基于区域健康医疗大数据脑卒中专病库的构建在中医药研究中的应用

目前国内基于区域诊疗数据构建的单病种数据库在中医药领域应用较少，本研究依托天津健康医疗大数据平台建立脑卒中专病库，探讨区域诊疗数据从形成专病库到深入挖掘临床问题所具有的独特优势和缺陷，为后续中医药领域区域诊疗数据下专病库的构建和应用提供参考。专病库构建需要经过跨机构数据采集、整合和治理，同时需兼顾数据安全，以此作为研究开展基础。建设完成后通过描述人群现状和预后情况，总结专病库概况和可开展的研究方向。最后对研究中数据治理、分析时存在的问题和解决方案进行梳理，从方法学角度讨论专病库应用与中医药研究的适配性。专病库覆盖天津市81家医院2016~2020年24万首发脑卒中患者诊疗数据，并链接公卫系统死亡登记数据，可进行人群特征分析、发病影响因素探查、诊疗路径优化、预测模型构建和卫生经济学研究。但也存在数据覆盖度不全和记录不完善的问题，并伴随真实世界研究中常出现的各种偏倚，需要在研究设计及方法学应用上加以调整。真实世界下的诊疗数据应用是探索和提升中医药疗效的必由之路，但正如脑卒中专病库展现出的研究优势和不足，数据挖掘是循序渐进的过程，需要通过将研究需求反哺于临床电子病历的记录以提升专病库质量，为承载后续更多高质量的中医药研究奠定基础。

Application of construction on stroke database based on regional health care big data in traditional Chinese medicine

At present，the specific disease database based on regional diagnosis and treatment data is rarely used in the field of traditional Chinese medicine in China. Our study relies on Tianjin Health Big Data Company to establish a special disease database for stroke，to explore the advantages and defects of regional diagnosis and treatment data from the formation of a special disease database to in-depth exploration of clinical problems，and provide references for the subsequent construction and application of a special disease database under regional diagnosis and treatment data in the field of traditional Chinese medicine. The construction of the special disease database needs to go through cross-institutional data collection，integration and governance，while taking into account data security as the basis for research. After the establishment，the general situation of the special disease database and the possible research direction were summarized by describing the population status and prognosis. Finally，the problems and solutions of data governance and analysis in the study were sorted out，and the adaptation of the application of special disease database to the research of traditional Chinese medicine was discussed from the perspective of methodology. The special disease database covers the diagnosis and treatment data of 240 000 first-episode stroke patients in 81 hospitals in Tianjin from 2016 to 2020，and is linked to the death registration data of the public health system. It can analyze the characteristics of the population，explore the influencing factors of the disease，optimize the diagnosis and treatment path，construct the prediction model and research in health economics. However，there are also problems of incomplete data coverage and incomplete records，accompanied by various biases that often occur in real-world research，and need to be adjusted in research design and methodological application. The application of diagnosis and treatment data in the real world is a necessary method to explore and improve the efficacy of traditional Chinese medicine. However，just as the advantages and disadvantages getting along with the research of stroke special database，data mining needs a gradual process. It is necessary to improve the quality of the special disease database by filling the clinical electronic medical records entirely according to research requirement，and lay the foundation for undertaking more high-quality Chinese medicine research in the future.