1 INTRODUCTION

The combination of theoretical knowledge and acquired manual skills is essential in dentistry. Thus, motor skill training is a crucial part of the dental curriculum. Little has changed in recent decades in the way that dental students are trained to acquire fine motor skills by using phantom models in the simulation clinic. In reconstructive dentistry, undergraduate students are trained using dental patient simulators where they can practice treatment with full crowns.1, 2 Exercises involving the preparation of crowns and grinding of artificial teeth can be frustrating for the students and the repetitive nature of the task can negatively impact motivation.3 Nevertheless, repeatedly practising pre-clinical techniques such as grinding the same tooth to enhance accuracy, quality and efficiency of tooth preparation is necessary to reach the level of skill needed for the student transfer to the clinical course with patient treatment. For the instructors, the repetitive work of evaluating identical tooth preparations is also tiring. Instructor calibration is mandatory to ensure an objective assessment of student's work, which is best conducted in a blinded approach.4-7

Digitalisation is a ubiquitous trend. Intraoral optical scanning (IOS) in particular has changed the clinical dental workflow.8, 9 Besides its routine implementation in disciplines such as restorative and reconstructive dentistry and orthodontics, IOS can be also considered a valuable tool in dental education.10, 11 Dental students can use IOS to capture their tooth preparations and receive objective evaluations.12-15 A recent review analysed the penetration of digitalisation in the dental curriculum and the educational quality enhancements arising from such techniques.10 The application of standardised digital surface mapping was identified as a valuable tool for providing an objective and reproducible feedback of wax-ups or abutment tooth preparations.10 The students of today are considered “digital natives”; therefore, working with digital tools is easy and enjoyable for them.16, 17 Nevertheless, several open questions remain, such as how can digitalisation be used as a standardised learning method within the curriculum? do students learn better with personal feedback face-to-face or are digital assessments more effective? and are faculty instructors still needed and to what extent?

The aim of this prospective cohort study was to investigate the influence of IOS combined with a tooth preparation evaluation software (prepCheck, Dentsply Sirona, Bensheim, Germany) on undergraduate student's performance during motor skill training in a university setting. In particular, the outcome of repetitive preparation of the same tooth and its influence on the acquisition and maintenance of motor skills was evaluated.

The following hypotheses were tested that (i) instructors’ and digital assessments of repeated tooth preparation do not differ, and (ii) visual feedback from IOS technology does not affect motor skill acquisition among dental students.

2 MATERIALS AND METHODS 2.1 Study cohort and experimental design

This study was exempt from oversight by the University… review board. Third-year undergraduate dental students with no previous experience in tooth preparation skills were recruited to the study. Between November 2018 and March 2019, students conducted a training course in abutment tooth preparation, in which typodent teeth (Frasaco AG-3 28) of the maxillary central incisor (FDI #21) were grinded for a veneered all-ceramic crown. The guidelines for crown preparation were taught in the theoretical lessons providing ideal dimensions of substance removal (Figure 1A).

Preparation and assessment criteria for veneered all-ceramic crowns. (a) Silicon key cut in a bucco-oral direction and repositioned on the model indicating preparation dimensions. (b) Criteria for conventional assessment by faculty instructors and students’ self-assessment

The training course comprised six practice sessions (runs) and was concluded with a practical examination. The study cohort was split into two groups (A and B) based on alphabetical order. Three of the six preparation runs took place in the evening after regular class lessons (5 pm–9 pm). Group A conducted runs one, three and six in the evening, while Group B performed runs two, four and five in the evening. The examination took place in the morning for all students with a time limit of 120 min for the abutment tooth preparation. The time of day for the practice sessions (morning or evening) and gender were recorded to determine potential influences on the outcome.

2.2 Conventional assessment of tooth preparation

All conventional assessments were performed by two calibrated faculty instructors, who received the encoded model after each preparation. They also received silicon keys prepared by each student and cut in bucco-oral and mesio-distal direction to visualise the amount of tooth substance removal (Figure 1A). Five predefined preparation criteria were assessed (Figure 1B), and each was graded from 2 to 0 scoring (2: clinically very good; 1: clinically acceptable, crown manufacturable; and 0: clinically unacceptable, crown cannot be produced). The five individual scores were summed up, and an average value was calculated and rounded up or down to the grades 2, 1 or 0. If one of the predefined no-go applied, the overall grading was zero.7

2.3 Digital assessment with IOS technology

A reference tooth (FDI #21) was prepared by a prosthodontic faculty member according to the predefined guidelines presented in Figure 1A. The reference tooth was then scanned and uploaded into prepCheck Wizard software program (prepCheck-application 3.1, Dentsply Sirona) and deemed as “master preparation” for digital comparison. All students’ preparations including the exam preparation were digitalised using an IOS system (Cerec Omnicam, Dentsply Sirona). The scanning process was carried out according to the manufacturer's recommendations and performed by one single experienced operator capturing the prepared tooth and the adjacent teeth. PrepCheck-application 3.1 (Dentsply Sirona) was used to compare and analyse the digitised preparations using the built-in algorithm.18 The parameter “substance removal” was used to analyse abutment tooth preparation and set at a limit of 0.2 mm permissible error (tolerance) compared with the ideal abutment. The built-in algorithm generated a colour-coded visualisation of the preparation areas within, above and below this tolerance range for the “substance removal” parameter. To investigate whether an instructor-student feedback session using the prepCheck system could improve student performance, the preparation from the third practice session was digitised and the colour-coded visualisation was discussed with the student before they continued with the fourth preparation.

2.4 Student's self-assessment and preparation time

Dental students completed a self-assessment of their preparations after each practice session and recorded the time required for each preparation. Self-assessment was scored using the same 2 to 0 scoring system used by the instructors (Figure 1B) using the cross-sectioned silicon key. For the examination, 120 min were allowed for the preparation and no self-assessed was performed.

2.5 Statistical analysis

Mean values and standard deviations (SD) were indicated for the relative areas derived from the digital evaluation parameters “within,” “above” and “below” the tolerance range. In cases of skewed distributions, median values and interquartile ranges (IQR) were also calculated. Counts and percentages were listed for the categorical parameters of conventional assessment and students’ self-assessment.

As verified by quantile comparison plots, the relative area values were analysed by regression analysis on normal scale. To study the time-course of the various evaluation parameters over the six practice sessions and the exam preparation, mixed-effects models were performed with the first preparation as reference. For the digital assessment, improvement was defined as an enlargement of the area deemed to be within the tolerance range. Potential interactions between findings and gender as well as the time of day for the practice sessions (morning or evening) were investigated.

To compare the agreement rate between student's self-assessment and instructor's assessment, the Spearman's rho coefficient was calculated. Moreover, the percentage of conformity was indicated.

All regression models were adjusted for gender. The level of significance was set at α=0.05. An adjustment of the significance level for multiple comparisons was omitted, because of the descriptive nature of the study. All analysis was performed with the statistical program R version 3.5.1.19

3 RESULTS 3.1 Conventional and digital assessment

Data were collected from 26 students (7 male and 19 female) who completed a total of 177 preparations. Five absences occurred during the practice sessions due to sickness. All students attended the examination. The highest scores determined by conventional assessment by the instructors occurred at the fifth preparation session, with 43.5% of the students received a score of 2. For the examination, 38.5% of students received score 2 (Figure 2A).

Conventional assessment of repetitive tooth preparation. (a) Assessment by faculty instructors. (b) Self-assessment by students

The digital assessment (prepCheck) of substance removal is presented in Table 1 and Figure 3A, and representative prepCheck colour-coded visualisations are shown in Figure 3B, C. Statistically significant improvements with higher areas within the tolerance range were observed between the first (24.5 ± 11.9%) and the second run (32.0 ± 10.2%; 7.5% improvement, 95% CI: 2.2%, 12.7%; p = 0.006), and between the first run and the exam preparation (31.2 ± 9.9%; 6.7% improvement; 95% CI: 1.7%, 12.5%; p = 0.011; Figure 3). Preparations from practice sessions three, four, five and six were not significantly improved compared with those of the first run.

TABLE 1. Digital assessment of substance removal for each preparation (mean ± SD) compared with the master preparation using prepCheck software. Preparation session Below tolerance range Within tolerance range Above tolerance range 1 22.5% (±21.1%) 24.5% (±11.9%) 53.0% (±28.7%) 2 30.7% (±15.3%) 32.0% (±10.2%) 37.3% (±19.2%) 3 34.5% (±18.8%) 25.8% (±9.1%) 39.6% (±22.7%) 4 32.6% (±14.0%) 27.2% (±12.4%) 40.2% (±17.4%) 5 42.6% (±18.4%) 28.0% (±8.5%) 29.4% (±19.5%) 6 44.0% (±25.1%) 23.5% (±1.0%) 32.5% (±24.1%) Examination 18.1% (±13.3%) 31.2% (±9.9%) 50.8% (±19.9%)

Digital assessment of repetitive tooth preparation and PrepCheck report of two preparations visualising the difference between study preparation and master preparation. (a) Substance removal (mean %) between preparations and master preparation, error bars indicate SD. (b) Less-invasive preparation with 57% of the area in green indicating not enough substance removal mostly at the palatal, incisal and chamfer site. Red areas are predominant at the buccal site meaning extensive substance removal in 17%. 26% of the area is displayed in light to dark blue indicating substance removal within the set tolerance range. (c) More-invasive preparation with 63% of the area in red. The palatal reduction and chamfer preparation are mostly within tolerance range (26%), whereas only 11% of the preparation is classified as not enough substance removed (green). The percentage distribution of the results in relation to the set target range is displayed on a scale: green =below tolerance range, light to dark blue =within tolerance range, red =above tolerance range

The individual instructor-student feedback-visualisation after the third preparation session had no direct effect on the performance of the fourth run, as indicated by similar areas within tolerance range (1.3%, 95% CI: 3.9%, 6.5% for session three and 2.7%, 95% CI: 2.6%, 7.9% for session four; p = 0.322). For the fifth and sixth preparation sessions, there was a trend towards less invasive preparation.

3.2 Preparation times and student's self-assessment

During the practice period, the average time required for the single crown preparation significantly decreased from 153 min for the first session to 104 min for the sixth session, which represents an improvement of time of approximately 32% (p < 0.001; Figure 4). During the examination, all students used the available 120 min. There was no difference regarding the time of day that preparations were done (morning vs. evening; data not shown).

Time recorded for preparation sessions 1–6; 120 min was given for examination

The best self-assessment score was recorded for the fifth preparation session, when 43.5% of students gave their preparations a score of 2 (Figure 2B). The overall agreement between students’ self-assessments and conventional instructor assessment was 66.1% with variations among the 6 practical sessions (Table 2). The highest agreement (76.0%) with the best correlation (Spearman's rho 0.78) was observed for the fourth run after students had access to the digital prepCheck feedback (Table 2). The correlation was maintained at a higher level also for the fifth and sixth run compared with the first three sessions.

TABLE 2. Correlation between instructors’ and students’ assessments. Preparation Agreement (%) Spearman's rho 1 73.1 0.57 2 61.5 0.51 3 53.8 0.52 4 76.0 0.78 5 73.9 0.63 6 58.3 0.69 Overall 66.1 – 4 DISCUSSION

The current study investigated students’ performance of abutment tooth preparation evaluated by digital analysis and conventional instructors’ assessment compared with student's self-assessment. Data revealed that an overall improvement of the quality of crown preparation was only observed when preparations were scored by conventional instructors’ assessment and could not be reproduced using the prepCheck software. The visual feedback from prepCheck software after the third run had only limited influence on the outcome of the subsequent crown preparations. Visual feedback did appear to improve students’ self-assessment skills, with better agreement to instructors scoring observed after the prepCheck feedback session. A significant training effect with repeated tooth preparations was indicated by reduced preparation time.

A limitation of the current study was the selection of a rather narrow tolerance threshold for the prepCheck assessment set to 0.2 mm. This was possibly too small to visualise improvements between the individual practice sessions; however, this tolerance was selected to train the undergraduates to the high level of accuracy required for patient treatment in the subsequent master course. While different evaluation tools are available with IOS,13, 20 the current evaluation focused on substance removal as the most reliable parameter, which affects all other parameters such as taper and location of the preparation margin. prepCheck assessment provided detailed feedback on too little substance removable, which is easily correctable, and too invasive substance removable representing an irreparable damage. Such a precise reporting was not feasible by faculty instructors who provided more detailed feedback about the quality of the preparation margin and its position, taper, undercuts or damage of adjacent teeth. A further limitation was the restricted access to the digital feedback which was provided only after the third preparation session. Following this feedback, no direct impact was observed in the fourth preparation session, though there was a trend towards a more careful preparation as shown by a higher percentage of less invasive substance removal in the prepCheck assessments of the fifth and sixth preparation sessions. At the examination, students showed an opposite behaviour and prepared more invasively. Examination stress could be an explanation for the more invasive preparation pattern.21

Providing students with unrestricted access to the digital prepCheck feedback would provide continuous feedback and possibly enable a greater impact. Traditionally, motor skills in dentistry were mostly assessed by the outcome, without looking at the learning process.22 Park et al.23 showed that IOS-based tools were especially helpful for weaker students by enabling them to improve their skills in specific aspects. Hence, a tailored approach could help encourage and maintain higher motivation levels over the current repetitive approach. Since the time at which the preparation sessions were conducted (morning or evening) did not significantly impact outcomes, it might be even possible that students could set an individual schedule for their preparation exercises. One must note that providing IOS-based tools for all students comes at high acquisition and maintenance costs that need to be critically assessed.24 In the current study only one digital software tool was available. Since students had no knowledge of the scanning process, and time was limited, scanning was conducted by one experienced instructor. Providing students with permanent access to digital self-scanning and self-assessment would be preferable, particularly with the majority of students today enjoying digital practice.17 In a previous study, in which students performed the scanning process, they complained about the additional time needed for the scanning process while practice time was reduced.25

The time required for abutment preparation in the present study decreased continuously as a result of the repeated practice sessions, indicating a significant improvement in motor skills. During the examination, students used the entire available time slot. The digital assessment with prepCheck showed improvements after the second preparation session and in the exam preparation. A possible explanation was the motivation level being high at the start of the course and rekindled during examination, while lower dedication was given in the interim phase. Previous studies indicated that motivation is a factor influencing motor skill improvement and that high motivation results in a high level of performance.3, 26

A moderate correlation was documented between student's self-assessment of their abutment preparation and instructors’ assessment. Learning to honestly assess one's own performance is an important part of dental education since students will ultimately become independent practitioners, often without further feedback from instructors.27-29 In the present study, the overall agreement between student and instructor scoring was 66.1% with a tendency for students to over-score themselves compared with the instructors’ score. Access to the prepCheck feedback led to an improvement in agreement between students’ self-assessment and instructors’ assessments (76% agreement for the session after visual feedback with the prepCheck system). This indicates that objective assessment tools with digital visualisation could be effective in developing self-assessment skills of dental students supporting motor skill acquisition.

5 CONCLUSION

The results of this study suggested that repeated abutment preparation sessions improved student's performance and reduced the time needed to produce the preparations. When examining repetitive crown preparation, IOS technology with small tolerance range of the prepCheck software can be used for direct feedback of motor skill improvement, while instructors’ feedback incorporates more detailed criteria including preparation margin and location. The prepCheck software is a supportive tool in motor skill training; it improved self-assessment skills and may help to keep students’ motivation high and supports students self-training. Further studies are required to adapt an optimal calibration between conventional and digital assessment suitable for examination assessment within the existing curricula.

ACKNOWLEDGEMENT

The authors thank Urs Simmen for statistical support. The authors also thank Dentsply Sirona, Bensheim, Germany, for making the prepCheck software available.

CONFLICT OF INTEREST

This research received no external funding and there are no conflicts of interest.