Generative AI adaptive narratives to enhance nursing diagnostic reasoning: a classroom innovation
Open accessAbstract
BACKGROUND: Nursing students often struggle to apply standardized nursing languages (NANDA-I, NIC, NOC) in a coherent and context-sensitive way. Narrative pedagogy supports clinical reasoning, but traditional patient narratives are static and offer limited variability. Generative artificial intelligence (AI) provides new opportunities to create adaptive, cue-rich simulations grounded in authentic patient experiences. OBJECTIVE: To develop and pilot-test an instructional design using generative AI–enhanced narrative simulations to support undergraduate nursing students’ diagnostic reasoning and NANDA–NIC–NOC linkage. METHODS: A mixed-methods pilot study was conducted with second-year nursing students (N = 46). Authentic coronary patient narratives were transformed into adaptive simulated cases using a structured generative AI pipeline. Students were randomly assigned to a control group (static narrative) or an AI-enhanced group (adaptive narratives). Outcomes included diagnostic accuracy, NIC–NOC coherence, self-efficacy, and qualitative reflections. RESULTS: Students in the AI-enhanced group demonstrated significantly higher diagnostic accuracy, stronger NIC–NOC coherence, and greater self-efficacy than the control group (all p < .001, large effect sizes). Qualitative analysis identified enhanced cue sensitivity, iterative refinement of diagnostic decisions, and increased perceived authenticity as key learning mechanisms. CONCLUSION: Generative AI–enhanced narrative simulation is a feasible and pedagogically valuable approach for strengthening nursing diagnostic reasoning using standardized nursing languages. By introducing controlled narrative variability grounded in real patient experiences, this method supports context-aware and reflective clinical reasoning in undergraduate nursing education. IMPLICATIONS FOR NURSING EDUCATION: Generative AI expands narrative pedagogy by enabling educators to create an unlimited range of realistic, ethically sound and cue-rich scenarios for teaching nursing diagnostic accuracy, clinical reasoning and patient-centred care planning. CLINICAL TRIAL NUMBER: Not applicable.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
These outcomes are particularly evident when learners receive real-time feedback and repeated practice within psychologically safe learning environments.
“These outcomes are particularly evident when learners receive real-time feedback and repeated practice within psychologically safe learning environments [4,8-11].”
From Artificial Intelligence-Based Adaptive Simulation Integrated With a Workforce Decision-Support System to Improve Simulated Nursing Leadership Performance: A Randomized Controlled Study (2026) · cites this paper as [8]
The Results (Quantitative findings; page unavailable) report better diagnostic accuracy, NIC–NOC coherence, and self-efficacy in the AI-enhanced group. The Discussion (Mechanisms of learning; page unavailable) links repeated exposure and revision across narrative variations to learning, but the paper does not establish that these outcomes are particularly evident with real-time feedback or in psychologically safe learning environments. Its description of safe, controlled complexity is not a finding about psychological safety. (LLM, conf=0.85)
Experimental — accuracy not yet measuredFindings vary across outcomes, with some studies reporting stronger effects for structured cognitive tasks than for complex interpersonal or psychomotor performance.
“However, findings vary across outcomes, with some studies reporting stronger effects for structured cognitive tasks than for complex interpersonal or psychomotor performance.”
From Artificial Intelligence-Based Adaptive Simulation Integrated With a Workforce Decision-Support System to Improve Simulated Nursing Leadership Performance: A Randomized Controlled Study (2026) · cites this paper as [8]
The Results > Quantitative findings (page unavailable) report large improvements in diagnostic accuracy and NIC–NOC coherence, supporting an effect on structured reasoning tasks. The paper does not measure interpersonal or psychomotor performance or compare effect sizes across those types of outcomes, so it does not support the claimed pattern of stronger effects. (LLM, conf=0.79)
Experimental — accuracy not yet measuredStudies involving AI-supported virtual, narrative, and generative simulations have reported improvements in self-efficacy, communication knowledge, perceived competence, diagnostic reasoning, and clinical decision-making.
“Studies involving AI-supported virtual, narrative, and generative simulations have reported improvements in self-efficacy, communication knowledge, perceived competence, diagnostic reasoning, and clinical decision-making.”
From Artificial Intelligence-Based Adaptive Simulation Integrated With a Workforce Decision-Support System to Improve Simulated Nursing Leadership Performance: A Randomized Controlled Study (2026) · cites this paper as [8]
The Results > Quantitative findings (page not specified) report higher self-efficacy, diagnostic accuracy, and NIC–NOC coherence for students using generative AI–enhanced narratives. The study used written narratives, not virtual simulations, and does not report improvements in communication knowledge or a separate measure of perceived competence or clinical decision-making. (LLM, conf=0.86)
Experimental — accuracy not yet measuredConcerns regarding realism, bias, and long-term transfer to practice have been reported.
“Concerns regarding realism, bias, and long-term transfer to practice have also been reported [4,8,9].”
Recent nursing studies of adaptive ECG simulation and AI-enhanced diagnostic narratives reported that intervention participants demonstrated better decision-making, diagnostic accuracy, and self-efficacy than control participants.
“Similar findings have been reported in recent nursing studies of adaptive ECG simulation and AI-enhanced diagnostic narratives, in which intervention participants demonstrated better decision-making, diagnostic accuracy, and self-efficacy than control participants [8,11].”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 12 · Cited by 8
Research works printed but not linked: 5
- [1]Herdman TH, Kamitsuru S, editors. NANDA International nursing diagnoses: definitions and classification, 2021–2023. Stuttgart: Thieme; 2021. 10.1055/b-006-162655. · DOI
- [2]Butcher HK, Bulechek GM, Dochterman JM, Wagner CM. Nursing interventions classification (NIC). 8th ed. St. Louis (MO): Elsevier; 2021.
- [3]Moorhead S, Swanson E, Johnson M, Maas ML. Nursing outcomes classification (NOC). 7th ed. St. Louis (MO): Elsevier; 2022.
- [10]Li B, Zhao Y, Zhang H. Generative AI for clinical case simulation: opportunities and challenges. Comput Biol Med. 2023;155:106640. 10.1016/j.compbiomed.2022.106640.36774889 · DOI
- [13]McKinnon T, Smith M, Currie K. Artificial intelligence in undergraduate nursing education: A rapid review. Nurse Educ Today. 2024;122:105928. 10.1016/j.nedt.2023.105928. · DOI
- Authors
- Díaz, M. J. F.
- Affiliations
- Universidade de Santiago de Compostela (ES)
- Corresponding author
- Díaz, María José Ferreira
- Venue
- BMC Nursing
- Published
- 2026-01-31
- Volume
- 25
- Issue
- 1
- ISSN
- 1472-6955 (ISSN-L)
- PMC
- PMC12947319
- OpenAlex
- W7126470659
- Cited by
- 8
- References
- 12
- Citations by year
- 2026: 8
- FWCI
- 36.57
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- BMC Nursing (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Health Sciences › Medicine › Physiology › Simulation-Based Education in Healthcare
- Health Sciences › Medicine › Family Practice › Clinical Reasoning and Diagnostic Skills
- Health Sciences › Nursing › Issues, ethics and legal aspects › Nursing Diagnosis and Documentation
- Keywords
- generative AI, NANDA-I, NIC, NOC, standardized nursing language, narrative pedagogy, clinical reasoning, self-efficacy, nursing education
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
