A Longitudinal Spaced‐Learning Approach to Support Faculty Calibration in Periodontal Diagnosis

Faculty development is essential in ensuring consistent, evidence-based clinical teaching in dental education. Yet, faculty often face challenges in accessing time-efficient, relevant, and scalable professional development opportunities [1]. This is especially relevant when faculty vary in their knowledge and experience to a dental school's curricular goals such as teaching the new classification scheme on periodontal diseases [2]. Although periodontal diagnosis is a foundational skill and commonly emphasized in everyday clinical instruction, studies on structured, asynchronous faculty development models in this area are limited. While traditional approaches rely on in-person workshops or lectures [3], these may not accommodate diverse faculty schedules, clinical responsibilities, or varying levels of subject familiarity. A web-based, asynchronous course was conducted from August to December 2023 at a University. This study used a longitudinal, assessment-driven model to promote self-directed faculty development through repeated information and immediate feedback on periodontal diagnostic content [4]. The course consisted of a 10-item online pretest, three monthly questions (September–November), and a 10-item posttest, all focused on basic periodontal diagnosis. (Figure 1). We utilized multiple best practices for adult learning strategies in this course. The asynchronous format and short assessments allowed continuous reinforcement of core concepts while minimizing intrusion into faculty schedules. Because these monthly assessments occur in a low-stake environment, faculty can honestly appraise between their perceived and actual knowledge. Therefore, assessment questions, particularly when followed by instant feedback, allow opportunities for individual reflection and knowledge improvement. An additional innovative feature of this course was that we tailored its delivery method to the needs of busy faculty with diverse academic schedules. The course allowed learner engagement any time of the day, such as a faculty member taking this course after hours. This provided faculty flexibility to engage with the content at a time that best fits their own schedule and learning to be sustainable rather than one-time, high-stake assessment or resource heavy event. To maintain question quality, one periodontist created assessment questions testing higher-level learning skills, and another periodontist checked these for relevance and validity. We also assessed statistical performance of questions such as the ability to distinguish low and high performing participants. Engagement was measured by recording time-stamped responses of participants to assessment questions. To ensure feasibility within a limited time frame and minimal disruption to faculty's schedule, all assessment items were designed to be completed within one minute. The format was restricted to multiple-choice items without supplemental information such as radiographs or periodontal charting. All questions were constructed to assess fundamental knowledge in periodontal diagnosis, emphasizing recall and simple application, instead of higher-level cognitive functions that required more time or different format. The questions embedded learning directly into calibration process by asking either common clinical scenarios or testing on straightforward knowledge. All questions were provided with immediate feedback with rationale and links to references for further exploration (Figure 2). All assessments were delivered through Microsoft Forms (Microsoft Corporation, Redmond, WA). No new lectures or separate studying materials were given to isolate the effect of test-enhanced learning on existing faculty knowledge. All licensed dental faculty who teaches clinical and preclinical courses were invited via email to voluntarily participate; 17 initially volunteered and 16 completed the course. Participants received 3.0 CE credits for completion. Participants had one month to complete the monthly question set and received automated email reminders on the first day of the month. If participating faculty did not complete all assigned questions at three-week timeline, individual reminder email was sent at 7 days prior, 3 days prior, and on the due date. After completing the course, participants filled out an anonymous 15-item survey via Qualtrics, evaluating course content, perceived learning outcomes, and overall experience using a 5-point Likert scale. Posttest scores (mean = 8.13 ± 1.36) were higher than the pretest scores (mean = 7.41 ± 2.6), though not statistically significant (paired student t-test, p = 0.31) (Figure 3). However, score variance decreased significantly (F-test, p = 0.016), suggesting more consistent performance among faculty by course end. This suggests the course successfully reached a goal of reducing variance and inconsistency in periodontal diagnosis within participating faculty. Fleiss Kappa increased from 0.50 to 0.62, reflecting enhanced agreement in posttest responses, supporting the potential of this model in calibrating and promoting reliable diagnostic reasoning longitudinally. Four faculty with the lowest pretest scores improved, while five with initially high scores declined slightly—potentially due to decreased engagement or interest over time. The testing effect [5] most likely helped the low performing faculty by targeting deficiency in their understanding and followed by learning from immediate feedback and rationale. Therefore, they were able to self-assess continuously and engage in self-directed learning right after, contributing to the overall group performance. Survey results (93.8% response rate) showed that 100% of participants agreed the questions were relevant and effective learning tools. Most (86.7%) preferred the longitudinal format over the traditional one-time course, and 80% reported improved confidence in diagnosing periodontal conditions (Table 1). Fewer than half (46.7%) saw benefit in anonymous peer comparison, suggesting that self-assessment alone may be more motivating for this cohort. While most agreed on accessing the rationale and thought it was beneficial, only 46.7% of participants reported they have accessed the reference link which required additional steps beyond the immediately available rationale. The additional step of clicking, logging in, and reading extra material likely required more time and effort than simply reviewing the provided explanations. Given that the rationales provided were likely sufficient, participants might not have felt the need to seek additional information. This highlights the importance of immediate feedback and most reinforcement of knowledge and deep learning may come from the rationale part. 1 (6.7%) 1 (6.7%) 1 (6.7%) 1 (6.7%) 4 (26.7%) 4 (26.7%) 1 (6.7%) 3 (20%) 3 (20%) 1 (6.7%) 4 (26.7%) 3 (20%) 4 (26.7%) 4 (26.7%) 4 (26.7%) 3 (20%) 1 (6.7%) 1 (6.7%) 3 (20%) 1 (6.7%) 1 (6.7%) 1 (6.7%) 1 (6.7%) This pilot shows the potential that a spaced, assessment-driven model can be an effective and low-burden approach to dental faculty development and calibration. It supports individual growth, encourages consistent knowledge recall, and provides actionable data for continuous faculty development. Limitations include small sample size and short duration. Implementing a longer course in the future that allows for individualized learning, difficulty levels, and retakes can greatly enhance learning outcomes. This approach will also enable the identification and targeted support of individuals who require additional resources and assistance. The authors received no specific funding for this work. The authors have nothing to report.