2.1 Participants
The study protocol was approved by the medical ethics committee of School & Hospital of Stomatology, Wuhan University. Conducted during the academic year 2022-23, the study involved first-year graduate dental students at the School of Stomatology, Wuhan University, who did not exhibit color blindness or other cognitive impairment. A total of 80 participants were selected and randomly assigned into two groups: the experimental group, taught with CBL (n=38) and the control group, taught with LBL (n=42). There was no statistical difference between the two groups in terms of gender, age, entrance achievement, self-study ability and subject preference (p>0.05). Table 1 illustrates the general characteristics of the 80 participants.
Table 1. Distribution of Number, Gender, and Age (Mean ± Standard Deviation) of Two Groups of Participants
|
CBL group
|
LBL group
|
Gender
|
Number
|
Age (Mean ± Standard Deviation)
|
Number
|
Age (Mean ± Standard Deviation)
|
Male
|
15
|
23.86±0.74
|
17
|
23.41±1.06
|
Female
|
23
|
23.61±1.12
|
25
|
23.40±0.76
|
2.2 Design
In the LBL group, students were received instruction through traditional lectures held in a classroom setting, covering a variety of topics related to DF. This course aimed to provide comprehensive understanding of the etiology, pathophysiology, clinical manifestations, diagnosis, operational guidelines of various evaluative methods for DF severity, with particular emphasis on VAS. The traditional learning sessions lasted $45min per course, with a total of 4 courses delivering the entirety of the teaching content.
Students in CBL group participated in small group sessions. Prior to these sessions, the instructor, with the assistance of physicians from the affiliated hospital, selected typical DF patients exhibiting varying severity and obtained their permission for disease history inquiries and clinical examination by students. Students were encouraged to engage in self-directed learning of DF-related information. During the teaching sessions, students conducted oral clinical examinations, analyzed each case, assessed severity using VAS and discussed factors contributing to the severity. At the end of each session, instructors summarized and evaluated students' discussion outcomes, highlighted key diagnostic points, addressed any teaching-related issues encountered, and solicited feedback from students for course evaluation. Similar to the control group, teaching sessions in the experimental group also lasted 45min per course, totaling 4 courses.
2.3 Effectiveness assessment
The effectiveness and satisfaction of the two teaching methods were evaluated by presenting the identical DF cases to students in both groups to assess their clinical decision making. Each evaluator received a written informed instruction detailing the study's purpose and procedures and signed a consent form to participate in the study.
2.3.1 Selection of DF cases
DF cases were selected from patients visiting the department of prosthodontics, Hospital of Stomatology, Wuhan University,between January 1st, 2022 and December 30th, 2023. The maxillary incisors and canines of these patients were evaluated and diagnosed as DF by prosthodontics specialist according to medical history collection and clinical oral examination. Each patient signed an informed consent form after receiving an explanation of the study's aims and procedures at the study's outset.
The inclusion criteria were as follows:
- Age 19-38
- Presence of two fluorosed maxillary central incisors
- No carious or non-carious lesion in these two teeth
- No filling or prosthesis in these two teeth
- No periodontal inflammation
- Acceptance to be taken frontal view photography of incisors
Exclusion criteria included:
- Poor general or dental health
- Pregnant or lactating
- Smoking habit
- Use of any fixed orthodontic appliances
- Previous use of bleaching agents
2.3.2 Photography
All intraoral photographs were taken using a Nikon D700 camera and Nikon 105mm-macro lens, with consistent lighting conditions. The magnification was set at 1:2, with an aperture of F25, a shutter speed of 1/125, and flash intensity of M/4. To expose the incisors, canines, and premolars, patients' lips were held open with a mouth gag, and only one photographer captured the images to ensure consistency. The photographs were subsequently transferred to a personal computer-based system for storage and viewing. Photoshop software was utilized to crop each image to focus on a single maxillary central tooth (wide 3.4mm and height 3.8mm). Totally, 105 images were obtained and evaluated by two prosthodontics experienced in DF assessment, following the grading standard of TSIF. From these images, 32 teeth with TSIF scores ranging from 0 to 7 were randomly selected and arranged for further analysis (Fig.1).
2.3.2 Evaluation
All the students in both groups rate the severity of each case presented through a slideshow, with each case displayed for ten seconds. Using a 100 mm horizontal VAS, students placed a vertical mark to indicate their assessment of each case. The left end of the scale represented the best or ideal tooth, while the right end represented the worst or unacceptable tooth. The photographs were shown in random order to each evaluator to minimize bias.
We measured the distance between the left end and the marked point made by each student for every photograph. Two weeks later, five students were randomly selected from each group and then rated the cases once again.
2.4 Data Analysis
Statistical analyses were conducted using SPSS 28.0 (SPSS, Chicago, IL, USA) for WINDOWS. The results of VAS scores were presented as mean ± SD. Two-way ANOVA were analyzed differences between groups and gender for all 32 cases. P < 0.05 was considered statistically significant.
2.4.1 Reliability
Reliability, in the context of this study, refers to the consistency of scores over time, between different evaluators, or across different tasks or items measuring the same construct. It is typically assessed using a reliability coefficient, which measures the proportion of variance in scores that is related to the true variance between the objects being measured. In this study, reliability, including both inter-examiner agreement and intra-examiner agreement, was assessed using the intra-class correlation coefficient (ICC) [13, 14].
Inter-evaluator reliability refers to the consistency of ratings among different evaluators. The Intra-class correlation coefficient (ICC) is utilized to measure the level of agreement among evaluators for each group. In this study, both "single ICC" and "mean ICC" were calculated. Single ICC represents the reliability for a single evaluator, while mean ICC represents the reliability for multiple evaluators. Furthermore, ICC was employed to examine whether any differences existed in how the CBL group perceived DF compared to the LBL group.[15].
Intra-evaluator reliability assesses the consistency of ratings made by the same evaluator over time. Five evaluators from each group were asked to repeat the test and rate the cases again. ICC was used to evaluate the intra-evaluator reliability for each of these five evaluators. Pearson correlation coefficients (R) were applied to assess the correlation of each evaluator's repeated evaluations, with benchmarks provided by Landis and Koch (<0.4 poor agreement; 0.4-0.6 moderate agreement; 0.6-0.8 substantial agreement; > 0.8 excellent agreement)[16, 17].
2.4.2 Validation
Validity refers to the accuracy and appropriateness of a measurement tool in assessing the intended construct[18]. The Spearman rank correlation coefficient (RS) is often used to assess the validity of a measurement by examining the strength and direction of the relationship between its scores and those of a criterion measure[19]. Higher RS indicate stronger validity, suggesting that the measurement tool reliably reflects the underlying construct[18, 20].
Validation of mean scores: Spearman rank correlation coefficients between the mean VAS score of each case in relation to DI and TSIF of that case;
Validation of total scores: Spearman rank correlation coefficients between the total VAS scores of each case assessed by all evaluators in relation to DI and TSIF of that case;
Validation of each evaluator: Spearman rank correlation coefficients between the VAS scores of each case assessed by each evaluator in relation to DI and TSIF of that case.