摘要目的:本研究旨在构建一套科学、系统和可操作的专病数据库质量评价指标体系，通过系统评估专病队列数据库建设、数据质量和队列管理情况，提升数据规范性和研究支撑能力，为疾病研究、医疗政策制定和生物医药研发等提供可靠的数据基础。方法:通过系统分析国内外临床研究数据质量评价文献和政策，初步筛选专病数据库质量评价指标，采用两轮德尔菲法构建专病数据库质量评价体系，并运用层次分析法和YAAHP 7.5软件计算各级指标的相对权重和组合权重，采用SPSS与R(lavaan包)对问卷信度与结构效度进行分析。结果:两轮专家咨询问卷有效回收率均为100.00%，专家权威系数均为0.81，第二轮一级和二级指标的肯德尔协调系数分别为0.311和0.218( P<0.05)，表明专家意见具有良好的一致性。最终构建的专病数据库质量评价体系包括数据库建设(权重31.82%)、数据质量(41.49%)和队列建设(26.69%)3个一级指标，10个二级指标和32个三级指标。指标体系的整体一致性检验结果CR＝0.003 8，信度分析Cronbach′s α＝0.96，结构方程模型拟合指标IFI＝0.83，GFI＝0.89，验证了其科学性与稳定性。该体系已在某市58个专病数据库中应用，评估结果显示数据库建设方面的得分提升最为显著，不同类型医疗机构的队列质量亦存在差异。 结论:本研究构建了具有良好信度与效度及指标权重合理的专病数据库质量评价指标体系，并在实证应用中展现出较强的适用性与推广价值。该体系可有效支持专病队列数据库的质量评估，促进数据治理能力提升，为临床研究、公共卫生政策与医药创新提供重要支撑。

abstractsObjective:This study aims to construct a quality evaluation index system for specialized disease databases. Through systematic assessment and optimization, it seeks to comprehensively enhance the quality and standardization of specialized disease cohort data. This initiative will provide more precise and reliable data support for disease research, the development of innovative drugs and medical devices, as well as policy formulation.Methods:By conducting a thorough analysis of domestic and international literature and policies related to clinical research data quality evaluation systems, preliminary quality evaluation indicators for specialized disease databases were established. Utilizing the Delphi method in two rounds, a quality evaluation system for specialized disease databases was constructed. The Analytic Hierarchy Process (AHP) and YAAHP 7.5 software were then employed to calculate the relative weights of indicators at various levels and their composite weights.Results:The two rounds of expert consultation achieved a 100.00% valid response rate, with an expert authority coefficient of 0.81 in both rounds. In the second round, the Kendall′s coordination coefficients for the first-level and second-level indicators reached 0.311 and 0.218, respectively ( P<0.05), indicating a good level of consensus among experts. The final specialized disease database quality evaluation system consists of 3 first-level indicators, 10 second-level indicators, and 32 third-level indicators. The first-level indicators include database construction, data quality, and cohort development, with weight coefficients of 31.82%, 41.49%, and 26.69%, respectively. The scientific validity of the indicator system was confirmed through reliability and validity analyses. When applied to assessing 58 specialized disease database projects from 36 medical institutions in a certain city, the results showed significant improvements in scores for database construction, data quality, and cohort development, with the most notable improvement observed in database construction. Conclusions:This study successfully developed a scientific, practical, and rationally weighted quality evaluation system for specialized disease databases, demonstrating high expert consensus and broad applicability.Validation studies have shown that this system effectively enhances the standardization and data quality of databases, providing robust technical support and assurance for specialized disease research and data resource sharing.