Systematic review of risk prediction models for hypoglycemia in diabetic patients
10.3760/cma.j.cn115682-20220612-02817
- VernacularTitle:糖尿病患者低血糖风险预测模型的系统评价
- Author:
Mengya YAN
1
;
Meijuan WANG
;
Yihong XU
;
Xiaolin LIU
;
Dan YANG
;
Yang GAO
;
Shanni DING
;
Hongying PAN
Author Information
1. 浙江中医药大学护理学院，杭州 310053
- Keywords:
Hypoglycemia;
Diabetes;
Prediction model;
Systematic review
- From:
Chinese Journal of Modern Nursing
2023;29(10):1325-1332
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To systematically review the risk prediction models for hypoglycemia in diabetic patients.Methods:The literatures published up to March 25, 2022 was retrieved from PubMed, Embase, Web of Science, Cochrane Library, CINAHL, MEDLINE, China National Knowledge Infrastructure, Wanfang, VIP and SinoMed. Two researchers independently screened the literatures, extracted information, and applied the PROBAST tool to evaluate the quality of the included models.Results:A total of 11 literatures and 13 models were included. The area under the receiver operating characteristic curve or C statistic of all models was 0.666-0.890, with a high risk of bias and a low risk of applicability, and the most included predictors were chronic kidney disease and age. The main reason for the bias in the model were insufficient number of events in the dependent variable, improper handling of continuous variables, and screening of predictors by single factor analysis. Conclusions:The existing hypoglycemia risk prediction models for diabetic patients are still in the development stage, and medical and nursing staff can choose the existing hypoglycemia models according to the results of this systematic review and clinical practice. In the future, we should improve the existing models based on tools or carry out large-sample, multi-center, prospective cohort studies, and build a high-quality hypoglycemia risk prediction model for diabetic patients that is more suitable for China based on more comprehensive and accurate statistical methods and clinical data.
