Validation and quality of AI-generated continuous quality improvement reports in nursing education

IF 2.5 Q2 NURSING

Teaching and Learning in Nursing Pub Date : 2026-10-01 Epub Date: 2026-06-11 DOI:10.1016/j.teln.2026.05.018

Sun Kyung Kim PhD, RN, Hye Ri Hwang MSN, RN, Jongmyung Choi PhD

{"title":"Validation and quality of AI-generated continuous quality improvement reports in nursing education","authors":"Sun Kyung Kim PhD, RN, Hye Ri Hwang MSN, RN, Jongmyung Choi PhD","doi":"10.1016/j.teln.2026.05.018","DOIUrl":null,"url":null,"abstract":"<div><h3>Aim</h3><div>This study evaluated the content and information quality of CQI reports generated by a large language model (LLM).</div></div><div><h3>Background</h3><div>CQI in nursing education relies on extensive data analysis. LLMs have emerged as effective tools for producing quality documentation.</div></div><div><h3>Design</h3><div>Mixed method design.</div></div><div><h3>Methods</h3><div>We developed a prompt in which refined in consultation with multidisciplinary experts. CQI reports were generated by GPT-o3, covering domains of statistical analysis, text analysis of learning diaries, root-cause identification and improvement planning.</div></div><div><h3>Results</h3><div>A total of 20 experts participated, 18 nursing professors and two professors specializing in pedagogy. All domains achieved a Content Validity Index of 0.95 or higher. Scores on information quality exceeded 4.0 across all components, demonstrating a high level of perceived accuracy, relevance, clarity, and usefulness. Analysis of the qualitative feedback identified three themes: acceptable report quality, demands for tailored reporting and in-depth analysis, and perceived reliability yet logical gaps.</div></div><div><h3>Conclusions</h3><div>Findings indicated that AI-generated CQI reports are valid and adequate. However, enhancing logical fluency and causal reasoning is still essential to satisfy professional standards.</div></div>","PeriodicalId":46287,"journal":{"name":"Teaching and Learning in Nursing","volume":"21 4","pages":"Pages e1559-e1565"},"PeriodicalIF":2.5000,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Teaching and Learning in Nursing","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1557308726001472","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/6/11 0:00:00","PubModel":"Epub","JCR":"Q2","JCRName":"NURSING","Score":null,"Total":0}

引用次数: 0

Abstract

Aim

This study evaluated the content and information quality of CQI reports generated by a large language model (LLM).

Background

CQI in nursing education relies on extensive data analysis. LLMs have emerged as effective tools for producing quality documentation.

Design

Mixed method design.

Methods

We developed a prompt in which refined in consultation with multidisciplinary experts. CQI reports were generated by GPT-o3, covering domains of statistical analysis, text analysis of learning diaries, root-cause identification and improvement planning.

Results

A total of 20 experts participated, 18 nursing professors and two professors specializing in pedagogy. All domains achieved a Content Validity Index of 0.95 or higher. Scores on information quality exceeded 4.0 across all components, demonstrating a high level of perceived accuracy, relevance, clarity, and usefulness. Analysis of the qualitative feedback identified three themes: acceptable report quality, demands for tailored reporting and in-depth analysis, and perceived reliability yet logical gaps.

Conclusions

Findings indicated that AI-generated CQI reports are valid and adequate. However, enhancing logical fluency and causal reasoning is still essential to satisfy professional standards.

查看原文本刊更多论文

人工智能生成的护理教育持续质量改进报告的有效性和质量

目的评价大型语言模型（LLM）生成的CQI报告的内容和信息质量。护理教育中的cqi依赖于大量的数据分析。法学硕士已经成为生产高质量文档的有效工具。设计混合方法设计。方法在咨询多学科专家的基础上，制定了一套完善的提示。CQI报告由gpt - 03生成，涵盖统计分析、学习日记文本分析、根本原因识别和改进计划等领域。结果共有专家20人参与，其中护理学教授18人，教育学专业教授2人。所有域的内容有效性指数均达到0.95或更高。信息质量的得分在所有组件中都超过4.0，显示出高水平的感知准确性、相关性、清晰度和有用性。对定性反馈的分析确定了三个主题：可接受的报告质量，对量身定制的报告和深入分析的需求，以及可感知的可靠性和逻辑差距。结论人工智能生成的CQI报告是有效和充分的。然而，提高逻辑流畅性和因果推理能力仍然是满足专业标准的必要条件。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Teaching and Learning in Nursing NURSING-

CiteScore

2.60

自引率

10.50%

发文量

119

审稿时长

37 days

期刊介绍： Teaching and Learning in Nursing is the Official Journal of the National Organization of Associate Degree Nursing. The journal is dedicated to the advancement of Associate Degree Nursing education and practice, and promotes collaboration in charting the future of health care education and delivery. Topics include: - Managing Different Learning Styles - New Faculty Mentoring - Legal Issues - Research - Legislative Issues - Instructional Design Strategies - Leadership, Management Roles - Unique Funding for Programs and Faculty