临床预测模型在糖尿病领域的应用

目的：分析临床预测模型在糖尿病领域的应用现状，为糖尿病预防相关研究提供方向和参考。方法：以2010-2021年Web of Science数据库中收录的糖尿病临床预测模型相关文献为基础，运用CiteSpace软件分析该领域的文献数量、国家、机构、作者、关键词和共被引文献。结果：糖尿病临床预测模型相关研究文献数量总体呈逐年增长趋势；在文献数量排名前10位的国家中，美国的文献数量和中心性最高，中国文献数量排在第2位但中心性最低，研究机构较为分散；心血管疾病、自我管理和妊娠期糖尿病是2020-2121年的研究热点，糖尿病肾病、低血糖和机器学习是未来临床预测模型的研究趋势。结论：应开发适用于我国糖尿病人群的预测模型，同时加强与英国、美国等国研究机构的合作。

Application of clinical prediction models in the field of diabetes

Objective By searching the core collection of Web of Science in the past 10 years, the application status of clinical prediction model in the field of diabetes was analyzed to provide the direction and reference for research related to diabetes prevention. Methods Based on the literature on diabetes-related clinical prediction models included in the Web of Science database from 2010 to 2021, the CiteSpace software was used to analyze the number of publications, countries, institutions, authors, keywords and co-citations in this field. Results The number of annual publications of diabetes- related clinical prediction models showed an increasing trend. The United States had the highest number of publications and centrality; China ranked second in the number of publications, but had the lowest centrality, and its research institutions were scattered. Cardiovascular disease, self-management and gestational diabetes have been the focus of research in the last 10 years. Diabetic nephropathy, hypoglycemia and machine learning are the research trends of future clinical prediction models. Conclusion We should develop prediction models suitable for Chinese with diabetes, and strengthen cooperation with the United Kingdom and the United States.