Abstract: Objective To systematically evaluate the risk prediction model of post-stroke cognitive impairment （PSCI）,so as to provide reference for clinical doctors and nurses to select appropriate risk prediction model. Methods We searched PubMed,Embase,Web of Science Cochrane Library,CBM,CNKI,WanFang,and VIP for literature related to risk prediction model of PSCI from the inception of database to August 27,2022. Two researchers independently screened the literature and extracted data,PROBAST was used to assess the bias risk and applicability of the included literature,and RevMan 5.3 software was used to conduct meta-analysis on the common predictors in the included models. Results A total of 8 articles were included. The overall risk of bias of the literature was high,and its applicability was good. The area under the working characteristic curve of subjects in the 8 models was 0.807~0.930. Meta analysis showed that age[OR=1.05,95%CI（1.03,1.07）],gender[OR=2.39, 95%CI（1.58,3.61）],diabetes[OR=3.55,95%CI（1.67,7.55）],and number of non-lacunar infarcts[OR=1.59,95%CI（1.21,2.09）] were all effective predictors of PSCI. Conclusion The PSCI risk prediction model has good prediction efficiency and high overall bias risk. When reporting the risk prediction model,future researchers should refer to PROBAST's reporting specifications,and clinical medical staff should timely assess the risk of PSCI in elderly stroke patients,women,patients with diabetes and non-lacunar infarction.

Key words: Stroke, Cognition Disorders, Models, Statistical, Systematic Reviews as Topic, Evidence-Based Nursing