Abstract:

Objective To systematically retrieve and evaluate the subsyndromal delirium （SSD） risk prediction models for ICU patients，aiming to provide reference for medical staff to construct and select scientifically effective risk prediction models. Methods The researches on SSD risk prediction models for ICU patients were retrieved in PubMed，Embase，Web of Science，Cochrane Library，CINAHL，China National Knowledge Infrastructure，VIP database，Wanfang database，and China Biomedical Literature Database，the retrieval time limit was from the establishement of database until July 19，2024. Two researchers independently screened literature，extracted data，and evaluated the bias risk and applicability of the included literature. Results A total of 8 studies on SSD risk prediction model for ICU patients were included，involving 8 models，with an area under the curve（AUC） of 0.710-0.956 for establishing the models. The overall applicability was good，but the risk of bias was high，mainly concentrated in the field of data analysis. Conclusion The existing ICU patient SSD risk prediction models have good overall performance，but have a high risk of bias. In the future，the existing models can be further optimized or higher quality prediction models can be developed.

Key words: Intensive Care Units, Subsyndromal Delirium, Forecasting Model, Evidence-Based Nursing, Systematic Review