数字化智能与传统临床研究模式在观察性项目中的应用与对比

Application and Comparative Analysis of Digital Intelligent and Traditional Clinical Research Models in Observational Studies

关键词:  医院信息系统  临床试验  数字化  回顾性研究  医学信息学

Key words:  hospital information systems  clinical trial  digitalization  retrospective study  medical informatics

摘要:

目的 在真实的临床研究环境中全流程模拟应用数字化智能临床研究模式，将其在研究成本和效率方面的表现与传统临床研究模式进行比较。方法 在心血管内科的一项真实世界观察性临床研究项目中，采用数字化智能临床研究模式进行包括患者招募、数据录入、试验监查、项目稽查流程在内的全过程模拟验证，统计分析工时和数据录入准确率等关键指标，并与研究传统模式下的结果进行回顾性比较。结果 本项临床研究项目真实入组人数52人，数字化智能临床研究模式下筛选出4240名潜在患者，预计入组人数66人，筛选总用时6小时；数据录入阶段，临床协调员的工时缩短(数字化智能临床研究模式105.71h VS 传统临床研究模式141.9h)且准确率提高(数字化智能临床研究模式97.23% VS 传统临床研究模式85.54%)；数据监查阶段，临床监查员监查总工时减少(数字化智能临床研究模式72.00 h VS 传统临床研究模式169.50 h)且质疑数量减少(数字化智能临床研究模式1156条 VS 传统临床研究模式1541条)。结论 在本研究中，数智化临床研究模式通过使用人工智能显著提高了患者招募效率和数据录入准确性，同时减少了操作时间和成本。数智化模式展现了在提高临床试验效率和数据质量方面的明显优势。未来研究需进一步优化技术并扩大其应用范围，以充分利用其潜力。

Abstract:

ABSTRACT: OBJECTIVE This study aims to simulate the full process application of digital intelligent clinical research model in a real clinical research setting and compare its performance in terms of research cost and efficiency with the traditional clinical research model. METHODS In a real-world observational clinical research project in the field of cardiology, the digital intelligent clinical research model was used to simulate and verify the entire process, including patient recruitment, data entry, trial monitoring, and project inspection. Key indicators such as working hours and data entry accuracy were statistically analyzed, and the results were retrospectively compared with those under the traditional research model. RESULT A total of 52 patients were enrolled in the clinical research project. Under the digital intelligent clinical research model, 4,240 potential patients were screened, with an expected enrollment of 66 people and a screening total time of 6 hours. During the data entry phase, clinical coordinators' working hours were reduced (digital intelligent clinical research model: 105.71h vs. traditional clinical research model: 141.9h) and accuracy improved (digital intelligent clinical research model: 97.23% vs. traditional clinical research model: 85.54%). In the data monitoring phase, total monitoring hours for clinical monitors were reduced (digital intelligent clinical research model: 72.00h vs. traditional clinical research model: 169.50h), and the number of queries decreased (digital intelligent clinical research model: 1541 queries vs. traditional clinical research model: 1156 queries). CONCLUSION In this study, the digitalized clinical research model significantly enhanced patient recruitment efficiency and data entry accuracy through the use of artificial intelligence, while also reducing operational time and costs. The digitalized model demonstrated clear advantages in improving the efficiency and quality of clinical trials. Future research needs to further optimize the technology and expand its application scope to fully leverage its potential.