Volume 15, Issue 6 And S7 (Artificial Intelligence- In Press 2025)
J Research Health 2025, 15(6 And S7): 7-7 |
Back to browse issues page
Ethics code: CRD420251089257
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Afkhami Teimouri G, Amadeh Taheri A, Moeinipour Y, Adineh Fathabadi M, Bagheri F. The Impact of AI-Based Nursing Documentation on Time Management and Patient Safety: A Systematic Review. J Research Health 2025; 15 (6) :7-7
URL: http://jrh.gmu.ac.ir/article-1-2940-en.html
URL: http://jrh.gmu.ac.ir/article-1-2940-en.html
Ghazaleh Afkhami Teimouri1 
, Arash Amadeh Taheri1 , Yasamin Moeinipour2 , Malihe Adineh Fathabadi1 , Fatemeh Bagheri3
, Arash Amadeh Taheri1 , Yasamin Moeinipour2 , Malihe Adineh Fathabadi1 , Fatemeh Bagheri3
1- Department of nursing and midwifery, MMS.C. Islamic Azad University, Mashhad, Iran
2- Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Islamic Republic of Iran
3- Department of nursing and midwifery, MMS.C. Islamic Azad University, Mashhad, Iran ,bagherifatemeh59@yahoo.com
2- Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Islamic Republic of Iran
3- Department of nursing and midwifery, MMS.C. Islamic Azad University, Mashhad, Iran ,
Abstract: (38 Views)
Background: The administrative burden of nursing documentation is a primary contributor to clinician burnout. Artificial Intelligence (AI), particularly emerging generative and ambient technologies, offers a potential solution, yet the evidence regarding its dual impact on efficiency and safety remains fragmented. This systematic review aims to synthesize the evidence on the effects of AI-driven interventions in nursing documentation on time management and patient safety outcomes.
Methods: A systematic review was conducted following PRISMA 2020 guidelines and registered with PROSPERO (CRD420251089257). A comprehensive search was performed in PubMed, Scopus, Web of Science, and Embase with no date or language restrictions. Primary research studies evaluating any AI intervention in nursing documentation for its effect on time or safety were included. Due to significant heterogeneity, a narrative synthesis was performed.
Results: From an initial 2,052 records, 18 studies met the inclusion criteria. The included studies were methodologically diverse, comprising randomized trials, quasi-experimental, and qualitative designs, with most assessed at a moderate risk of bias. The findings indicate that generative and ambient AI tools can significantly reduce documentation time and improve efficiency. However, the impact on patient safety was shown; some AI tools directly prevented adverse events (e.g., medication errors) or identified safety risks more effectively, while others improved safety indirectly by enhancing documentation quality. Critically, several studies highlighted the emergence of new risks, such as AI-generated inaccuracies ("hallucinations") and a lack of clinical nuance, underscoring the necessity of human oversight.
Conclusion: AI-driven documentation systems significantly enhance clinical efficiency by reducing documentation time and cognitive workload, thereby improving workflow and allowing greater focus on patient care. However, their reliability for autonomous use remains limited, underscoring the need for human oversight to maintain clinical accuracy and safety. Persistent challenges, including data heterogeneity, interoperability gaps, and ethical concerns, must be addressed through standardized frameworks, advanced Natural Language Processing (NLP) development, and transparent validation. Future large-scale, multi-center studies should evaluate the sustained effects of AI-assisted documentation on efficiency, clinician well-being, and patient outcomes to enable safe, trustworthy, and equitable integration into clinical practice.
Methods: A systematic review was conducted following PRISMA 2020 guidelines and registered with PROSPERO (CRD420251089257). A comprehensive search was performed in PubMed, Scopus, Web of Science, and Embase with no date or language restrictions. Primary research studies evaluating any AI intervention in nursing documentation for its effect on time or safety were included. Due to significant heterogeneity, a narrative synthesis was performed.
Results: From an initial 2,052 records, 18 studies met the inclusion criteria. The included studies were methodologically diverse, comprising randomized trials, quasi-experimental, and qualitative designs, with most assessed at a moderate risk of bias. The findings indicate that generative and ambient AI tools can significantly reduce documentation time and improve efficiency. However, the impact on patient safety was shown; some AI tools directly prevented adverse events (e.g., medication errors) or identified safety risks more effectively, while others improved safety indirectly by enhancing documentation quality. Critically, several studies highlighted the emergence of new risks, such as AI-generated inaccuracies ("hallucinations") and a lack of clinical nuance, underscoring the necessity of human oversight.
Conclusion: AI-driven documentation systems significantly enhance clinical efficiency by reducing documentation time and cognitive workload, thereby improving workflow and allowing greater focus on patient care. However, their reliability for autonomous use remains limited, underscoring the need for human oversight to maintain clinical accuracy and safety. Persistent challenges, including data heterogeneity, interoperability gaps, and ethical concerns, must be addressed through standardized frameworks, advanced Natural Language Processing (NLP) development, and transparent validation. Future large-scale, multi-center studies should evaluate the sustained effects of AI-assisted documentation on efficiency, clinician well-being, and patient outcomes to enable safe, trustworthy, and equitable integration into clinical practice.
Type of Study: Review Article |
Subject:
● Artificial Intelligence
Received: 2025/10/13 | Accepted: 2025/11/8 | Published: 2025/11/8
Received: 2025/10/13 | Accepted: 2025/11/8 | Published: 2025/11/8
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |