Development and validation of prompts for generating simulation scenarios in nursing education using generative artificial intelligence.

• Strategic integration enables more effective use of generative AI in nursing education. • Prompts were developed to generate simulation scenarios using prompt engineering. • Based on the plan–do–check–act model, prompts underwent four refinement cycles. • Scenarios generated using prompts were val...

Descripción completa

Detalles Bibliográficos
Publicado en:Clinical Simulation in Nursing Vol. 113
Autores principales: Lee, Miji, Shin, Sujin
Formato: research Journal Article
Publicado: Elsevier B.V. Apr2026
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=192457102&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 192457102
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        18761399
        85TC
      jtl: Clinical Simulation in Nursing
      issn: 18761399
      maglogo: N
    pubinfo:
      dt: Apr2026
      vid: 113
      pid: 467
      pub: Elsevier B.V.
      place: New York, New York
    artinfo:
      ui:
        192457102
        192457102
        192457102
        10.1016/j.ecns.2026.101922
        192457102
      ppct: 1
      formats:
      tig:
        atl: Development and validation of prompts for generating simulation scenarios in nursing education using generative artificial intelligence.
      aug:
        au:
          Lee, Miji
          Shin, Sujin
        affil: College of Nursing, Ewha Womans University, Seodaemun-gu, Seoul 03760, Republic of Korea
      sug:
        subj:
          Artificial Intelligence, Generative
          Program Development
          Program Evaluation
          Engineering
          Nurse Educators
          Simulations
          Education, Nursing
          Curriculum Development
          Human
          Methodological Research
          Nursing Models, Theoretical
          Vital Signs
          Resuscitation, Cardiopulmonary
          Postoperative Care
          Content Validity
      ab: • Strategic integration enables more effective use of generative AI in nursing education. • Prompts were developed to generate simulation scenarios using prompt engineering. • Based on the plan–do–check–act model, prompts underwent four refinement cycles. • Scenarios generated using prompts were validated as appropriate through expert evaluation. • The prompts allow nurse educators to generate realistic and robust scenario drafts. Generative artificial intelligence's (GenAI) implementation in the development of simulation scenarios for nursing education is undergoing, highlighting its potential as a transformative tool. The next step is to explore how GenAI can be harnessed more effectively. This study aimed to develop and validate structured prompts that enable nurse educators to generate realistic and pedagogically robust simulation scenarios using GenAI. This methodological study employed the plan–do–check–act (PDCA) model to guide prompts' iterative refinement. Four iterative cycles were conducted during prompt development, each corresponding to an application of the PDCA model. Informed by three preliminary studies, the initial prompts were refined through three rounds of prompting, one round of expert consultation, and two rounds of content validity evaluation, resulting in four rounds of prompt refinement. In the fourth cycle, simulation scenarios were developed for the following three core nursing cases: unstable vital signs, cardiopulmonary resuscitation, and postoperative nursing care. Expert evaluation of the scenarios' educational applicability and clinical representativeness demonstrated that most items exhibited a content validity score of 0.80 or higher. Two exceptions were noted: the item regarding the inclusion of essential patient-case elements scored 0.60 across all cases, while the item concerning the reflection of real clinical situations in Case 2 scored 0.60. The k* values were 0.67 for educational applicability and 0.94 for clinical representativeness. Consequently, 48 prompts were developed. The findings indicate that structured prompts enhance simulated scenarios' educational applicability and clinical representativeness. Further research is required to evaluate these scenarios' effectiveness in educational settings. The validated prompts allow nurse educators to efficiently generate realistic and pedagogically robust simulation-scenario drafts. This study presents a comprehensive approach to developing complete simulation-scenario components using GenAI.
      pubtype: Academic Journal
      doctype:
        research
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N