Objective

2

This study evaluated the effectiveness of an AI-assisted information traceability system in risk early warning for nursing quality and operational efficiency assessment in the Central Sterile Supply Department (CSSD)

providing evidence-based support for the intelligent and refined management of the CSSD.

Methods

2

A self-controlled study was conducted in the CSSD of a Grade A tertiary hospital. The control phase (January to June 2025) employed the conventional barcode-based traceability system

whereas the observation phase (July to December 2025) utilized the AI-assisted information traceability system. In the control group

the management model comprised process recording

manual spot-check quality control

and manual statistical efficiency management. For the observation group

AI modules were developed based on the existing traceability system

incorporating four core functional modules: full-process data collection

three-tier risk early warning

dynamic efficiency evaluation

and closed-loop rectification. The two groups were compared across nursing quality and risk control indicators

operational efficiency indicators

and nursing staff work experience indicators. All statistical analyses were performed using SPSS 26.0 software.

Results

2

In the observation group

the qualified rates for instrument cleaning

packaging

sterilization

and risk early warning accuracy were 99.94%

99.97%

99.99%

and 98.62%

respectively

outperforming the control group (99.71%

99.78%

99.83%

and 82.15%). The incidence rates of quality defects

adverse events

and healthcare-associated infection (HAI) related events in the observation group were 0.07%

0.03%

and 0

respectively

which were lower than controls (0.32%

0.18%

and 0.11%) (all

P

<

0.05). The average instrument turnaround time

emergency item processing time

and daily manual verification time in the observation group were 16.58 ± 2.37  h

1.21 ± 0.32  h

and 0.85 ± 0.21  h

respectively

all significantly shorter than the corresponding values in the control group (22.43 ± 3.26  h

2.64 ± 0.57  h

and 4.12 ± 0.75  h). The instrument reuse rate

equipment utilization rate

and completeness rate of traceability information were also significantly higher in the observation group than in the control group (all

P

<

0.05). Furthermore

nursing staff in the observation group reported significantly higher satisfaction with work efficiency and system use

along with a significantly lower risk management pressure score

compared with the control group (all

P

<

0.05).

Conclusion

2

The AI-assisted information traceability system enables pre-event alerting

real-time interception

and post-event traceability for risks throughout the entire workflow in the CSSD. The AI-assisted system significantly enhanced nursing quality and safety

and reduced HAI risk. Furthermore

it facilitates dynamic and refined evaluation and optimization of departmental operational efficiency

alleviates the workload of nursing staff

and meets the requirements of smart hospital construction and high-quality nursing development. Collectively

these findings support the system’s substantial potential for clinical translation and broad applicability.