Autopsy-Based Digital Cases Boost Medical Students’ Pathology Scores in Controlled Study

Autopsies have long been the quiet backbone of medical knowledge, the place where clinical suspicion meets anatomical truth. Yet in modern curricula, autopsy teaching has steadily faded, squeezed by shrinking autopsy rates, crowded timetables, and the logistical difficulty of bringing large classes into a mortuary or specimen room. A new controlled educational study from Shantou University Medical College, published in BMC Medical Education, suggests that digitizing the autopsy experience and weaving it into undergraduate pathology teaching can measurably improve how medical students reason through cardiovascular disease. The research, led by Zhijin Lei, Junwei Chen, Bin Zhang, and colleagues under the correspondence of Runhua Lin, reports consistent gains in stage assessment scores across two independent student cohorts, alongside rich qualitative feedback about how the approach changed the way students thought about disease.

The study addressed a familiar complaint in pathology education: that the subject is often taught in fragments. Students memorize gross images and microscope slides, but the connective tissue between what a clinician sees, what the pathologist sees, and why a patient died can remain abstract. The Shantou team built a teaching model around autopsy-derived digital case resources, converting real autopsy material into digital assets that could be deployed before, during, and after class. Rather than replacing the autopsy, the approach aimed to scale it, allowing every student to walk through a complete clinicopathological narrative without the constraints of physical access, specimen availability, or biosafety limitations that increasingly constrain traditional autopsy demonstrations.

The pedagogical core of the intervention was a four-step model. First, case introduction presented students with the clinical context of a deceased patient, including history and presenting problems. Second, lesion observation and analysis guided learners through the gross and microscopic findings, now available in digital form. Third, clinicopathological correlation forced the explicit linkage between the lesions observed and the clinical events that produced them, the reasoning step that educators most often identify as the weakest link in conventional teaching. Fourth, summary and reflection asked students to consolidate what the case revealed about disease mechanisms and diagnostic logic. The digital resources supported each phase, so students could preview material before lectures, interrogate lesions during class, and revisit the full case afterward for revision.

To test whether this model actually mattered, the researchers ran a two-cohort controlled study within the undergraduate cardiovascular pathology module. The first cohort comprised 87 students in the intervention group and 87 in the control group; the second, larger cohort included 134 students in each arm. Control students received conventional lecture-based teaching supported by static pathological images and routine slide observation, the standard diet of most pathology courses. Intervention students experienced the autopsy-based digital case model. The primary quantitative outcome was the stage assessment score, a measure of performance within the module, allowing a direct comparison between the two teaching approaches in each cohort and then across both combined.

The results were consistent and statistically robust. In the first cohort, the intervention group achieved a mean stage assessment score of 85.40 with a standard deviation of 15.61, compared with 78.05 plus or minus 22.82 in the control group, a difference that reached statistical significance at P equals 0.014 and corresponded to a Cohen’s d effect size of 0.38. The second cohort replicated the pattern almost exactly: 86.64 plus or minus 15.07 for the intervention group versus 79.48 plus or minus 22.62 for controls, with P equal to 0.003 and a Cohen’s d of 0.37. When the two cohorts were pooled, the intervention group remained ahead, 86.15 versus 78.91 on average, with P below 0.001 and an effect size of 0.37. Effect sizes of this magnitude are modest by conventional benchmarks but meaningful in educational research, where dozens of uncontrolled variables, from prior attainment to test anxiety, typically dilute measurable differences.

Importantly, the researchers checked whether the effect was an artifact of one particular year or group of students. The group-by-cohort interaction test returned a P value of 0.960, meaning there was no statistical evidence that the size of the between-group difference varied across the two cohorts. In practical terms, the benefit of the autopsy-based digital approach appeared stable across independent groups of students taught in different iterations of the module. Replication across cohorts is a strength that many single-class educational studies lack, and it lends credibility to the conclusion that the teaching model, rather than some idiosyncrasy of one student population, drove the improvement.

Beyond the numbers, the study captured how students experienced the new model. Intervention-group students completed an anonymous post-course questionnaire spanning eight learning domains. The domains rated most consistently highly across both cohorts were clinicopathological correlation, diagnostic and differential reasoning, and understanding of disease processes. These are precisely the higher-order skills that pathology educators struggle to cultivate with static images alone. A student can memorize what a myocardial infarction looks like on a slide, but understanding why that infarction produced the clinical picture, the complications, and ultimately the autopsy findings requires a narrative integration that the digital case format appears to supply.

A keyword-based thematic analysis of students’ written learning experiences added a subtle layer of insight. The two cohorts emphasized different things. In the first cohort, interest-oriented and practice-oriented expressions dominated, suggesting the digital cases sparked curiosity and a sense of connection to real clinical work. In the second cohort, thinking-oriented, integration-oriented, and discussion-oriented expressions were more frequent, indicating that students engaged more deeply with reasoning and collaborative analysis. The authors do not over-interpret this shift, but it hints that the model can support multiple dimensions of learning simultaneously, from motivation to metacognition, and that different groups may take different pathways through the same material.

The technical achievement underlying the study deserves attention in its own right. Creating autopsy-based digital case resources requires careful digitization of gross specimens, documentation of microscopic findings, and the assembly of complete case narratives that preserve the ethical dignity of the individuals who donated their bodies to medicine. The authors acknowledge the colleagues who supported the organization and digitization of the pathology case resources, and the work was funded by the Teaching Reform and Research Project of Shantou University Medical College and a 2025 education and teaching research project of the Guangdong-Hong Kong-Macao Greater Bay Area University Online Open Course Alliance. The study received ethics approval from Shantou University Medical College, was conducted in accordance with the Declaration of Helsinki, and collected questionnaire responses anonymously with informed consent from all participants.

The authors are appropriately measured about causality. This was a non-randomized, controlled educational study, so self-selection or cohort effects cannot be fully excluded, and the findings demonstrate association rather than proof of cause. Even so, the consistency of the quantitative gains, the replication across two independent cohorts, and the alignment between the measured outcomes and the qualitative feedback make a compelling case that autopsy-derived digital case resources offer a practical, scalable route to more integrated pathology teaching. As medical schools worldwide grapple with declining autopsy rates and the challenge of teaching genuine diagnostic reasoning, the Shantou model suggests that the autopsy’s educational value need not die with its decline in the mortuary. Digitized, structured, and deployed across the learning cycle, the complete clinicopathological story that only an autopsy can tell may become one of the most powerful tools in the undergraduate pathology classroom.

Subject of Research: Autopsy-based digital case resources in undergraduate cardiovascular pathology education

Article Title: Integrating autopsy-based digital case resources into undergraduate pathology teaching: a two-cohort controlled study in cardiovascular pathology

Article References: Lei, Z., Chen, J., Zhang, B., Zhao, S., Xu, Z., & Lin, R. (2026). Integrating autopsy-based digital case resources into undergraduate pathology teaching: a two-cohort controlled study in cardiovascular pathology. BMC Medical Education. https://doi.org/10.1186/s12909-026-10549-4

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10549-4

Keywords: autopsy, digital case resources, pathology teaching, cardiovascular pathology, medical education, clinicopathological correlation, diagnostic reasoning, undergraduate medicine, instructional design, two-cohort study, Shantou University, BMC Medical Education