Failing to fail phenomena

Abstract Introduction Clinical competence is the backbone of competence‐based dental education. Over time, there has been a paradigm shift toward training students who are capable of independent practice, as opposed to mere academic success. Methods A mixed‐method study was undertaken by anonymised email questionnaire to all restorative tutors at the UK Dental School. Demographics and teaching experience were ascertained, along with key questions on the utilisation of online assessment software iDentity. The assessment process for tutors was explored, and barriers experienced when grading students were reported. Results The questionnaire was sent to all 51 restorative tutors with a response rate of 59% (n = 30). Only 3.5% of tutors provided verbal feedback and grading to students in person, with 20.7% only completing iDentity gradings following an email reminder. The majority of staff (93.3%) felt comfortable in raising concerns; however, one of the three clinical tutors admitted they had allowed a failing student to a pass. Qualitative analysis demonstrated several themes why tutors were reluctant to fail students: maintaining good relationships, limited supervision, time delay of grading, one‐off event and the student's first attempt. Conclusions Grading students as competent as a one‐off experience could potentially mask a recurring problem with a student, in turn impacting the student's ability to assess their own weakness and believe themselves to be competent, and potentially be overconfident. Fair and accurate assessment has a significant benefit to student and staff, enabling targeted development to motivate the students and improve the quality of care provided to the patients.

such pieces of information, but also is capable of demonstrating how a procedure is done and is able to perform such procedures on their own.
At the heart of CBE lies the assessment. It is crucial for the educators to be able to demonstrate how and when a student progresses from one level of the pyramid to another. Consequently, the literature has been exponentially expanding on this topic. [14][15][16][17][18][19][20][21] Out of a variety of toolkits available in the toolbox of assessment, longitudinal assessment of performance over time has been described as an optimal method at the highest level of clinical competence. 22,23 and quoted as the most popular method within medical and dental schools in the United Kingdom. [24][25][26] Longitudinal methods of assessment, however, bring their own challenges and deficiencies into the equation. They are usually expensive to set up, require extensive infrastructure and most importantly require a great deal of clinical supervisors' buy-in. 25 This assessment method is heavily reliant on assessors' willingness to be part of the assessment process and more importantly being able to make a sound judgment on the performance of their students. This on its own casts a shadow of doubt over validity and reliability of such an assessment method. Fraser's report confirmed that incompetent midwifery students 'slip through the net' 27 and initiate a 'buzz' of disbelief amongst many clinical professions.
Healthcare professionals such as medics, 28,29 nurses, 30-38 social workers, 39,40 occupational therapists, 41 and dentists 42 have raised concerns that clinical modes of assessment can result in incompetent students graduating and being placed on healthcare professional registers.
The bachelor in dental surgery (BDS) program at the UK Dental School is designed based on a CBE curriculum with the current fluid concept of milestone assessments, where a student's clinical competence is evaluated throughout their undergraduate education. 3 The emphasis placed on longitudinal assessment is regarded as a highstake mode of assessment, influencing the progression of students at each year, as their clinical progression as a whole is evaluated. The assessment data are captured using a commercial electronic portfolio system traded under the name of iDentity. [43][44][45] iDentity allows tutors to grade the students on every single clinical encounter using one of the six possible grades: H, U, Sp, Sv, I and I+ (Table 1). All supervisors are briefed to make their judgment based on comparison with what they expect from a newly qualified dentist, therefore irrespective of the BDS year they are in. When processing the data at the Division level, grades of H and U are considered "fail" for the students in lower years whilst grades of H, U and Sp are considered "fail" for the students in the final year.
The iDentity system is heavily reliant on the judgment of tutors, many of whom are part-time general dental practitioners (GDPs) who could be supervising the students for as little as half a day per week. Though tutors are invited to annual training on iDentity, there is little evaluation on the supervisors' process of grading the students. Therefore, the aim of this research was to evaluate the failure to fail the students during clinical activity. F I G U R E 1 Distribution of seventeen assessment methods in the 2008 survey of assessment practices in US dental schools amongst Miller's pyramid of professional competence levels

Scale
Title Description

H Harm
The student's performance, knowledge or action has resulted or could have resulted in harm to the patient, their relatives or members of staff.

U Unsatisfactory
The student's performance, knowledge or action is below standards. However, this has not resulted in a compromise of the safety of the patients, their relatives or staff members.

Satisfactoryprocedural intervention
The tutor (or other healthcare professional) has to intervene to produce a satisfactory outcome.

Satisfactoryverbal intervention
Satisfactory clinical outcome was achieved after verbal prompts from the tutor.

I Independent
The student is capable of starting and finishing clinical tasks independently. The role of the tutor is purely supportive. No intervention from the tutor was needed.

Excellent independence
The student's clinical work is beyond that expected in their foundation training year.

TA B L E 1 iDentity marking descriptor
The objectives were: • To understand tutors' confidence in using the iDentity grading system.
• To understand if tutors are allowing failing the students to pass.
• To explore barriers why tutors may fail to fail poorly performing students.

| ME THOD
The study was completed at the UK Dental School. A short an- Demographics and teaching experience were ascertained, along with key questions on the utilisation of online assessment software iDentity, and the assessments process for tutor was explored. The options were populated from the most common features published in the literature, and free text was also utilised to discuss barriers they had experienced when grading the students.

| Analysis of the data
The data collection was anonymised and recorded on Microsoft Excel 2010. Data analysis of the free-text sections was completed using thematic analysis. 46

| RE SULTS
A total of 51 restorative staff were invited with a response rate of 59% (n = 30). One blank response was excluded. The staff who re- average.
Whilst 87% of the tutors were confident with the grading system, the remaining lacked confidence in grading via iDentity. Thirtyseven percent of the clinical supervisors admitted that they have allowed the failing students to pass on at least one occasion. Of the different possible reasons (tutors were able to select multiple answers), "giving the student the benefit of the doubt" and "considering the poor performance as a one-off incident" were the most likely reasons with 63% and 59% respectively. This was followed by "good knowledge but inability to apply the knowledge to practice" (50%), "feeling uncomfortable to fail students" (33%), "lack of training on iDentity grade boundaries" (33%), "being afraid of negative consequences" (30%) and "aim to maintain good relationship with the student" (30%).

| Barriers to failing students
A thematic analysis was undertaken by the free-text comments, from which six themes emerged.

| Maintaining good relationship
A key theme noted was the importance of having a good relationship with the students and the impact of providing the students with a grade of H or U. Several tutors' responses used emotive language such as "hurt" or "worry" toward the student as a consequence when providing low grades on iDentity feedback: "Failing them with a bad mark after they have shown the effort to do things right will put them down instead to continue encourage to improve to do better." (tutor 1-2 years' experience).
"I think sometimes we worry about knocking the student's confidence too severely." (tutor 3-5 years' experience).
"Some students have a theoretical knowledge and need more practice to achieve satisfactory results." (tutor 10+ years' experience).
These responses highlight that some tutors view the grading as an emotional assessment, by not wanting to hurt the student's feelings.
This results in prioritising avoidance of the potential negative impact on a student's confidence, if a poor grade was recorded, rather than providing an objective mark of the students' clinical performance.
Some tutors did not want to expose the students to the recorded consequence of failure, viewing verbal feedback sufficient in order to improve the student's clinical skill set and resolve the situation: The challenges of providing fail grades to students were also raised. Where instructions had not been followed or there was poor professional conduct, staff avoided low grades to prevent confrontation.
"Work was of a good quality but professionally it was a

| Limited supervision
Tutors raised concerns regarding their ability to appraise students' knowledge from the single clinical session each week, whilst others commented on the clinical pressures resulting in difficulty to assess the student's competence and therefore provide an appropriate grade: "Only looking after students one day a week can be difficult to assess the students' knowledge and then grade appropriately…." (tutor 3-5 years' experience).
"Being heavily booked on the clinic means that students can be left unsupervised…." (tutor 10+ years' experience).

| Time delay of grading
The completion of iDentity was a noted theme in the feedback, with a delay students' part to upload their own responses onto iDentity after the procedure and tutors not having sufficient time to complete student feedback face to face: The perceived value of the iDentity process was also a theme noted by the staff, with limited understanding in the CBE process and its role in the students' clinical progression. Another concern was noted that the tutor's views were dismissed by the academic "…Tutors' concerns ultimately dismissed." (tutor

3-5 years' experience).
It was also noted by tutors that the role of iDentity was only to record negatively if harm had been caused, even if the student had demonstrated inadequate clinical knowledge, rather than understanding the whole range of grades.
"The most likely problem is that the student has poor knowledge and/or is ill prepared for the clinical sessionbut no harm has been done." (tutor 1-2 years' experience).
"I find it far more useful to discuss with the student on clinical any problems that may have occurred" (tutor >1 years' experience).

| A one-off event
The students were given the benefit of the doubt on the procedure, as a one-off event rather than recording the grade of the viewed procedure: "…when student started cleaning and shaping canals without WL xray. After discussing the situation with student it was clear it was a one off and she did understand the procedure. When

| The first attempt
There were examples where suboptimal clinical performance was described when the students were completing the procedure for the first time. In these cases, the tutors showed empathy, as it was their first attempt, rather than providing an independent assessment of their clinical practice:

| Raising concerns
Ninety-three percent felt comfortable raising concerns about a student, with two staff members feeling they were not comfortable. A free-text section allowed the tutors to explore what pathways they would utilise in raising concerns regarding the failing students. The majority of responses noted they would liaise with the year lead regarding concerns about a student. Others noted they would explore concerns with another tutor before escalating. Two members of staff noted they would discuss the issue with the student initially, providing a warning verbally of potential provision of a "red card", which is then raised with year lead. The final option discussed was to speak to the undergraduate director of the concerns regarding a student.

| DISCUSS ION
The emotional link between staff and students appears to dominate the feedback process, which is a common theme in clinical education literature. 47 In turn, the students were potentially shielded from a true assessment of their clinical performance and limiting their development to independent clinical practice. By not wanting to expose the student to the consequence of failure, it could be argued as a conflict to the duty to protect the public. In examples where clinical care did not cause harm, but was not to a satisfactory standard, for example "over prepared crown", "ledges" or "poor margins". In these cases, is there a duty of candor to the patient to explain that the students' performance was not satisfactory and to discuss the implications of this? Shielding the students from the consequence of failure and avoiding confrontation with the patient and student fail to develop on core skills required as a dental clinician and maintain good practice in line with the General Dental Council Standards.
The systematic review by Yepes-Rios (2016) highlighted that there is often parental approach by tutors when providing feedback to protect the students. 47 Another barrier to clinical development is the tutors' own person views on how they would be perceived by others for failing a student. 39,41,47 A CBE program is heavily dependent on the tutors to perform an objective assessment of the student's performance. If the foundation of this assessment has a bias toward protection of the student and their perceived relationship, there is a risk that the students may slip through the net and graduate as students with significant deficiencies in their skill set that they were unaware of. To fail a student also required self believe and confidence in their own clinical judgment. Fear of legal action has been reported by dental, medical and nursing specialities during clinical education as a barrier to failing a student. 28,41,43 The environmental constraints of the busy clinic and the ability to provide tailored feedback have been raised as a barrier using electronic assessments. 45 The role of feedback as an appraisal and informational support was often overlooked due to time constraints, and completed days after the procedure had been undertaken, limiting the staff and student's reflection of their performance.
Only a small proportion of tutors provide students with immediate feedback on their clinical work. There are potential problems with this behavior, most importantly loss of valuable face-to-face feedback and also the potential for memory attrition. Therefore, the grade awarded may not be a true reflection of the student's performance. In improving the staff and student compliance in completing iDentity, the students must complete iDentity on clinic, with the gold standard of the tutors providing immediate feedback and recording on iDentity the same day. 45 Where feedback is delayed, reducing the email reminder to daily rather than after one week may improve staff compliance. The delayed interaction with the software poses the question is the current evaluator tool a barrier to failing students? As noted by Bush et al (2013), where tools lack objectivity and explicit evaluation can reduce engagement for staff and thus value of tool. 42 Tutors presented with a range of clinical experience in dental education, with several over 10 years of experience, demonstrate a bias to their own previous teaching styles and impact on engagement with software due to time contraints. 45,47 The utilisation of general dentists as tutors provides them with a range of clinical supervisors with experience outside the hospital setting. However, with some tutors only supervising half a session per week, interaction on an institutional training and the study highlighted further training opportunities that have been addressed. The need for continued assessment training is a current theme within clinical education to ensure the standards are maintained. 28,43

| Limitations of the study
The authors acknowledge the limitations of the results as a singlesite evaluation, and further multicentred research would allow a more comprehensive overview on the current assessment process in dental education.

| CON CLUS ION
Continuous assessment of performance is a paramount pillar of competency-based education and is utilised in undergraduate and postgraduate training within dental education. The study demonstrated a training needed in the utilisation of iDentity and understanding of its integral role in the assessment process and individual evaluator development. The utilisation of standardised software enables all students to be assessed against the same core standards providing high volume of data to evaluate student's progression across a range of clinical disciplines as a clinical logbook of competence. However, in order for the software to be utilised to its full potential, staff must be confident to fail and fairly grade the students on their clinical activity.
The desire for some staff to shield the students from failure often comes from a foundation of good intentions, trying to protect or motivate the student. However, if each tutor provides the student the benefit of the doubt, the student in turn is unable to assess their own weakness and believe themselves to be competent, and potentially overconfident. 28,29 Without this written evidence, the students may "slip through the net" with no supporting evidence of concern available at yearly student progression meetings. The study highlights the continued barriers within healthcare education, and the need continued the discussion for formalised training nationally in dental education.
Due to this phenomenon, the students with significant incompetence (clinically or professionally) may become incompetent tutors when graduating. The need for honest and accurate assessment has a significant benefit to the student and staff, enabling targeted development to motivate the students and improve the quality of care provided to the patients.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

AUTH O R CO NTR I B UTI O N S
C Dixon and R Roudsari have both made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; and have been involved in drafting the manuscript or revising it critically for important intellectual content; and have given the final approval of the version to be published. Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.