Tutor uncertainty in dealing with unprofessional behaviours of medical students and trainees: a mixed methods study

Background: Despite acknowledgement of medical students’ expected professional behaviours and attitudes, there remains widespread reluctance to report students that behave inappropriately. Existing literature focuses on why faculty fail to fail, overlooking the clinical teachers who deal with students day to day. We investigated how clinical teachers address inappropriate behaviours and attitudes in students and trainees. Methods: A mixed methods study was carried out consisting of a survey and two focus groups with clinical teachers. Seventeen clinical teachers from the at University of Limerick School of Medicine, Ireland, took part in the survey (n=22%) and eight clinical teachers participated in two focus groups during the 2018-2019 academic year. Results: Findings suggested that 59% of tutors would take a different approach in regards to professional identity formation (PIF) between addressing unprofessional behaviours witnessed in medical students and trainees. 88% of tutors said they intervened on a professionalism issue with 52% saying ‘once in a while’ In contrast to the survey, tutors in the focus groups expressed a lack of condence in addressing some behaviors due to a lack of time, not seeing the outcome of process/remediation etc. Tutors indicated a strong preference for case-based training on assessing PIF. Conclusions: We found tutors typically work closely with students on a day-to-day basis and manage unprofessionalism issues. Clinical tutors valued regular communication about policies and procedures about appropriate conduct as well as support, advice and/or oversight from independent member of the university staff. This research also highlights the need for training designed for busy clinical tutors as a distinct type of medical teacher.


Background
The development of professional behaviour and attitudes in medical students and trainees remains a focus for medical educators with much interest in understanding how this can be optimised. Clear guidelines and standards exist for the professionalism expected of medical students and trainees. 1,2 In addition, all schools and training institutes have local policies on assessing professional behaviour and for disciplining and remediating poor professional behaviour. Yet, educators often fail to adequately identify and report unprofessional behaviour. [3][4][5] This has been described at the level of senior academic faculty and school administration and has been attributed to multiple factors. 6 In many medical schools, a signi cant educational role is played by non-faculty tutors. Such tutors are uniquely positioned to observe students' behaviour as they regularly interact with them in a variety of settings from classroom to bedside teaching, often in small groups, delivering both formal and informal teaching. 7,8 Their central role in guiding students' professional development is one that deserves greater attention in the discourse about teaching and assessing professionalism. 9 This is an under-reported population in this context and training should be tailored to their speci c roles; tutors commonly juggle clinical and teaching duties 10 and faculty development initiatives to address their educational needs are often unsuccessful in clinical settings. 11,12 Gaining an insight into how tutors view their role in managing unprofessional behaviour will enable better training and support for those at the frontline in fostering students' and trainees' professionalism.
In order to develop successful faculty development and clinical teacher development programs, we need to further understand the barriers that exist in addressing professionalism issues at the clinical teacher level. In this study we used a mixed methods approach to investigate tutors' experiences of and attitudes to identifying and managing unprofessional behaviours in medical students / trainees. The issues explored include tutors' reluctance to report professionalism issues and the underlying reasons for their di culties in identifying and remediating those issues.

Study context
The University of Limerick's BMBS medical programme is a four-year graduate entry degree. Multiple strategies to teach and assess behaviours and attitudes are embedded across the four year programme with a speci c focus in an integrated Professional Competencies module which runs throughout all four years of the programme. All clinical teachers, many of whom are long-serving tutors, are all medically quali ed.

Study design
In order to gain a wide tutor perspective of unprofessional behaviours in medical students and trainees a mixed method approach was deemed appropriate. An online survey was designed with two follow-up focus groups; one to be based with clinical and problem-based learning tutors (PBL) (

Quantitative data collection
An online survey was conducted between October and December 2018. The survey used (see Appendix A) was designed with input from experienced tutors and piloted with two staff members. All PBL and clinical tutors (n = 76) were invited to take part in this study via an independent gatekeeper at GEMS. Informed consent was given by way of clicking a tick box prior to completing the survey. Data was collected via an online survey with 17 clinical teachers. The response rate was 22%, typical of an online survey completion. 15 Demographic data collected included gender, age, years working as a tutor, clinical background and where they trained. Data was also sought from each tutor on how they had addressed inappropriate behaviours in the past and any further supports / training they would like to be provided.

Qualitative data collection
Two follow-up focus groups (n = 8) took place based December 2018-January 2019; one was based in the medical school and the other in a hospital setting. Volunteer sampling was used to identify participants for focus group discussions. All PBL and clinical tutors (n = 76) were invited to participate in a focus group and provided with information sheets (See Appendix B) via an email from an independent gatekeeper.
Participants were predominantly female in the medical school group and predominantly male in the clinical site group. Other demographics such as age varied in both groups. All participants provided their written consent (See Appendix C) and were aware they could withdraw from the group if they wished and assured that all the data collected would be anonymised upon analysis and used strictly for the purpose of this study.
These focus groups were facilitated using a guide (See Appendix D) comprised of open-ended questions relating to the role of each participant in addressing professionalism issues; good and bad behaviours typically seen, the effectiveness of addressing issues, any additional supports services required by tutors, and suggestions for improving the current system.
The duration of focus groups was an average 45 minutes. Focus groups were recorded on a digital recorder and transcribed. The groups were facilitated by a researcher (DOD), an experienced qualitative researcher, who was an impartial member of staff without teaching responsibilities.

Quantitative Data
We analysed our quantitative data using SPSS version 24. Descriptive analysis was completed on survey data.

Qualitative Data
Focus group audio recordings were transcribed by an external, third party company to reduce bias by researchers. Transcripts were then analysed using NVivo 10 using thematic analysis, outlined by Braun & Clarke 16 . This process was completed by the team in an iterative and inductive process, reviewing codes and revising themes in a number of stages (see Appendix E). Any identifying information was removed as part of the data analysis process. DK, DOD, SH and HMK reviewed the nal codes and themes and a consensus was reached.

Results
Quantitative Results Table 1 below gives an overview of clinical and PBL tutor (n=17) demographics. Educational level -Student versus trainee Tutors described differences in their reporting of and responses to unprofessional behaviour when the individual concerned was a medical student versus a quali ed intern / trainee. 58.87% (n=10) reported that the action they would take upon witnessing an unprofessional behaviour would differ if the individual was a student / intern / trainee.
Open-text comments alluded further as to tutor's reasons for this response: Participant 2: "As someone has progressed through their training it would be expected that they have attained a standard of professional behaviours. You would probably be more lenient with a student than an intern or postgraduate trainee".
Participant 3: "I think we are more tolerant of students in years 1 and 2….due to their lack of experience" Participant 4: "They (students) are still learning" Medical students / interns / trainees were viewed differently and as separate roles, each with their own responsibilities. The majority of tutors agreed that as medical students were still learning, any unprofessional behaviour would be dealt with more leniently, compared to a trainee who had nished their training. This may be seen as an opportunity for a 'learning moment' for the student.
Interventions with students / trainees on professionalism issues Most tutors indicated that they had intervened with a student / trainee in a professional issue (88%, n=15). Of those that had, the majority had only done so 'Once in a while' (53%, n=9). The most common reason for triggering an intervention with a student / trainee was a 'Single unprofessional incident' (52.9%, n=9). Ninety-four percent (n=16) of participants also noted that they had an informal conversation with a student / trainee to address the issue raised. 47% found the student / trainee to be 'Moderately receptive' to the discussion and 29% 'Quite receptive'. 100% of tutors felt that it was their responsibility to address lapses in professionalism with students / trainees.

Feedback and Reporting to students / trainees
Participants were asked about how often they acknowledged good professional behaviours in students / trainees by explicitly telling the student / trainee. Responses varied from 'Once in a while' (29.4%), 'Sometimes' (23.5%), 'Often' (29.4%) and 'Almost always' (17.6%). This explicit acknowledgment however was not always recorded for future feedback or as a log of behaviours for students / trainees. The majority of tutors (41%, n=7) 'Almost never' routinely recorded professional behaviours for feedback to students / trainees or as a log of behaviours (Table 2). Of those that did record these behaviours, most were recorded at least four times (53%), with only 5.9% (n=1), recording after each session (Table 3). A majority of tutors noted that they were 'Quite con dent' however in providing feedback to students / trainees (71%, n=12).

Training Needs of Tutors
Almost a third of tutors (29%) stated that a reason for not addressing lapses in professional behaviour amongst students / trainees was due to their lack of training. Only 18% (n=3) of tutors noted having received any training in recognising and assessing professional behaviours in students / trainees.
This highlights a signi cant overlap and calls for further training to be provided. When asked what type of assistance would help tutors intervening with unprofessional behaviour, the majority (82%, n=14) noted case-based reference examples. Where tutors oversaw substantive behaviour change in students, the majority (47%, n=8) felt 'Quite supported', similarly when asked on delivering consequences on lapses in professional behaviour to students / trainees.

Qualitative Results
Seven core themes were highlighted as part of qualitative analysis with a number of minor themes (Table   4). Appendix F in supplementary material provides a description of our coding tree / thematic roadmap.  Participants spoke to their approach in dealing with unprofessionalism issues with students and trainees. These issues were discussed as part of several minor themes, relating to their approach taken dependent on the training stage of a student / trainee, the nuances between remediation and the insight of students.

a) Stages of Education
Tutors outlined how they would address lapses in professionalism, which was dependent on the stage of medical training the student / trainee. These lapses as a medical student were seen as training moments, whereas lapses by quali ed interns / trainees were viewed more seriously.
CP2 (Group 2): "I think once you've quali ed I mean there are still things to be learnt, but I think there is a different measurement value attached to that, because you affect the team in clinical practice, it affects the patient" b) Remediating knowledge and skills versus behaviour In reviewing lapses in professionalism, tutors discussed the difference in relation to remediating knowledge and skills rather than behaviours in students / trainees. Issues relating to knowledge and skills were distinctly outlined differently compared to poor behaviour.
CP3 (Group 2): "…with the junior trainee I've had less concerns with professionalism but more concerns with competence and ability." CP2 (Group 2): "Competence is easy to assess and poor competence doesn't make someone a bad doctor, you know… so I think competence we need to assess and that's important but that's far more changeable." c) Insight of student Tutors also discussed addressing professionalism lapses with students / trainees with many highlighting that for the most part, students / trainees do acknowledge and accept ownership when an issue is highlighted to them.
CP3 (Group 2): "they just have a lack of insight... so when you draw it to their attention then they are… appreciative, but it's very rare that you would get someone who would get their back up by what you say"

Issues related to an unprofessional incident(s) a) Formally / Informally
Tutors discussed how they had previously managed professionalism issues with students / trainees, often were done informally. Tutors noted that formal complaints were only done when informal discussions has failed to address issues.
CP3 (Group 2): "I don't think I've a formally reported anyone for professionalism issues, I've come across, I guess a couple of students… which I might have discussed informally with colleagues and with them. But, not fundamentally ever to the extent whether I've made a formal complaint" b) Level of severity of issue Tutors often chose to report professionalism issues depending on the level of severity of the issue, determined by the tutor. Where tutors felt there might be a higher level of risk or where patient safety was being affected, issues would then be escalated. Whilst tutors often dealt with more minor issues, they were cognisant that in not addressing these that it could potentially escalate over time to more serious. Tutors discussed how they often felt that they had a lack of knowledge on outcomes following the reporting of a lapse in professionalism with students / trainees. Tutors felt that if they were more aware of the reporting and assessment process then this would provide them with the con dence in reporting.
CP4 (Group 2): "we had one episode last year where there was actually a problem with aggressive behaviour with a consultant on a ward round and that was escalated to the CEO, but I never had to bring (it) any further than escalating it, but often I didn't necessarily see the resolution or what the outcomes were." CP2 (Group 2): "So, the biggest barrier is knowing or con dence I guess that being an accurate fair assessment as part of the process, and I think that encourages reporting for me… " CP2 (Group 2): "…everybody's right in saying that if you are going to report you do want to have some con dence that the input is going to be taken seriously…but it has to be properly accessed in a valid and a structured way and a consistent way…you have to be very clear what is and what is not happening then there has to be an opportunity to remediate and to access that and to see what change there is."

Reluctance to report a) Perceived consequences for the student
Tutors explored what they believed to be the consequences for students if reported for unprofessional conduct. Many felt that reporting should also have supportive actions rather than punitive for students, allowing a learning opportunity for students.

CP1 (Group 2)
: "if you suspect that this (reporting) might be more putative rather than supportive. You might end up thinking maybe I will give this person extra chance, because if it's a case of where the focus is more on making a note of this person as unprofessional rather than actually trying to educate that person as to where they went wrong." GP1 (Group 1): "…like …was is that bad, or when I think about it or maybe it wasn't that bad, how, you know, do they need to be punished. But, then when you read the medical council guidelines for a student it is very clear, they are very clear, it is just that maybe we are not, but they are very clear and what is that we might chose" b) Perceived consequences for the tutor Tutors identi ed potential consequences for themselves in escalating and reporting lapses in professionalism. A lack of knowledge of an outcome and feedback led to frustration and tutors questioning whether their report was valued and held to the same level of severity by superiors.
CP4 (Group 2) "…very frustrating that if you bring something up and it doesn't go any further, it's kind of 'a' it shows that there is no repercussion for the person. And 'b' -you are then brought out to be kind of a trouble maker"

Oversight a) Lack of a clear feedforward practice
Tutors discussed at length the lack of a clear feedforwarding process of information on students who may have been remediated or could potentially need remediation. This perceived lack of knowledge on reporting processes impacted tutor con dence. GP1 (Group 1): "how do we informally notify each other as well, you know, and part of that I suppose is con dence on our side as well" CP2 (Group 2): "I think it's a big problem because we see people and there's very good examples for over last few years…in both those cases it was impossible to change, and yet, kind of as we discovered everybody knew about it, but there wasn't a system in place to take X and move him forward as an undergraduate" b) Independent Oversight Whilst tutors discussed lack of a feedforwarding practice, several suggestions were made that might address this. One proposed solution was the appointment of an academic liaison/pastoral support role that would be independent of training and assessment.

Discussion
In the challenging healthcare system, the assessment of behaviours and attitudes suffer because not only is the clinical teacher under pressure due to lack of time, we found a lack of con dence in remediation process also hampers chances of improvement. This study addresses uncertainties in recognising and managing lapses in professional behaviours in medical students and trainees. Results highlight that tutors are generally happy to address minor breaches informally and feel con dent in identifying serious breaches in professional behaviour. They are often unaware of the reporting procedures in place, hence affecting tutor con dence, leading to under reporting of behaviours. Factors which also affect identifying and managing unprofessional behaviours include lack of time with the student, peer and institutional culture, educational stage of the individual and the level of severity of the issue. Tutors expressed the need to improve their knowledge on addressing lapses in professionalism, and they felt case-based references would be most suitable. There are many suggestions in the literature that teaching professionalism didactically, as therapeutics and skills are taught, is not likely to produce the best results. 17 The preference of clinical tutors for case based training and also the suitability of case based training has been highlighted as potentially advantageous as it can make moral issues easier to identify and confront. 17 This nding is unique in this study and will aid in our planning of faculty development programs for clinical teachers. Ziring et al. 18 found that faculty in medical schools expected to be central to identifying unprofessional behaviours in medical students however less than half the schools had a formal program to train staff to do so. Of the schools which did offer this professionalism awareness training to faculty, this was done so via faculty development programs. 6, 19 Our study highlights a gap knowledge and availability of appropriate resources aiming to teach clinical teachers, rather than faculty, about the identi cation and assessment of professional identity formation.
In this study, there was some disparity between educators survey answers about experiences of reporting unprofessional behaviours in students and trainees and the answers given in focus group setting; essentially social desirability bias. 20 82% found it 'Somewhat & Moderately Easy' to discriminate between minor, moderate and serious breaches in professional behaviour. Where an intervention with a student / trainee did occur, 77% tutors felt it was 'Moderately / Quite Effective'. Overall, 47% of tutors felt 'Quite Con dent' in dealing with professionalism issues in an informal manner. 59% also reporting being 'Quite / Extremely Supported' in delivering consequences on lapses in professional behaviour to students / trainees. However, as part of the qualitative discussions, many participants' comments were quite polarising to quantitative data. Many tutors discussed the barriers in reporting professionalism issues (formal and informal) such as lack of time with the student whilst on rotation to identify patterns of behaviours, perception of the lack of a clear reporting and feed forward process and lack of time for remediation.
Tutors were tolerant of minor lapses in students as they contended that students were still in the process of learning and constructing their professional identity. Participants took a holistic, personal transformation approach to professionalism seeing it as a personal transformation from medical students to junior doctors. This approach has been found in the literature to more accurately portray the complex, contextual nature of desirable behaviours and attitudes of those practicing in the eld of medicine. 21,22 Mixed method results also highlighted this fact in relation to how tutors would often treat lapses in professionalism with medical students compared to trainees. Furthermore, tutors felt they were more likely to encounter situations where they witnessed a temporary lapse in professional behaviour rather than a bad student.
Tutors commented on how time impacted their ability to identify issues with student as they may not be able to assess if a concerning incident is part of a pattern of behaviours. Tutors also lack the time to see remediation of students and whether, if at all, remediation had its intended effect on students' PIF. Ziring et al. 6 also identi ed time as a barrier to reporting of unprofessional behaviours. Insu cient time for discussion with student and also for institutional reporting practices have also been described previously. 19 Limited time to identify and managing unprofessionalism issues can also lead to a lack of uncertainty for tutors in reporting incidences, either formally or informally. 23 This was highlighted in relation to con dence in dealing with a professionalism issue. It has been reported that some faculty see formal professionalism curricula as an attempt to "force all students into the straightjacket of political correctness" 4 and this may explain some of the reluctance of our tutors to follow a formal route when intervening.
Clinical tutors should strive to provide students with a suitable learning environment, where clinical competencies are learnt and practised, and it is likely that tutors often go above and beyond for medical students and trainees. For some, clinical tutors often act within a student support / student advocate role. It is important to be cognisant of this, particularly when reviewing the barriers which tutors face in identifying and reporting unprofessional behaviours. This element is crucial which much of the existing literature lacks -tutors who are 'at the coal face' are not only trying to ensure that students are gaining the appropriate clinical skills, knowledge and competencies, but also remain balanced in being a student advocate and ensuring appropriate institutional procedures and policies are followed and adhered to in relation to unprofessional behaviour. This may causes dilemmas for tutors, compared to others in more senior positions / deans of medical education. 6, 18 Tutors may have built up a rapport with students, therefore may be more willing to ignore more minor lapses in professionalism. It is therefore essential that those in the frontline with students and trainees are both aware of the appropriate procedures and feel supported by their institution in reporting professionalism issues but also con dent in the process that suitable remediation is put in place, related to the severity of the issue. Tutors who can identify and address more minor incidents in the earlier years of a medical student / trainee therefore are essential in reducing the number of more severe lapses in professionalism at a later stage, given the relationship between early issues and serious lapses in professionalism at later career stages. 24 One of the areas of uncertainty discussed by tutors was related to lack of a clear feed-forwarding process or communicating student information between tutors. A study conducted by Ziring et al. 18 highlighted this 'feedforward' practice; de ned as how information about a student experiencing professionalism issues was communicated to future supervisors. Ziring et al. 18 found that factors that in uenced the decision to 'feed-forward' this information included the stage of training the student was at and the type of professionalism lapse. The issue of student con dentiality and the potential for bias was also discussed. Our study ndings are quite re ective of the uncertainty and issues also found by Parker et al. 25, van der Vossen et al. 26 and Ziring et al. 18 Tutors in our study put forward recommendations for a person responsible for communicating student information formally and con dentially who would have independent oversight. It is notable that this role is ful lled in our institution by a senior academic, which emphasises the role of communication in medical schools to raise awareness of pathways to deal with professionalism issues.

Study limitations
There have been very few studies exploring the views of tutors in dealing with behaviours and attitudes issues outside of large teaching hospitals. The response rate for the quantitative survey (22.3%) with eight tutors taking part in qualitative focus groups. Due to the busy nature of work of tutors, it was di cult to try and improve sample size and response rate. This may have improved using other methodology such as semi-structured interviewing which authors recognise and may choose to use in the future.

Conclusions
It is important that tutors, who are front-line staff in dealing with medical students and trainees on a day to day basis are suitably trained in identifying and addressing professionalism issues. Tutors should also strive to ensure that unprofessional behaviours are documented and suitably reported and / or remediated for the betterment of future doctors. Tutor uncertainty should be mitigated by ensuring tutors are aware of existing reporting procedures and faculty should recognise the need for additional tutor training to address lapses in professional behaviour in medical students and trainees. In giving tutors the con dence to report the most minor of occurrences to the most serious, we are providing them with the toolkit for educating future doctors.

Declarations
Availability of data and materials The datasets generated and/or analysed during the current study are not publicly available due to the information originating at one medical school at a local level but are available from the corresponding author on reasonable request.