1예수병원 간호국
2전주비전대학교 간호학부
3전북대학교병원 간호부
4우석대학교 간호대학
5전북대학교 간호대학ㆍ간호과학연구소
1Department of Nursing, Presbyterian Medical Center, Jeonju, Korea
2Department of Nursing Science, VISION College of Jeonju, Jeonju, Korea
3Department of Nursing, Jeonbuk National University Hospital, Jeonju, Korea
4College of Nursing, Woosuk University, Wanju, Korea
5College of Nursing, Research Institute of Nursing Science, Jeonbuk National University, Jeonju, Korea
© 2025 Korean Society of Nursing Science
This is an Open Access article distributed under the terms of the Creative Commons Attribution NoDerivs License (http://creativecommons.org/licenses/by-nd/4.0) If the original work is properly cited and retained without any modification or reproduction, it can be used and re-distributed in any format and medium.
Conflicts of Interest
Seok Hee Jeong has been the editorial board member of JKAN since 2024. but has no role in the review process. Except for that, no potential conflict of interest relevant to this article was reported.
Acknowledgements
None.
Data Sharing Statement
Please contact the corresponding author for data availability.
Author Contributions
Conceptualization or/and Methodology: all authors. Data curation or/and Analysis: all authors. Investigation: all authors. Project administration or/and Supervision: SHJ. Validation: all authors. Visualization: all authors. Writing: original draft or/and Review & Editing: all authors. Final approval of the manuscript: all authors.
Characteristic | No. (%) |
---|---|
Publication year | |
2022 | 1 (1.7) |
2023 | 42 (72.4) |
2024 | 15 (25.9) |
The first author’s affiliated country | |
North America | 27 (46.6) |
USA | 24 (41.4) |
Canada | 3 (5.2) |
Europe | 10 (17.2) |
UK | 4 (6.9) |
Belgium | 2 (3.4) |
Others (Italy, Malta, Netherlands, Sweden) | 4 (6.9) |
Asia | 20 (34.5) |
South Korea | 3 (5.2) |
Japan | 3 (5.2) |
China | 2 (3.4) |
Hong Kong | 2 (3.4) |
Taiwan | 2 (3.4) |
Turkey | 2 (3.4) |
Others (India, Indonesia, Israel, Qatar, Singapore, Cambodia) | 6 (10.5) |
Oceania | 1 (1.7) |
Australia | 1 (1.7) |
Publication type | |
Research papers | 14 (24.1) |
Review papers | 2 (3.4) |
Discussion | 12 (20.7) |
Editorial | 23 (39.8) |
Letter to editor | 2 (3.4) |
Others (teaching strategies, teaching tip) | 5 (8.6) |
Generative AI typea) | |
LLM | 9 (14.0) |
ChatGPT | 49 (75.4) |
Google Bard | 2 (3.1) |
Bing Chat | 1 (1.5) |
Claude-2 | 1 (1.5) |
Llama-2 | 1 (1.5) |
Perplexity | 1 (1.5) |
Scispace | 1 (1.5) |
AI, artificial intelligence; ChatGPT, chat generative pre-trained transformer; DISCERN, a tool developed by the National Health Service Research and Development Program and the British Library to assess the quality and reliability of online health information, particularly treatment options; DNP, doctor of nursing practice; NANDA, north american nursing diagnosis association; NCLEX-RN, national council licensing examination for registered nurses; PICOT, population, intervention, comparison, outcome, and time.
Characteristic | No. (%) |
---|---|
Publication year | |
2022 | 1 (1.7) |
2023 | 42 (72.4) |
2024 | 15 (25.9) |
The first author’s affiliated country | |
North America | 27 (46.6) |
USA | 24 (41.4) |
Canada | 3 (5.2) |
Europe | 10 (17.2) |
UK | 4 (6.9) |
Belgium | 2 (3.4) |
Others (Italy, Malta, Netherlands, Sweden) | 4 (6.9) |
Asia | 20 (34.5) |
South Korea | 3 (5.2) |
Japan | 3 (5.2) |
China | 2 (3.4) |
Hong Kong | 2 (3.4) |
Taiwan | 2 (3.4) |
Turkey | 2 (3.4) |
Others (India, Indonesia, Israel, Qatar, Singapore, Cambodia) | 6 (10.5) |
Oceania | 1 (1.7) |
Australia | 1 (1.7) |
Publication type | |
Research papers | 14 (24.1) |
Review papers | 2 (3.4) |
Discussion | 12 (20.7) |
Editorial | 23 (39.8) |
Letter to editor | 2 (3.4) |
Others (teaching strategies, teaching tip) | 5 (8.6) |
Generative AI type |
|
LLM | 9 (14.0) |
ChatGPT | 49 (75.4) |
Google Bard | 2 (3.1) |
Bing Chat | 1 (1.5) |
Claude-2 | 1 (1.5) |
Llama-2 | 1 (1.5) |
Perplexity | 1 (1.5) |
Scispace | 1 (1.5) |
ID/author (year) | Population | Context | Concept | Types of generative AI | Conclusion |
---|---|---|---|---|---|
A1. Ahn (2023) | Nursing professors | Nursing education | Development of a case-based nursing education program utilizing generative AI | ChatGPT | Six modules suitable for case-based learning were designed. Generative AI can enhance case-based learning. It can be integrated into nursing education. |
A6. Branum (2023) | ChatGPT | Nursing research | Evaluation of PICOT clinical questions using ChatGPT | ChatGPT | ChatGPT provided unreliable answers and fabricated references. It is an unreliable tool for clinical questions. |
A10. Chang (2024) | Nursing undergraduates | Nursing education | Quasi-experimental study in a nursing education design course integrating ChatGPT | ChatGPT | Students’ performance, critical thinking, and satisfaction improved. Generative AI has the potential to enhance nursing education. However, further research is needed. |
A12. Cox (2023) | NCLEX-RN educators | Nursing education | Comparative analysis of NCLEX-RN questions generated by AI and human educators | ChatGPT | Questions generated by generative AI demonstrated a comparable level of clarity. However, they required educator revisions. Although generative AI can assist in creating NCLEX questions, human review is essential. |
A13. Dağci (2024) | Nursing texts generated by ChatGPT | Nursing practice | Analysis of 40 care plans based on NANDA diagnoses using the DISCERN tool | ChatGPT | Nursing care plan texts generated by generative AI demonstrated a moderate level of reliability and required improvement. |
A16. Epstein (2024) | Hospital librarians, nurses | Nursing research | Evidence-based practice education using AI tools and practical experience | ChatGPT, Claude-2, Llama-2, Perplexity | AI tools are useful in evidence-based practice education, particularly in evaluating evidence and retrieving information. ChatGPT and other AI tools can save educational time and improve the speed of information delivery. |
A23. Huang (2023) | ChatGPT and nursing license exam questions | Nursing education | ChatGPT's performance in Taiwan nursing license exam | ChatGPT | ChatGPT scored an average of 51.6 to 63.75 points out of 100 points in the nursing licensure exam, with some incorrect answers generated. While ChatGPT can be used as an auxiliary tool in nursing education, there is a risk of incorrect responses. |
A25. Taira (2023) | ChatGPT and national nursing exam questions | Nursing education | Evaluation of ChatGPT’s performance in Japanese national nursing licensure exam | ChatGPT | ChatGPT demonstrated a good understanding of basic knowledge. However, it failed in more complex areas. While it has potential, improvements are needed to enhance its consistency. |
A32. Moons (2024) | Patient information texts | Nursing practice | Improving readability of patient information using ChatGPT: a proof of concept | ChatGPT, Google Bard | Although ChatGPT improved the readability of patient information, it did not reach the recommended standards. Generative AI can enhance readability. However, content accuracy is not consistently maintained. |
A39. Parker (2023) | Nursing writing texts for students | Nursing education | Exploring ChatGPT for assessing academic writing in nursing education | ChatGPT | ChatGPT provides more rigorous grading than human evaluators and provides detailed feedback on writing. ChatGPT has potential as an automated writing assessment tool. It can improve the speed and quality of feedback without increasing instructor workload. It can support students’ self-directed learning. |
A40. Quattrini (2024) | Registered nurses, doctors of nursing practice | Nursing education | Analyzing the effectiveness of nursing DNP educational activities using clinical data generated by ChatGPT: Supporting clinical decision-making | ChatGPT | Through clinical scenarios generated by ChatGPT, students can develop critical thinking and strengthen clinical judgment skills through assessments. ChatGPT can facilitate higher-order thinking. It has been proven to be useful in clinical education. However, its application within educational programs requires careful consideration. |
A41. Saban (2024) | Emergency room registered nurses, nursing students | Nursing practice | Evaluating ChatGPT's contribution to triage and clinical decision-making in emergency care | ChatGPT | ChatGPT can support clinical decision-making. However, it requires human oversight. |
A54. Zaboli (2024) | Emergency room clinical scenarios | Nursing practice | Evaluating whether human intelligence or AI achieves better performance in patient triage | ChatGPT | Human nurses outperformed ChatGPT in predicting 72-hour mortality rates. ChatGPT cannot replace human expertise yet in prioritizing emergency room patients. |
A55. Kim (2023) | The responses of ChatGPT-3.5 and ChatGPT-4. | Nursing education | Evaluating the potential use of ChatGPT in biological nursing science education | ChatGPT | GPT-4 demonstrated higher accuracy than GPT-3.5 in responding to questions in Korean. Both models achieved 100% accuracy for questions in English. ChatGPT can be useful for understanding complex concepts in biological nursing science. However, it requires integration of up-to-date data. |
Domains | Contents | ID |
---|---|---|
Nursing education (n=26) | ||
Strengths | Save time in educational design and implementation | A4, A8, A28, A36, A44, A46, A51, A56 |
Support educational content creation | A3, A8, A28, A36, A43, A49 | |
Customized individual learning and feedback | A3, A11, A24, A28, A46, A58 | |
Useful for learning diagnosis and self-directed learning support | A3, A11, A24, A36 | |
Exceptional accessibility to information | A3, A11, A43, A46, A56 | |
Provision of guidance for problem-solving steps | A9, A35 | |
Updating and maintenance of the latest resources | A3, A11 | |
Accessibility to professional resources | A8 | |
Useful for disease information education | A22 | |
Providing systematic learning management and interactive experiences | A35, A48 | |
Overcoming traditional educational barriers | A3, A58 | |
Providing consistent education | A3 | |
Support for writing | A46 | |
Weaknesses and limitations | Algorithmic bias and generation and spread of misinformation | A3, A4, A5, A11, A24, A28, A35, A36, A37, A51, A56 |
Potential for ethical issues | A2, A3, A11, A17, A27, A28, A35, A51 | |
Information security issues | A2, A3, A9, A17, A27, A36 | |
Fake references and inaccurate sources | A8, A9, A24, A36, A46, A49 | |
Lack of human interaction | A3, A5, A9, A11, A28, A36 | |
Decline in critical thinking and problem-solving skills | A3, A4, A28, A46, A49 | |
Potential for bias | A28, A49, A51 | |
Need for fact-checking and lack of accountability | A21, A35, A58 | |
Excessive simplification | A2, A14 | |
Emergence of copyright issues | A3, A51 | |
Limitations in capturing human emotions | A2, A14 | |
Limitations in intuitive decision-making | A2, A44 | |
Generation of irrelevant responses | A8, A46 | |
Lack of access to untrained latest information | A3, A8 | |
Inability to verify the truthfulness of statements | A8 | |
Threat to scientific writing | A17 | |
Not a substitute for on-site training | A8 | |
Inability to generate new ideas | A8 | |
Widening gap between industry and education and intensification of social inequalities | A24 | |
Consideration of accessibility-based equity issues and individualized decision-making requirement | A2, A3, A11, A28, A36, A43 | |
Recommendations and guidelines | Developing policies and guidelines for AI utilization | A2, A4, A8, A11, A17, A28, A36, A44, A46, A49, A51 |
User education and support | A5, A8, A11, A35 A36, A37, A44, A46, A51 | |
Promotion of critical thinking | A2, A8, A14, A21, A44, A46, A51, A56 | |
Utilization as a complementary tool to traditional teaching methods | A11, A44, A49 | |
Supplementation and updates of additional systems | A5, A24, A28, A35 | |
Integration into educational curriculum | A8, A24 | |
Research recommendations from nursing educators | A8, A44 | |
Use of reference management software is recommended | A57 | |
Provision of assignments encouraging critical and creative thinking | A11, A35, A46 | |
Need for cross-verification and content evaluation | A11, A17, A21 | |
Understanding current trends in policy and legislative actions | A17, A21 | |
Avoid entering sensitive information | A37 | |
Need for protection from harmful information | A37 | |
Reference to OpenAI guidelines and author suggestions | A8, A21 | |
Nursing research (n=15) | ||
Strengths | Enhanced efficiency in the paper writing process | A15, A17, A52 |
Promoting equity in access to information | A29 | |
Improved speed in paper dissemination | A29 | |
Weaknesses and limitations | Risk of information manipulation and plagiarism | A47, A50, A52 |
Insufficient reliability of output results | A47, A52 | |
Erosion of academic foundations | A2 | |
Lack of ethical and legal accountability | A50 | |
Inappropriate as an author | A21, A47 | |
Information oversimplification and readability issues | A18 | |
Production of inaccurate and biased information | A18 | |
Recommendations and guidelines | Restriction on authorship eligibility | A4, A5, A18, A21, A30, A45, A50 |
Need for guidelines and policies on research utilization | A29, A50, A52 | |
Necessity for critical attitudes and transparency in usage | A18, A45 | |
Disclosure of ChatGPT usage and language development | A8 | |
Ensuring usability and reliability of generative AI | A18 | |
Review and update codes of conduct for research utilization | A27 | |
Nursing practice (n=15) | ||
Strengths | Potential for increased efficiency in broad nursing practice areas | A20, A31, A42 |
Improving nurse-patient communication | A21, A22, A42, A53 | |
Providing the latest technology and information | A20, A38 | |
Increasing convenience for routine tasks | A19, A42 | |
Providing high-quality healthcare services | A31, A53 | |
Identification and prediction of nursing situations | A19, A20 | |
Sustainable conversations | A7 | |
Speed in information retrieval and responses | A38 | |
Improvement in patient self-management abilities | A22 | |
Weaknesses and limitations | Low reliability of output results containing errors | A7, A13, A21, A27, A31, A54 |
Inappropriateness for providing quality nursing care | A8, A20 | |
Neglecting ethical considerations | A33, A34 | |
Lack of governance | A53 | |
Decreased nursing proficiency due to overreliance | A42 | |
Irreplaceable human element in nursing | A38 | |
High demand for information accuracy in the medical field | A21, A34 | |
Recommendations and guidelines | Creative utilization based on critical thinking | A2, A21 |
Responsible use with ethics in mind | A21, A38, A42 | |
Exercise caution when applying to nursing situations | A20, A21 | |
Need for regulations in utilization | A2 | |
Critical approach necessity, up-to-date information usage maintenance, and supplementary tool utilization for clinical decision-making enhancement | A26, A34 | |
Nursing administration (n=0) | ||
Strengths | NA | |
Weaknesses and limitations | NA | |
Recommendations and guidelines | NA |
No. | Databases | Search query | Results | ||
---|---|---|---|---|---|
1 | KoreaMed | (“nurses OR nurse”[ALL] AND “ChatGPT OR Conversation A.I. OR Generative A.I.”[ALL]) | 0 | ||
2 | KMbase (Korean Medical Database) | (간호사|간호|total) AND (ChatGPT|챗GPT|인공지능챗봇|대화형 인공지능|생성형 인공지능|total) | 0 | ||
3 | KISS (Koreanstudies Information Service System) | 전체 = “간호사|간호” and 전체 = “ChatGPT|챗GPT|인공지능챗봇|대화형 인공지능|생성형 인공지능” | 0 | ||
4 | Science ON | “전체=간호사|간호 AND 전체=ChatGPT|챗GPT|인공지능챗봇|대화형 인공지능|생성형 인공지능” | 29 | ||
5 | RISS (Research Information Sharing Service) | 전체 : 간호사|간호 <AND> 전체 : ChatGPT|챗GPT|인공지능챗봇|대화형 인공지능|생성형 인공지능 | 2 | ||
6 | DBpia (DataBase Periodical Information Academic) | 전체 : “간호사”|“간호” <AND> 전체 : “ChatGPT”|“챗GPT”|“인공지능챗봇”|“대화형 인공지능”|“생성형 인공지능” | 2 | ||
7 | CINAHL | TX ( “nurses” OR “nurse” ) AND TX ( “ChatGPT” OR “Conversation A.I.” OR “Generative A.I.” ) | 251 | ||
8 | PubMed | No. | Query | Results | 57 |
#1 | “nurses” | 262,793 | |||
#2 | “nurse” | 290,959 | |||
#3 | “ChatGPT” | 2,577 | |||
#4 | “Conversation A.I.” | 21 | |||
#5 | “Generative A.I.” | 3,694 | |||
#6 | (“nurses”) OR (“nurse”) | 461,770 | |||
#7 | ((“ChatGPT”) OR (“Conversation A.I.”)) OR (“Generative A.I.”) | 6,287 | |||
#8 | ((“nurses”) OR (“nurse”)) AND (((“ChatGPT”) OR (“Conversation A.I.”)) OR (“Generative A.I.”)) | 57 | |||
9 | Embase | No. | Query | Results | 46 |
#1 | ‘nurses’ | 341,987 | |||
#2 | ‘nurse’ | 444,302 | |||
#3 | ‘chatgpt’ | 2,647 | |||
#4 | ‘conversation a.i.’ | 0 | |||
#5 | ‘generative a.i.’ | 1 | |||
#6 | #2 OR #3 | 601,562 | |||
#7 | #4 OR #5 OR #6 | 2,648 | |||
#8 | #7 AND #8 | 46 | |||
10 | CENTRAL | No. | Query | Results | 0 |
#1 | “nurses” | 19,754 | |||
#2 | “nurse” | 26,106 | |||
#3 | “ChatGPT” | 23 | |||
#4 | “Conversation A.I.” | 0 | |||
#5 | “Generative A.I.” | 0 | |||
#6 | #1 OR #2 | 36,975 | |||
#7 | #3 OR #4 OR #5 | 23 | |||
#8 | #6 AND #7 | 0 |
AI, artificial intelligence; LLM, large language model; UK, united kingdom; USA, united states of america. Total exceeds N due to multiple AI types per study.
AI, artificial intelligence; ChatGPT, chat generative pre-trained transformer; DISCERN, a tool developed by the National Health Service Research and Development Program and the British Library to assess the quality and reliability of online health information, particularly treatment options; DNP, doctor of nursing practice; NANDA, north american nursing diagnosis association; NCLEX-RN, national council licensing examination for registered nurses; PICOT, population, intervention, comparison, outcome, and time.
AI, artificial intelligence; ChatGPT, chat generative pre-trained transformer; NA, not applicable. A total of 44 unique studies were included; however, when accounting for overlapping categories, the total count increased to 56.