Nursing Documentation in the AI Era: A Comparative Systematic Review and Meta-Analysis of Efficiency, Mistakes, Stress, and Quality of Care

This comparative systematic review and meta-analysis, led by Dr. Fernan N. Torreno and Famiela Torreno, evaluates the impact of artificial intelligence (AI)-assisted nursing documentation versus traditional charting methods. Nursing documentation is essential for patient safety and care continuity but remains a major source of workload, stress, and potential error for nurses worldwide. By synthesizing global evidence, this study explores whether AI can enhance efficiency, accuracy, and care quality while reducing stress.

Following PRISMA 2020 and ENTREQ guidelines, the authors systematically reviewed and meta-analyzed 32 studies (n≈6,200 nurses) published between 2010 and 2025. Studies included randomized controlled trials, observational cohorts, mixed-methods, and qualitative research. Risk of bias was appraised using RoB 2, ROBINS-I, and CASP tools, while certainty of evidence was assessed through GRADE and CERQual frameworks.

Quantitative synthesis demonstrated that AI-assisted documentation reduced documentation time by an average of −32 minutes per shift (95% CI −40 to −24), improved accuracy and completeness (RR 1.21; 95% CI 1.10–1.34), and lowered stress scores (SMD −0.38; 95% CI −0.55 to −0.21). Errors of omission decreased, though new challenges such as transcription or autocorrect errors were noted. Qualitative themes emphasized relief from administrative burden, yet also raised concerns regarding trust, usability, and professional deskilling.

Evidence suggested improvements in quality of care, with more bedside time and increased patient coverage per shift. However, variations across settings and infrastructural limitations in low- and middle-income countries tempered certainty.

To translate findings into practice, the authors developed a SMART Policy Roadmap, aligning with the WHO Global Digital Health Strategy and ICN’s call for AI literacy. Key milestones include AI literacy integration by 2027, verification safeguards by 2028, stress audits in implementation by 2029, and equitable adoption in LMICs by 2030.

In conclusion, this study led by Dr. Fernan and Famiela Torreno provides timely and policy-relevant evidence that AI-assisted documentation enhances efficiency, accuracy, and workforce well-being. At the same time, it calls for ethical safeguards and equity-focused strategies to ensure AI supports, rather than undermines, nursing practice worldwide.