Adding AI intervention to perioperative ERAS nursing in general surgery improved patients' disease knowledge, 24-hour pain, first ambulation time, and satisfaction versus routine care

Embedding AI intervention into the perioperative ERAS nursing pathway in general surgery was associated with simultaneous improvement in disease knowledge, postoperative pain, time to first ambulation, and nursing satisfaction. Prior references focus on single-link AI uses such as AI education robots, AI-assisted chronic disease management, or AI combined with a specific teaching model; this study organizes AI intervention into a full-process nursing system of preoperative intelligent assessment, postoperative intelligent warning, and continuous intelligent follow-up, and compares it against routine ERAS care. Single-center controlled study with random number table allocation, 84 elective surgery patients, and all four outcomes reported as mean±SD with t values and P<0.05.

The AI intervention showed larger between-group differences in disease knowledge and nursing satisfaction, and moderate differences in pain and time to first ambulation. The study extends the reported effects of AI intervention from a single satisfaction or education outcome to four perioperative outcomes, providing controlled data on the same intervention across multiple endpoints. Between-group differences were about 24 points for disease knowledge (t=33.191), about 9 points for satisfaction (t=19.338), about 2 points for 24-hour pain (t=8.105), and about 1.4 days for first ambulation (t=9.203), all reported with P<0.05.

The study offers an organizational pattern for AI nursing that can be implemented in general surgery, mapping intelligent assessment, intelligent warning, and intelligent follow-up onto the preoperative, postoperative, and post-discharge phases. Rather than treating AI as a single-point tool, this three-phase division aligns AI intervention with the ERAS timeline so nursing teams can assign tasks by phase. The methods state that the observation group built a full-process nursing system of preoperative intelligent assessment, postoperative intelligent warning, and continuous intelligent follow-up, and the results report corresponding changes in perioperative recovery indicators.

The study addresses elective surgery patients admitted to the general surgery department of a hospital in Wuchuan, Guangdong, and applies to perioperative nursing settings where AI intervention is added on top of routine ERAS care; for nursing teams, its value lies in offering a three-phase division of labor across preoperative intelligent assessment, postoperative intelligent warning, and continuous intelligent follow-up that can be referenced when organizing workflows in comparable general surgery wards.

The loaded text is incomplete, containing only the abstract, keywords, and references, without method details, instrument sources, randomization and blinding procedures, sample size justification, or adverse events; therefore the specific technical composition of each AI module, the assessment instruments, and follow-up frequency remain open questions. The text also contains an internal inconsistency in the reported group sizes, so the actual number of patients in the observation and control groups should be confirmed against the original. As a single-center study, the stability of its effect sizes under larger samples and multicenter conditions remains to be observed.