供需联动导向下基于智能检后随访的慢性肾脏病早筛流程重组研究

The increasing evidences and our previous survey show that the prevalence of chronic kidney disease ( CKD ) in the general population is as high as 10%. The end-stage renal disease and related complications caused by it are seriously threatening people's health and the burden of CKD is growing. CKD is with insidious onset and long asymptomatic period. The course of CKD can be well controlled and even reversed if it can be found and intervened early. However, the awareness rate of CKD population in China is only 12.5%, and the monitoring mechanism of high risk population is lack. Health examination is not only the "outpost" of the chronic disease management, but also an important source of information for the risk assessment of chronic diseases. Accordingly, based on the established cohort and follow-up platform, this project plans to develop early CKD management technology ( risk assessment scale, prognostic index, action willingness model, and standard operation procedure) and integrated them into follow-up information system. It is proposed on the basis of evidence-based medicine, statistics, behavior change theory and standard operation guidance. The artificial neural networks will be used in the study of constructing early prediction of CKD. In order to verify its value, we plan to carry out the pilot intervention and effect evaluation. It is expected to provide work ideas and reference for the establishment of an appropriate early screening mode of chronic disease for physical examination.

大型流调及我们研究均显示，慢性肾脏病（CKD）患病率已超10%。其导致终末期肾病及并发症严重威胁民众健康，疾病负担日益加剧。CKD起病隐匿，早发现及干预可控制甚至逆转病程。但我国CKD人群知晓率仅12.5%，且缺乏高危人群监测机制。健康体检是慢病健康管理“前哨”，也是开展慢病风险评估重要数据来源。鉴于此，拟基于已建体检队列及随访平台，在临床指南、风险评估、健康信念模型及质量管理理论指导下，利用多学科交叉，通过文献分析、专家咨询、现场调查、焦点组访谈、回归分析、BP人工神经网络、软件编制等方法，研制CKD筛查干预管理工具（预后风险指数、筛查意愿指数和标准化随访作业规程）并嵌入随访信息系统，形成以循证、定量模型及智能软件导航的旨在促进CKD筛查的供需联动检后随访模式，并采用整群随机对照进行试点干预及效果评价。项目对提高CKD早诊，促进健康体检在慢病防控中发挥更大作用具重大创新及应用价值。

我国CKD人群知晓率低，缺乏高危人群监测机制。本项目拟基于已建体检队列及随访平台，在循证证据、健康行为干预及质量管理理论指导下，利用多学科交叉，研制CKD筛查干预管理工具并嵌入随访信息系统，形成基于循证、定量模型及智能软件的旨在促进CKD筛查的检后随访模式并进行效果评价。项目基本按计划顺利实施，主要预期目标基本实现。. 经过三年实施，课题组研制出了一套基于数据建模（通过深度学习构建CKD风险评估模型和以健康信念模型为框架的CKD筛查意愿模型）的高危人群CKD筛查促进标准化随访流程，开发了CKD风险感知和筛查意愿测量工具，优化了体检CKD慢病健康管理随访信息功能模块。课题组按照上述重组后的检后随访干预流程及信息化辅助系统，实施了类实验干预效果评价。通过比较干预前后变化了解干预效果，同时设立对照组以分析干预净效应。评价内容包括调查对象有关CKD 筛查的知识、态度、行为等，以干预效果、净效应百分比和自然效应百分比作为评估指标。干预前1364人（应答88.1%）、干预后1294人（应答94.2%）和对照组973人（应答84.5%）。调查对象对不同时点进行的干预和评价调查依从性达到76.2%以上。干预组对于“高血压病、糖尿病是CKD的重要危险因素”的认知上升了26.8%，对于“平稳控制血压和血糖的目的还在于保护肾功能”的认知上升了23.3%，对于超重/肥胖标准和合适盐摄入量的认知分别增加了27.1%和24.4%。干预组常吸烟率、常饮酒率和定期体育锻炼率在干预前后无显著性差异。干预组常测量血压的比例增幅大于11.2%。以上改变归因于干预措施的效应比均大于50%。. 依托项目资助，已公开发表4篇学术论文，其中包括2篇SCI论文和2篇CSCD论文；获得2项相关软件著作权；获评全国学术年会优秀论文2次；国内学术会议报告2次和省内质控培训2次。课题组还将开发的评估量表和辅助随访信息模块嵌入并运用于高危人群检后随访管理中，并于所在平台的健康医联体合作，拓展干预功能社区和人群。项目对提高CKD早诊，促进健康体检在慢病防控中发挥更大作用具一定创新及应用价值。