Modern Moulage:  Evaluating the Use of 3-Dimensional Prosthetic Mimics in a Dermatology Teaching Program for Second-Year Medical Students FREE

In most medical schools, limited training time is dedicated to skin biology and clinical dermatology. Sixty-nine percent of medical students surveyed agreed that insufficient emphasis in their medical training was placed on learning the skin examination, and only 28.2% rated themselves as somewhat or very skilled in the skin examination.¹ Furthermore, core curricula are based on print media and lectures with 2-dimensional (2D) images, and it is unknown whether content learned through this approach transfers to practical long-term recognition of important lesions and eruptions. It has been shown that although medical students accurately identify melanoma in clinical photographs, they may not transfer this knowledge into identification of suspicious lesions during standardized patient examinations.² As such, it is important to develop effective standardized methods that improve long-term learner outcomes without using more curricular time and resources to ensure that medical students graduate with specific and standardized competencies in dermatology.

Clinical instruction reaching an entire class typically occurs in the form of large-capacity lectures with projected 2D images. However, lesions and eruptions are 3-dimensional (3D) structures with clinical features best appreciated by touch and subtle visual clues that are often lost in 2D photographs. Based on this concept, mouleurs of the Renaissance era produced detailed morphologic wax casts that were used as teaching tools throughout Europe.^3- 4 Although now an almost forgotten art, instruction through moulage paralleled the growth of dermatology as an independent specialty during the 19th century. Similarly, we explore a teaching method that uses 3D virtual mimics of common serious lesions and eruptions to train preclinical students in acquiring dermatologic recognition and initial management skills. The aims of the present study are to evaluate the student learning response to using 3D models, to compare that response to evaluation of traditional lectures using 2D images, and to measure learning outcomes before, immediately after, and 3 months after the teaching interventions.

This study received approval by the human subjects committee of the institutional review board at the University of Massachusetts Medical School. This study represents a collaboration between the Biology of Disease course, an organ-based course focused on the basic science, physiology, and clinical spectrum of normal and abnormal integumentary processes, and the Physical Diagnosis II course, which teaches the principles and practice of physical examination.

Second-year students at the University of Massachusetts Medical School were randomly assigned to 1 of 2 intervention groups: 3D (n = 49) or 2D (n = 41). Both cohorts underwent baseline assessment (T0) using 2D images that assessed recognition and basic management skills for several common serious lesions and eruptions. This pretraining assessment was followed by a single 1-hour clinical teaching intervention in which the 2D cohort underwent a traditional content-based lecture using projected 2D images and in which the 3D cohort underwent a clinical teaching session using 3D nontoxic, hypoallergenic platinum-based silicone prosthetic mimics (MEDART-FX Inc, New York, New York) of lesions and eruptions affixed to a standardized patient (Figure). All the students were assessed using a computer-based module and 2D images immediately (T1) and 3 months (T2) after their respective teaching intervention. Students took assessments at individual computer terminals. The order of the questions was the same for all the students, and each module contained unique clinical images. All 3 modules consisted of only multiple-choice questions, and the modules were equivalent in their level of difficulty. The T2 assessment was expanded to include more questions to better discriminate between the cohorts' retention of content across time.

Primary outcome measures at each assessment included morphology recognition, lesion and rash recognition, lesion and rash initial management, and overall performance, which is an aggregate score of all the performance measures. Morphology questions assessed a student's ability to correctly identify primary morphologic features from images. Lesion and rash recognition questions assessed a student's ability to correctly identify common serious diagnoses from 2D images. Lesion management questions assessed a student's ability to correctly identify the initial management (ie, biopsy or reassure) of a lesion. Rash management questions assessed a student's ability to correctly identify the initial management (ie, administration of an anti-inflammatory, antibiotic, or antiviral medication) of an eruption. Table 1 lists common serious lesions and eruptions on which the teaching content and assessments were based. Along with the T1 assessment, students were surveyed anonymously regarding general impressions of their respective instructional sessions using a 5-point Likert scale.

One-way analyses of variance were conducted to determine differences between the 2D and 3D groups in each period. Improvement across time was calculated for each group using paired t tests. Subject content, teaching time, and assessments were the same for both cohorts. To minimize repeated testing bias, students were not given correct responses or scores for any of the 3 assessments. None of the study participants had previous instruction in dermatology. After the final assessment, each cohort was offered the other's training intervention so that each student would ultimately have had the same learning experiences.

All T0, T1, and T2 outcomes are summarized in Table 2. At baseline (T0), tests of knowledge showed that mean percentage scores for overall performance, morphology, lesion and rash recognition, and lesion and rash management were similar and not significantly different between the 3D and 2D groups before any teaching intervention, as tested using analysis of variance.

Both groups showed a similar pattern of significant improvement from T0 to T1 in paired t tests. One-way analysis of variance found that the 3D group had significantly higher T1 mean percentage scores than the 2D group for overall performance, lesion recognition, and rash management (Table 2). The mean percentage scores at T1 for morphology, rash recognition, and lesion management were the same or higher in the 3D group than in the 2D group, but these differences did not reach statistical significance.

Both groups showed a similar decrease in scores 3 months after the single teaching intervention (Table 2). At T2, the 3D group still had a significantly higher mean percentage score than the 2D group for lesion recognition and a higher mean percentage score for overall performance that neared significance compared with the 2D group. The mean percentage scores for morphology, rash recognition, lesion management, and rash management were higher in the 3D group compared with the 2D group, but an analysis of variance indicated that these differences did not reach statistical significance.

Paired t test subanalysis of lesion recognition scores showed that the 3D group had a significant increase in mean score from T0 to T1 and a significant decrease in mean score from T1 to T2. The 3D group also showed a ignificant increase in mean score from T0 to T2, suggesting that the 3D group was better able to recognize lesions at 3 months compared with at baseline. Likewise, the 2D group showed a significant increase in mean score from T0 to T1 and a significant decrease in mean score from T1 to T2. However, the difference in 2D mean scores between T0 and T2 was not significant, suggesting that the 2D group was no better at recognizing lesions at 3 months compared with at baseline.

The 3D cohort agreed or strongly agreed that the 3D session was (1) more enjoyable than classroom lectures (96%), (2) effective in facilitating recognition of lesions and rashes (94%), and (3) perceived to be a more effective teaching method than lectures with 2D images (94%). The 3D cohort also believed that the prosthetic pieces (1) appeared realistic enough to learn from (79%), (2) facilitated close examination of specific morphologic features (92%), and (3) had a learning benefit that outweighed their artificiality (96%). In the 2D cohort, 42% of the students agreed or strongly agreed that it was difficult to examine specific morphologic features of lesions and rashes projected as 2D images, and another 30% were neutral on this issue.

As a modern perspective on the practice of moulage, the present method using silicone-based 3D prosthetic mimics of common serious lesions and eruptions resulted in significantly improved immediate clinical skills acquisition with respect to lesion recognition, rash management, and overall performance compared directly with traditional lecture-based teaching with projected 2D images. Regarding long-term skills retention, the 3D group significantly outperformed the 2D group with respect to lesion recognition 3 months after a single teaching intervention. Furthermore, the 3D group better recognized lesions at 3 months compared with at baseline, whereas the 2D group was no better at recognizing lesions at 3 months compared with at baseline. We took particular notice of lesion recognition performance for 2 reasons. First, we believed that any 1-hour teaching intervention, regardless of method, would be limited in its ability to demonstrate long-term retention of knowledge 3 months after the intervention. This result suggests that despite limited curricular time, use of the 3D method may effectively facilitate long-term retention of lesion recognition skills. Second, we believed that long-term retention of lesion recognition skills was the most essential of all the skills because initial recognition of melanoma or basal cell carcinoma, for example, would ultimately lead to intervention, either by a primary care physician or via referral to a dermatologist. The remaining immediate and long-term learner outcomes in the 3D group were at least as good as those in the 2D group. This study reinforces that lesion management, rash recognition, and rash management are likely higher-order skills that require a “higher dose” of intervention to facilitate acquisition and retention of knowledge. These skills should continue to be targeted in later years of medical school. Although the 3D group realized statistically significant improvements at 3 months, absolute scores at 3 months (T2) for both cohorts were lower than were scores immediately after the teaching session (T1). We believe that this finding underscores the need for additional teaching and reinforcement of dermatologic skills after students complete the core curriculum in the second year. Because most students do not take a clinical elective in dermatology, we advocate including this 3D session in third-year clerkships as part of the Objective Structured Clinical Examination.

The general impressions and attitudes toward the 3D method were highly favorable among students. The 3D method was thought to be enjoyable, effective, and authentic, and it was also believed to have offered advantages over 2D learning.

There are several limitations to this study. Prostheses do not represent exact replicas of clinical lesions or eruptions. Prosthetic lesions were of higher fidelity than were eruptions, and this may, in part, account for why students performed better with prosthetic lesions. It is also possible that the unique experience of the novel 3D teaching method may have encouraged students to pay more attention and, consequently, learn more than they otherwise might have through a 2D-based lecture, similar to the way a student may better recall a basal cell carcinoma after having experienced a clinical encounter with such a lesion on a patient. The improvement in 3D learning is modest in some areas. It is also important to underscore that the 3D group scored the same or better than the 2D group in all outcome domains at T1 and T2 despite assessments having been performed via 2D, which we would expect to have favored the 2D group. Outcome measures may become more disparate between the groups with the second-generation of improved prosthetics, an increased number of study participants, more discriminating assessments that are not skewed to favor either group, and an assessment interval that reasonably measures long-term learning resulting from a single 1-hour intervention. Although students were not given answers to assessments and each assessment was composed of unique images, we cannot exclude the possibility of learning due to repeated testing in either group. The most appropriate assessment would have involved clinical evaluation of actual lesions, but this would have been difficult to coordinate for melanoma, for example. Finally, results in this study require further validation in other medical student classes.

To our knowledge, this study is the first of its kind to rigorously assess and demonstrate the value of a prosthetics-based teaching method in dermatology. These findings are important especially in the context of limited curricular time allotments for dermatology. These results provide a framework to further develop and evaluate an undergraduate teaching program that incorporates the prosthetics method across an experience vertically oriented across 4 years.

Correspondence: Amit Garg, MD, Department of Dermatology, Boston University School of Medicine, 609 Albany St, Room J207, Boston, MA 02118 (agarg@bu.edu).

Accepted for Publication: August 14, 2009.

Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Garg and Hatem. Acquisition of data: Garg. Analysis and interpretation of data: Garg, Haley, and Hatem. Drafting of the manuscript: Garg. Critical revision of the manuscript for important intellectual content: Garg, Haley, and Hatem. Statistical analysis: Haley. Obtained funding: Garg. Administrative, technical, and material support: Garg and Hatem. Study supervision: Garg and Hatem.

Financial Disclosure: None reported.

Funding/Support: This study was supported by an Innovations in Medical Education Grant and by the Association of Professors of Dermatology (Dr Garg).

Role of the Sponsors: The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.