Learning What Matters Through Case Comparison in Undergraduate Nursing: A Randomised Controlled Trial of Reasoned Clinical Decision Transfer

Background and rationale: In clinical education, recalling the correct answer to a case is not the same learning outcome as identifying which finding changes the recommended course of action in a new situation. Tanner's model of clinical judgment emphasizes noticing, interpreting, responding, and reflecting, and this study focuses on the noticing, interpreting, and priority-response components. Case-comparison (analogical encoding) research suggests that comparing paired examples can help learners extract the structure that determines a decision and apply it to new problems, but this evidence is not specific to nursing students, and a meta-analysis indicates that the benefit of case comparison depends on how the comparison is structured. This trial therefore tests, under content- and time-matched conditions, whether an explicit case-comparison exercise improves near transfer of reasoned clinical decisions relative to individual case reasoning; it does not assume the approach will work in nursing before it is tested.

Design: This is a two-arm, parallel-group, 1:1 individually randomized, active-controlled, outcome-assessor-blinded educational trial conducted at a single site (Department of Nursing, Faculty of Health Sciences, Agri Ibrahim Cecen University) with undergraduate nursing students (Years 1-4, age 18 and older). There are no real patients and no clinical interventions; all training and assessment cases are fictional. Training is delivered with printed, coded booklets in a supervised classroom; no digital platform, audio/video recording, or tracking software is used, and there is no qualitative interview component.

Procedure: After consent and a baseline assessment of four cases, eligible volunteers are individually randomized (1:1, permuted blocks of 4 and 6, stratified by class year and session, with allocation concealed in sequentially numbered, opaque, sealed envelopes managed by staff independent of the investigators). In the comparative-case arm, students review the eight training cases as four side-by-side pairs, write their priority approach to each case, and then write the decision-relevant difference between the paired cases and the general principle it illustrates. In the individual-case arm, students review the same eight cases one at a time and write their priority approach and rationale for each. Total content, response burden, and feedback are equal between arms. Approximately seven days later (acceptance window Day 5-9), all participants complete eight new transfer cases presented one at a time, without any comparison prompt, in one of two counterbalanced orders.

Outcome measurement: The single primary outcome is the Day-7 reasoned clinical decision transfer score: each of the eight transfer cases is scored 0-3 with an integrated rubric by each of two independent, blinded raters (working from de-identified, coded response sheets), the two raters' totals (0-24) are averaged and rescaled to 0-100. Secondary, exploratory measures derived from the same responses include the proportion of appropriate first-approach decisions, the rate of delayed help-seeking on urgent cases, and the rate of unwarranted escalation on non-urgent cases. Before the main data collection, training and assessment materials undergo independent expert review (content/clinical-safety panel) and a small two-phase student pilot (cognitive interviews, then a feasibility run, maximum 24 students, not included in the main trial) against pre-specified acceptance criteria; materials that do not meet these criteria are revised before the main randomization.

Analysis: The primary analysis is an ANCOVA of the Day-7 transfer score by randomized arm, adjusted for baseline score, class year, and session block, with a two-sided alpha of 0.05, reported as the adjusted mean difference with a 95% confidence interval. The trial targets 360 randomized students to allow for an anticipated 15% loss to follow-up while retaining approximately 306 evaluable students (approximately 130 evaluable students per arm), which was judged adequate for the pre-specified planning difference. Missing follow-up data are handled by multiple imputation under a missing-at-random assumption, with pre-specified sensitivity analyses (complete cases, an extended follow-up window, and pattern-mixture scenarios). A time-stamped analysis plan and registration (OSF) will be finalized before the first participant is randomized.