Illness Script Training Linked to Higher Self-Reported Diagnostic Thinking in Medical Students

Medical students who completed a structured 12-week illness script–based clinical reasoning program reported stronger diagnostic thinking than peers receiving routine clinical education, according to a cluster-randomized study published in BMC Medical Education. Researchers at Shahid Sadoughi University of Medical Sciences in Yazd, Iran, conducted the trial in 2025 with 128 medical clerkship and internship students and found both higher self-reported diagnostic thinking scores and greater learner satisfaction in the intervention group.

Illness script theory provides a cognitive framework for organizing clinical knowledge and supporting diagnostic reasoning. The approach proposes that clinical knowledge is structured as scripts that learners can consciously build and compare, and structured evidence on illness script–based education among medical clerks and interns has remained limited. The new study was designed to evaluate whether participation in such a program was associated with differences in self-reported diagnostic thinking.

The trial took place at two teaching hospitals affiliated with the university. Before any participants were recruited, the two hospitals were randomly allocated to the intervention or control condition using a pre-specified coin-toss procedure, meaning entire clusters rather than individual students were assigned. Of the 128 students who took part, 65 were in the intervention group and 63 in the control group.

Students in the intervention arm received the 12-week structured illness script–based clinical reasoning program in addition to routine clinical education. Those in the control arm continued with routine clinical education and regular lecture-based teaching. Diagnostic thinking was assessed before and after the intervention using the 41-item Diagnostic Thinking Inventory, a self-report instrument comprising the structure of memory and flexibility in thinking domains. Learner satisfaction was also measured with an eight-item questionnaire.

The adjusted results favored the illness script group. On the post-intervention inventory, adjusted for baseline score, age, sex, and educational level, the intervention group’s estimated marginal mean total score was 151.05 compared with 136.41 in the control group — an adjusted mean difference of 14.64 points, with a 95 percent confidence interval of 9.46 to 19.81 and a p value below .001. The overall adjusted model was statistically significant (F(5,122) = 7.92, p < .001, R² = 0.245), and the study group effect was itself significant (F(1,122) = 29.39, p < .001, partial η² = 0.194).

The researchers also tested the robustness of the primary finding. A sensitivity analysis using a generalized linear model with robust covariance estimates, which guards against heterogeneity of error variances, yielded a consistent group effect, with a coefficient of −14.64 and a 95 percent Wald confidence interval of −19.76 to −9.51 (Wald χ²(1) = 31.30, p < .001). Learner satisfaction was higher in the intervention group than in the control group (35.23 ± 2.25 versus 32.00 ± 2.70; p < .001).

Despite these results, the authors frame the findings as preliminary. Because the study included only two hospital clusters, and because the primary analysis was conducted at the individual level, the design cannot fully separate the effect of the illness script program from potential hospital-level differences — an inherent constraint when the unit of randomization is the institution but the unit of analysis is the student.

The self-reported nature of the outcome adds a further caveat. Higher Diagnostic Thinking Inventory scores reflect how students perceive their own diagnostic thinking, not directly observed diagnostic performance. The authors note that objective performance-based outcomes are needed to determine whether illness script–based education translates into improvements in actual clinical reasoning performance.

Looking ahead, the researchers call for multicenter studies with a larger number of clusters, cluster-appropriate analytical methods, and objective performance-based outcomes. Within those limits, the trial offers preliminary evidence supporting the potential value of structured illness script–based activities in clinical reasoning education for medical clerks and interns.

Subject of Research: Illness script–based clinical reasoning education and self-reported diagnostic thinking among medical students

Article Title: An illness script approach in clinical reasoning education: self-reported diagnostic thinking among medical students

Article References: Gholinataj Jelodar, M., Mirzaei, S., & Keshmiri, F. (2026). An illness script approach in clinical reasoning education: self-reported diagnostic thinking among medical students. BMC Medical Education. https://doi.org/10.1186/s12909-026-10290-y

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10290-y

Keywords: clinical reasoning, illness scripts, medical education, diagnostic thinking, cluster-randomized trial, Diagnostic Thinking Inventory, medical students, clerkship, internship, cognitive framework, illness, script