health care domain, have demonstrated that uncertainty provokes fear, worry and anxiety, perceptions of vulnerability, and avoidance of decisionmaking. " These factors combine to explain why educators may not feel confident to offer teaching around uncertainty or, indeed, to disclose the full extent or nature of their own uncertainties.
Despite these many potential reasons for health professions' educators' reluctance to engage with uncertainty, it seems clear that they act as an influential presence in helping learners to cope with uncertain situations, e.g. through role modeling and mentoring [14,24]. It is likely, then, that educators may need assistance to reach their potential here, making the ways that they themselves respond to uncertainty explicit and tangible as they guide students. As Domen [25] states, "a greater emphasis should be placed on the teaching of ambiguity to residents and faculty who, ultimately, have the greatest influence on the qualities and behaviors we hope to instill in our students, residents, and other learners. " There is, however, a surprising lack of research into faculty development around strategies for uncertainty [26,27], and little guidance on how to equip faculty to recognise and engage with students' experiences of uncertainty. It is reasonable to consider that effective faculty development interventions may empower health professions' educators to notice and harness opportunities to support learning, both formal and informal, around uncertainty.
The purpose of this study is to explore a faculty development intervention designed to help educators unpack simulated experiences of student uncertainty. Using role-played vignettes derived from real-life experiences to trigger discussions around uncertainty, we aimed to deepen educators' understanding of where uncertainty manifests within health professions' education, and how students can be helped to manage its accompanying challenges. With regards to the latter, we wanted to gather educators' perspectives on the student attributes (knowledge, skills and attitudes) that would represent foundational competence in managing uncertainty [28]. By specifically defining the attributes that make up the construct of uncertainty management and tolerance, we hoped this would make both implicit and tacit learning around uncertainty more explicit, as well as providing educators with a framework for supporting students when they reflect on uncertain situations.
Our faculty development intervention employed Han and colleagues' [29] taxonomy of uncertainty as a conceptual framework. This taxonomy organises experiences of uncertainty according to three dimensions: source, issues and locus (Fig. 1). The locus of the uncertainty is the person to which the uncertainty relates. The sources of uncertainty are where the uncertainties arise from. These can be categorised as probability (the indeterminacy of a future event occurring), ambiguity (the lack of adequate, reliable or credible information) or complexity (aspects of the situation that make it difficult to understand). The issues of uncertainty are the substantive issues about which an individual is uncertain. These can be categorised as scientific or data-centred (uncertainty related to a medical condition), practical (uncertainty related to the system, structures, or processes) or personal (uncertainty related to the individual). We chose this taxonomy as it has been demonstrated as a practical way to facilitate "an organized approach to the problem of uncertainty" [29] in a wide range of healthcare contexts [30][31][32][33][34][35][36]. Using this taxonomy also allowed us to situate our study within the wider body of literature on uncertainty in healthcare, helping to contribute to "a more systematic program of research based upon shared, integrative conceptual models" [37].
Here, we used the taxonomy to help faculty participants to "distinguish and understand … specific uncertainties" [33], and reflect on how uncertainty arises and unfolds in health professions' education. Through applying the taxonomy to role-played scenarios, we reasoned that participants would gain deeper insight into what attributes could help undergraduate students to manage uncertain situations. Our specific research questions were: • "In what ways does the use of Han et al. 's taxonomy of uncertainty support health professions educators' understanding of this concept when used experientially to support faculty development?"; and, • "What do health professions' educators consider as key attributes for undergraduate learners with respect to managing uncertainty?" To the researchers' knowledge, it is the first time that that this conceptual framework has been used in an educational setting to generate a deeper understanding of uncertainty in a context of faculty development.

Study design
We used a qualitative study design which was informed by a constructivist methodological approach, where understanding about uncertainty was co-constructed by participants, using simulated scenarios as triggers for conversation and exploration of this concept [38]. This study is reported in accordance with O'Brien et al. 's [39] Standards for Reporting Qualitative Research.

Intervention
The faculty development session was underpinned by a social constructivist theory of learning. Using this approach, participants were encouraged to share meaning around the nature of uncertainty in health professions' education. The specific instructional design of the intervention was inspired by Armitage-Chan and Whiting's [40] use of role-play to provoke faculty discussion around student professionalism learning outcomes, with an overall objective of helping attendees to "practice, reflect on, and develop important skills in a predictable and safe learning environment".
Each session was 90 min in length and attendees were notified of two learning outcomes, i.e., that, by the end of the session, they would be able to (i) analyse uncertain situations using a recognised taxonomy of uncertainty, and (ii) identify attributes that could help prepare undergraduate health professions' students to navigate uncertain situations. There were three components to each session: opening plenary (30 mins); focus group work (30 mins); and final whole group work (30 mins).
In the opening plenary, JM addressed contemporary definitions of uncertainty and highlighted early findings from a scoping review which explored how undergraduate health professions' students learned to engage with uncertainty which had been carried out by three members of the research team (JM, JH and TP) [14]. This was followed by a description of Han et al. 's taxonomy of uncertainty [29].
Next, the participants were invited to engage in a focus group activity. Participants had been pre-assigned to a group by facilitators according to profession, so that each focus group had a multi-professional profile. Group sizes ranged from four to six participants. In these focus groups, participants were presented with two roleplayed scenarios which they observed and discussed in sequence.
The role plays were scripted using content from firsthand student accounts of uncertainty documented in the literature [41][42][43],and enacted by experienced facilitators (Table 1). A live role play was used during the face-toface focus group, and a pre-recorded video role play was used during the online focus group.
In the first scenario, participants watched "Alice", a final-year medical student, disclose her uncertainty when, during her hospital placement, her patient's unexpected communication and behaviour made her feel uncomfortable, limiting her capacity for clinical historytaking. The role play captures her debrief with a clinical tutor. Participants were asked to analyse Alice's experience using Han et al. 's taxonomy as a guiding framework, i.e., "If you place Alice at the "locus" of uncertainty here, what sources and issues of uncertainty existed for her?" The participants were then asked: "What knowledge, skills or attitudes might have helped Alice to manage uncertainty in this situation?" In the second scenario, participants watched "Dena", a first-year nursing student, discuss her uncertainties around a poorly organised problem-based learning (PBL) session with a fellow student. Again, participants were asked to apply Han et al. 's taxonomy, this time with Dena at the locus of uncertainty. They were also asked to consider what knowledge, skills or attitudes might have helped her to manage uncertainty.
After both scenarios had been observed and discussed, participants were invited to reconvene as a large group. Participants in the group were then asked to contribute to a list of knowledge, skills and attributes that could help undergraduate health professions' students to prepare for uncertainty more generally.

Participants
The faculty development intervention was offered on two occasions, with two distinct recruitments of participants. The first session was offered as part of the Irish Network of Healthcare Educators' annual conference [44]. The session was promoted to all conference attendees through the conference website and programme. The second session was offered to members of staff at RCSI, a health professions-specific university with an international campus (RCSI, 2000). This online session was promoted through email lists and the university's in-house social media platform (WorkVivo; Cork, Ireland), and was facilitated using Zoom web-conferencing software (Zoom; San José, USA). Participants in both sessions were informed that they could take part without being included in the research study. To qualify for inclusion in the study, participants needed to have an active role in supporting learning for health professions' students, and to have provided consent for their comments to be captured. There were no specific exclusion criteria. No incentives were offered to take part in this study.
The faculty development sessions were attended by 30 participants, and all attendees chose to take part in the study. The face-to-face session had 14 participants (13 Table 1 Uncertainty role play scenarios Available online at: https:// rcsi. cloud. panop to. eu/ Panop to/ Pages/ Viewer. aspx? id= 72807 b43-baa6-4e56-84c0-aba60 0a84c ad female, 1 male; three focus groups): eight medical educators, three pharmacy educators, two dentistry educators, and a medical education researcher. The online session had 16 participants (13 female, 3 male; three focus groups): five pharmacy educators, two medical educators, two pre-clinical sciences lecturers, two health professions' education administrators, one physiotherapy educator, one nursing educator, one physicians' associate educator, one psychology lecturer and one simulation educator.

Researcher characteristics
All members of the research team have expertise across health professions' education, workshop facilitation and qualitative methodologies. JM is a faculty developer with a research interest in uncertainty in health professions' education. EAC is a faculty developer with expertise in professional identity. JH is a veterinary educator with a research interest in students' management of uncertainty. SK is a clinical medical educator with a research interest in assessment methods. TP is an educationalist with expertise in clinical communication skills, role play-based training and several decades of experience as a principal investigator in health professions' education studies.

Data collection
Data were collected in a range of different formats. First, group artefacts, i.e., flip chart pages (face-to-face focus group) and shared presentation slides (online focus group), were captured, and text was extracted from these. Second, focus group discussions were audio-recorded (face-to-face focus group) or video-recorded (online focus group) and these were transcribed. Finally, field notes were kept by facilitators during the focus group discussions.

Data analysis
Data from both study cohorts, face-to-face and online, were combined. Text from the focus group artefacts and from the transcribed discussions were organised using NVivo 12 (QSR International; Melbourne, Australia). Data were examined using a two-stage, hybrid approach of deductive and inductive thematic analysis based on a process proposed by Fereday and Muir-Cochrane [43]. In the first stage, JM created a coding framework based on questions that participants were asked to discuss in their focus groups which drew on Han and colleagues' [29] taxonomy of uncertainty. These were: • What sources of uncertainty exist here for Alice? • What are the substantive issues of uncertainty that Alice faces here?
• What knowledge/skills/attitudes might help Alice manage uncertainty here? • What sources of uncertainty exist here for Dena? • What are the substantive issues of uncertainty that Dena faces here? • What knowledge/skills/attitudes might help Dena manage uncertainty here?
Data were categorised by the researchers according to each question using themes and sub-themes. In the second stage, the data were examined again by JM using an inductive approach; this was carried out to screen for unexpected themes which may have been relevant to our research questions. The application of the coding framework to the dataset was discussed by JM, JH and EAC, and the findings of the analysis overall were discussed by all members of the research team.

Results
Data analysis of the focus group interactions revealed four themes overall. Three themes related to categories that had been pre-determined by the coding framework: "Sources of uncertainty", "Issues of uncertainty", "Uncertainty attributes" (Table 2). One further theme was identified as a result of our second stage of inductive analysis: "Learning environment" ( Table 3).

Sources of uncertainty
Using Han et al. 's taxonomy of uncertainty as a guide, focus group participants discussed many different sources of uncertainty for the students in the role played scenarios. These sources were categorized by the researchers into sub-themes which reflected the taxonomy: probability, ambiguity and complexity.
For example, participants noted how the future indeterminacy of events, or probability, could provide a source of uncertainty for the student. In the case of the clinical student, Alice, the participants talked about the different ways in which the situation could unfold for her:

"She does express an uncertainty about her grading and how she will be perceived based on that interaction although it wasn't clear unless the woman complained, she probably wouldn't get into trouble. " [Medical education researcher, female]
Participants also shared ideas about how a lack of clear or trustworthy information (ambiguity) contributed to uncertainty in both scenarios: "There seems to be an ambiguity in terms of the grading process as well, so that might be another source. " [Medical educator, male] "They had been given a lecture, I suppose, but they • There is a specific lack of details around psychosocial information concerning the patient • It is not clear whether or not the patient's response is a symptom of a psychiatric disorder • There is a lack of trust in information related to consent, i.e., "Is this the right patient?"; "Is consent in place?" • Alice lacks clear options on how best to proceed, i.e., stay with patient or exit? which communication skills to use?
• Alice lacks clear information on how she will be assessed on this interview PBL is a teaching and learning strategy which naturally incorporates elements of ambiguity • There is a lack of induction to PBL and why it is used • There is a lack of instructions as to the specific task, e.g., details around goals, student roles, learning outcomes • There is a lack of clear guidance as to how assessment will take place Complexity • The patient's situation and presentation are inherently complex • Consent and capacity to consent are complex concepts • There appears to be a rapid change in the patient's condition which culminates in withdrawal of consent • Alice experiences tension between two roles: a student who will be assessed and a healthcare professional, i.e., "How will this reflect on me?" versus "How am I to proceed in managing this patient?" • Alice's apparent lack of experience and/or training adds difficulty to the situation • PBL is an inherently complex teaching and learning strategy • The students are in their first year and lack experience with PBL • The environment, e.g., room set up, was not conducive to a successful PBL session • Key people, e.g., the facilitator and other group members, are absent from the session • Dena is balancing the priorities of different tasks, i.e., engaging with the PBL session or studying for an anatomy test  • Help-seeking (i.e., knowing when/how to ask for help, knowing your team and who to call, knowing limits of capabilities) • Consent and capacity • The importance of the patient (i.e., knowing to place the patient at the centre of the learning) • The nature of uncertainty: "It's not personal" • More knowledge about PBL, its purpose and value • Uncertainty is part of the process in PBL • More knowledge about the specific session (e.g., the outcomes required, the specific assessment methods) • More knowledge about her classmates and the facilitator, and their roles and potential issues affecting them • More knowledge about the PBL topic and its importance • How group work takes place • How the session fits with the end goals of Dena's profession • The nature of uncertainty: "It's okay to not know"; "Sometimes the answers aren't going to be perfect"; "Sometimes there is uncertainty and it's just part of the process and you just have to go with a path") Discussions also highlighted aspects of the situations which made it harder for each student to understand (i.e., complexity). For example, participants picked up on the tension experienced by Alice in balancing her identity as a medical student with that of a new healthcare professional: "So for the student, the greater consequence is 'How will this reflect on me?' as opposed to 'What are the issues for the patient?' So the student has the dilemma of having to fulfill both roles, one, in how they're performing and being assessed, and, two, as a… as a physician, even in their early days, 'How am I to proceed in managing this patient?'" [Medical educator, female] Finally, there were occasional instances where participants expressed confusion as to what aspects of the students' experiences could be classified as sources or issues.

Issues of uncertainty
Our participants were also able to identify different issues of uncertainty for each student which were organised by the researchers into sub-themes according to Han et al. 's taxonomy [29]: scientific, practical and personal.
With regards to scientific, or data-centred, issues, participants observed how the situation had led to specific knowledge gaps for the students around their learning experience: There were multiple comments which related to how the system around the student, or practical issues, represented specific issues of uncertainty: Finally, participants were able to discern different ways in which the impact of these uncertainties could be experienced by students from a personal perspective. Personal issues of uncertainty that were mentioned included:

Uncertainty attributes -role-play specific
Our third theme, uncertainty attributes, stemmed from a section in the session where participants were asked to discuss what knowledge, skills and attitudes may have helped students to manage uncertainty within the context of the specific role play ( Table 2).
With regards to knowledge, participants highlighted that more background knowledge of the topic at hand, e.g. a better patient history for Alice or a better understanding of MRSA for Dena, could have helped the students. They also listed specific types of knowledge for each student. For example, they considered that Alice could have benefited from more knowledge around setting boundaries and knowing when and how to ask for help, whilst Dena could have benefited from more knowledge about PBL, its purpose and value.
The participants also commented that the students might find it useful to have more insight around the nature of uncertainty in health professions' education:

"Sometimes you just have to go with a path. " [Physician's associate educator, female] "It's okay to not know…" [Pre-clinical sciences lecturer, female]
Participants listed numerous skills that could have helped these students. Notably, communication, teamwork and problem-solving skills appeared relevant to both Alice and Dena. Similarly, many attitudes were discussed with openness, growth mindset and positivity/ positive outlook observed across both lists.

Learning environment
The final theme that was identified was "Learning environment. " This theme covered a range of different comments which addressed participants' perspectives on how the learning environment could play a role in the uncertainty experienced by these students, or other students in similar situations. The comments were categorised according to four sub-themes: "Reducing uncertainty"; "Role of the educator"; "Evidence-based teaching strategies"; and, "Addressing the culture around uncertainty. " The first sub-theme focused on the ways that uncertainty in the learning environment can be reduced for students: "If you're running a course, it's very important to make sure that all the guidance documents and the resources are available and arranged and provided in a logical order. So maybe it's a good idea sometimes to get somebody who's not directly involved to proofread everything, you know, as a fresh eye just so that, um, things are optimal for the students coming in.

" [Health professions' education administrator, male]
The next sub-theme highlighted the importance of the educator and educator-student relationships in helping students to navigate such uncertain situations.
"I suppose that maybe the tutor is available or it might be that they feel they can report back uncertainties. You

Attributes
Finally, our participants were asked to contribute to a list of attributes, knowledge, skills and attitudes, which they considered would provide a foundation to undergraduate health professions' students in managing uncertain situations. These included an awareness of the nature of uncertainty within healthcare practice (i.e., "health professions' work has many grey areas as opposed to black/ white ones"; "uncertainty is not always bad"), an ability to recognise uncertainty, and adopting attitudes of adaptability, positivity, and resilience. A full list of proposed uncertainty management attributes is presented in Table 4.

Discussion
Revisiting our research question, we reflect on the ways in which Han et al. 's taxonomy can support health professions educators' to better recognise and conceptualise uncertainty. The findings from this study indicate that this taxonomy of uncertainty can be pragmatically applied to a faculty development setting. Specifically, the framework allowed participants to achieve a greater depth of understanding around students' experiences of uncertainty than they might have achieved had they worked without it.
Using the taxonomy, participants were able to identify multiple sources and issues of uncertainty for the students, thus unveiling the extent of uncertainties encountered by health professions' learners in relatively commonplace circumstances. Whilst it may seem paradoxical to want to unpack an experience of uncertainty into further, multiple uncertainties, this can be an important opening step to managing such situations.
As noted by Han et al. [29], "uncertainty is not a monolithic phenomenon. There are multiple varieties of uncertainty, which may have distinct psychological effects and thus warrant different courses of action. " By recognising and teasing out the separate sources and issues, educators can identify, or help students to identify, more adaptive responses to uncertainty.
It was also considered that the taxonomy worked well in the focus group setting. Previous studies that have used Han et al. 's taxonomy to organise experiences of uncertainty have tended to describe one-toone approaches such as in-depth interviews [31,34,35]. To the authors' knowledge, this is the first time that the taxonomy was used in a group setting. The level of detail provided by the participants around the students' uncertainties was surprisingly high and suggests that using the taxonomy in a group setting, where individuals hold different personal responses and perspectives, can uncover a more complete range of sources and issues of uncertainty. Perhaps this process, beyond the taxonomy itself, stimulated reflection on what constitutes 'good' management of uncertainty or, indeed, 'good' teaching around management of uncertainty?
This was also the first time, to our knowledge, that the taxonomy was applied to simulated rather than real-life experiences. We chose this approach as we considered it a low-risk way of introducing educators to student uncertainties. In practice, this took the form of a perspective-taking exercise, where educators were asked to focus on the students in the vignettes. There was some evidence that this approach helped educators to develop more open approaches to student uncertainty; one participant described a greater empathy towards one of our role play students as a result of engaging with the taxonomy. This is an important faculty development issue, considering the salience of the educatorstudent relationships in mitigating the uncertainties that students encounter in the course of their studies.
Overall, we consider that this framework provides a valuable tool for faculty to approach conversations around uncertain situations. As Han et al. [33] explain, the taxonomy has an important function in "promoting shared awareness of otherwise unconsidered sources and issues of uncertainty, and enabling stakeholders to approach these uncertainties in an organized manner. " In the context of education, this taxonomy can provide an entry point and guiding framework, promoting dialogue around uncertainty that might otherwise be avoided. The taxonomy could also be used to help educators and students engage in shared reflection around the specific sources and issues of uncertainty in any given situation, facilitating shared mental models and improved decision-making processes.
Whilst most of our data related to the taxonomy and activities used in the intervention, our analysis also revealed a theme around how the learning environment contributed to the uncertainty experienced by these students. Despite being asked to focus on the knowledge, skills and attitudes that could help these students, the discussion often drifted towards what the participants might have done differently in organising teaching. The issues highlighted were: reducing uncertainty for the students, the role of the educator, using evidence-based teaching practices, and addressing the culture around uncertainty. This digression from the task at hand may • The purpose and value of teaching strategies (i.e., "priming" for learning) • An awareness of others' issues and roles (e.g., other classmates and patients) • The centrality of the patient in healthcare • The nature of uncertainty within healthcare practice (i.e., "health professions' work has many grey areas as opposed to black/white ones"; "uncertainty is not always bad") • have resulted from the choice of role plays; in both cases there was a perception that students had been let down by some level of failure in their education. Alternatively, this might have represented participants' efforts to prevent uncertainty from arising in teaching, rather than acknowledging that students will likely experience uncertainty as part of their learning experience.
Our study revealed that participants, when faced with this in-depth and extensive appreciation of uncertainty within the scenarios, tended to respond in different ways. There were responses that were oriented towards students' learning and development (e.g., using these to make explicit the nature of uncertainty to students, and defining the skills to cope with such situations). However, there were also responses that were oriented towards reducing or removing the uncertainty for the student, with many suggestions as to how to "fix" the situations that Alice and Dena found themselves in. A drive to reduce uncertainty has been well documented within the health professions' practice. As Han et al. [45], say: "Physicians and other health care providers manage these effects and their experience of uncertainty itself through various strategies, but principal among these is the effort to seek information to reduce uncertainty. Nearly every major clinical activity that physicians undertake-diagnostic, prognostic, and therapeutic-is part of this overarching effort. " However, our findings here suggest that that this drive to reduce uncertainty extends beyond clinical activities into those of teaching.
Whilst reducing uncertainty for students and patients is important, it may not always be possible, or, indeed, the right course of action. The value of identifying these uncertainties is not so much that they can be prevented by preparing the student, but more that they can be used as focal points for reflection and discussion, supporting the student to manage, respond to and cope with uncertainty. In this study, participants did seem to recognise that there are times for educators to step in and times to step back. For example, it may be appropriate to reduce uncertainty for students at specific points in their training e.g., to support them as they enter clinical rotations or during first exposure to problem-based learning, with a gradual 'stepping back' of this supervisory management of the environment, being replaced by exposure to uncertainty. It's likely that such a scaffolded approach to uncertainty can be approached practically through shared reflective strategies where students and tutors employ tools such as the taxonomy used here [46].
Within the theme of learning environment we also noticed the appearance of a more positive, accepting narrative around the culture of uncertainty. Past research has highlighted that uncertainty can be viewed from a negative perspective by educators and students [14].
Here, participants commented that uncertainty should be highlighted as normal and "a good thing", and that development of skills to manage uncertainty can be viewed as part of the maturation process within undergraduate health professions' education. This may suggest that faculty development interventions, such as the one described here, can promote a 'stop and reflect' approach for educators, i.e., encouraging them to pause to consider, and communicate to students, the importance of productive uncertainty in the learning process.
Our second research question enquired as to what health professions' educators considered as key attributes for undergraduate learners with respect to managing uncertainty. When asked to consider the knowledge, skills and attitudes that could help our role play students, and undergraduate health professions' learners more generally, to manage uncertainty, the participants listed multiple and diverse attributes. Here we will focus on the more general attributes (Table 4) as these were deemed of greatest relevance to our research question.
Perhaps somewhat unsurprisingly, our participants considered that having a firm foundation in core medical knowledge would help reduce uncertainty for health professions' learners. They also commented that it would be helpful for students to have a greater understanding of their role, as well as that of their colleagues and the patient, in healthcare settings. Some comments related to the nature of uncertainty and how it would benefit students to understand that uncertainty is "normal" and "not always bad" in healthcare. Other additions to the list, i.e., "knowing what to do when you don't know what to do", served to reinforce how even experienced educators have difficult pinning down specific, observable steps to help students in managing uncertainty.
With regards to skills that may help students to manage uncertain situations, these included communication, emotional regulation, problem solving, information management, ethical decision making and an ability to self-assess. Each of these skills are supported within the existing health professions' literature [14]. Again, there were additions to the list, e.g., "taking the initiative", that were more vague and harder to define. One addition to the list, "recognising uncertainty" was of particular interest. Whilst we know that health professions' learners meet uncertainty in multiple stages of their training, it's likely that many learning opportunities are missed. The literature signposts that experiences of uncertainty commonly happen "under the radar" in healthcare settings, e.g., Mackintosh & Armstrong [47] highlight that "uncertainty work… may not be directly experienced or categorised as such by those undertaking it. " This leads us to ask if it may be helpful to prime both educators and students to notice when uncertainty emerges in the course of learning. It may be that using a "lens" of uncertainty to explore difficult situations could be a useful approach in the context of shared educator-student or student-student reflections.
Finally, attitudes thought to be helpful in managing uncertainty were: openness, adaptability, positivity, and a growth mindset. Again, these have been supported in the literature [48][49][50]. Although student attitudes are typically viewed as more difficult to expand on, such 'intangibles' can be developed experientially and through structured reflection, e.g., through small group 'debrief ' tutorials or personal journals.

Strengths and limitations
A key strength of this study is the successful application of Han et al.'s taxonomy [29] to a faculty development session, opening the door to further explorations of its use within educational contexts. We consider that the taxonomy is a practical and versatile framework which can help to empower health professions' educators when working with students to manage uncertain situations derived from real-life experiences. Whilst this study uses the taxonomy to focus on the perspective of a student (i.e., the student as the locus of uncertainty), our intervention could be easily adapted to explore the uncertainties faced by educators and clinicians from a wide range of socio-cultural and professional backgrounds. We also consider that overall instructional design of this session could be adapted to the context of undergraduate health professions' education, providing a valuable opportunity for students to diagnose and classify aspects of uncertainty that they meet during their training.
Although there were occasions where the participants expressed confusion with regards to the sources and issues of uncertainty, these were not frequent, and did not appear to detract from the utility of the tool overall. On this basis, we recommend Han et al.'s taxonomy as a conceptual framework of interest, which has potential application in a wide variety of uncertainty management educational interventions.
The study also resulted in a list of knowledge, skills and attitudes which can be used in discussions around how to manage uncertainty in the context of health professions' education. Our list of attributes, whilst not exhaustive, can be used by educators and students alike in considering how best to work with, or respond to, an uncertain situation.
There were, however, several limitations associated with this study. Our study sample size was small, and, due to the opt-in nature of both faculty development sessions, it's likely that participants held some pre-existing interest in uncertainty management. Further to this, almost all participants were based at health professions' education institutions in Ireland. Individuals' approaches to uncertainty can be heavily influenced by professional and socio-cultural variables [51][52][53], and, thus, our findings may not be representative of the views and experiences of a more diverse faculty cohort.
In addition, offering the intervention as part of a conference setting added time restrictions to its design. Participants were presented with just two role-played scenarios which aimed to balance depth of enquiry with feasibility. More time, and a greater number and diversity of scenarios, may have yielded different results. Finally, we believe that any future delivery of a similar intervention should include the active participation of health professions' students. Our focus in this study was the development of faculty and our intervention design used a perspective-taking approach, i.e., our participants were asked to consider uncertainty from the viewpoint of a student. Whilst we consider this approach was a valid way to provoke thought and discussion around uncertainty attributes, it would have helped to have health professions' students take part in the facilitation of the session to bring real-world depth to our role-played scenarios.

Conclusions
This study highlights the successful use of Han et al.'s taxonomy of uncertainty within a faculty development setting. The taxonomy provides a effective conceptual framework which educators can use to identify a wide range of sources and issues of uncertainty for students within simulated scenarios. Our findings suggest that Han et al. 's taxonomy is a practical and versatile tool in designing faculty development interventions around uncertainty management. Participants in this study also contributed to a list of uncertainty attributes, i.e. knowledge, skills, and attitudes, which may be useful to undergraduate health professions' learners in managing uncertainty. Finally, our findings suggest that health professions' educators can sometimes feel compelled to reduce or remove uncertainty for students, which may not always be the most appropriate course of action. We propose that more faculty development in this domain is likely to be required.