脾胃系疾病计算机自适应测评体系的临床解释度和属性考评研究

Many developed instruments were never used in clinical practices and researchers, because the patients and doctors did not understand the scores meanings and the scales usually have extensive burdens. To solve the problem, we applied for this project entitled 'Clinical interpretation and psychometric assessment of Patient Reported Outcomes Instrument for Patients with Gastrointestinal Diseases Based on Computer Adaptive Test'. This project aims to: 1) describe the meaningful interpretations of health status using the developed Patient Reported Outcomes Instrument for Patients with Gastrointestinal Diseases Based on Computer Adaptive Test (GD-CAT); ii) develop intellectual platforms for GD-CAT assessments in clinical environments and patient's daily life, such as in computer and telephone, etc; iii) explore the clinical interpretation research methodologies and translational ways for primary clinical practices. This study adopted the combination of prospective, longitudinal, cohort and retrospective design. In step 1, the patients with Gastric Stuffiness or Gastric Pain (Chronic gastritis or Functional dyspepsia in western medicine) will be prospective enrolled in the field out-department and in-department of hospital and online survey. With the Patient Reported Outcomes Instrument for Patients with Gastric Stuffiness Based on Computer Adaptive Test (GS-CAT) and Patient Reported Outcomes Instrument for Patients with Gastric Pain Based on Computer Adaptive Test (GP-CAT), data will be collected on the admission and 14 days later. Meanwhile, the retrospective data from previous studies performed between 2010 and 2012 were also included. After statistical process control, SPSS13.0, RUMM2030, Lisert8.70, MULTILOG7.03, FASTtest Web, et al. will be used for analysis. Items and scales properties will be evaluated with comparative fit index (CFI), Root Mean Square Error of Approximation (RMSEA), Factor loadings, item information curves (IIC), item characteristic curves (ICCs), etc., and the health status, 90% Confidence interval (90%CI), reliable change index (RCI) etc. will be also evaluated for interpretations establishment. In step 2, to evaluate the scientific and clinical properties of GS-CAT and GP-CAT interpretations, we will first develop the intellectual platforms in independent computer, mobile telephone or network for GD-CAT assessments which used in clinical environments and patient's daily life. Adopted the prospective, longitudinal, cohort study design, data will also be collected on the admission and 14 days later with GD-CAT intellectual platforms and primary clinical interpretation system. SPSS13.0, RUMM2030, Lisert8.70, MULTILOG7.03, FASTtest Web, et al. will be used for analysis in items, scale property, patients health and clinical feasibility assessment. The final interpretation system will be established after above mentioned analysis. It will decrease the burdens and promote the clinical application of developed instruments.

基于计算机自适应测试和项目反应理论下的量表临床意义解释研究是促进量表向一线临床实践转化的重要途径。本研究采用前瞻、纵向、队列研究和回顾性病例调查相结合设计，通过医院和网络调查招募胃痞和胃痛（慢性胃炎和功能性消化不良）患者，分别在纳入研究和随访结束时以胃痞患者报告结局量表计算机自适应测试版、胃痛患者报告结局量表计算机自适应测试版进行数据收集，使用RUMM2030、Lisert8.70、MULTILOG7.03、FASTtest Web等软件分析各条目、领域特征并计算受试者健康水平等，进而建立量表初步的临床解释方法与体系； 然后，在上述工作基础上采用相同研究设计，以基于独立电脑和互联网络研发胃痞和胃痛患者报告结局量表的网络智能终端测试体系在纳入研究和随访结束时收集数据，对量表的临床解释体系进行科学性考评，评估量表临床解释标准的临床可信性，进而降低量表回答负担和管理负担，促进量表向临床实践转化。

临床结局和量表研究已被国际药物经济与疗效研究协会、美国国家食品药品监督管理局、欧洲医学研究所等政府机构和顶级国际研究组织广泛接受。然而，测试效率低及结局解释体系薄弱是当前国内外量表研究所面临的两大难题。为提高量表测评的应用效能，丰富量表的测评结局解释方法，本研究基于项目反应理论及计算机自适应等现代测量方法，引进国内外多领域的优秀测量和评价技术，着眼于临床结局量表的解释体系构建以及临床高效测评方法构建两方面，寻求临床结局量表研究的新突破，拟一方面建立脾胃系疾病患者报告结局量表计算机自适应测试版的临床意义解释体系，另一方面建立量表智能测评平台，使患者能随时通过网络等在短时间内测评自身健康水平。. 本研究采用队列研究与回顾性调查结合设计，运用多种方式结合进行病例采集随访，共采集202例功能性消化不良、102例慢性胃炎、250例健康人临床信息，其中功能性消化不良患者和慢性胃炎患者采用功能性消化疾病生存质量量表和脾胃消化疾病患者报告结局量表慢性胃炎模块，完成间隔14天随访。然后，基于结构方程模型、项目反应理论及计算机自适应测试方法，综合运用SPSS，RUMM，AMOS及R等软件，数据进行整理后， 遵循评估基线和反应度、确定量表得分的可信度、建立量表得分的分布谱、确定统计学可信变化的指标群、确定临床可信变化的阈值、基于综合证据的临床推荐等流程，分析各条目和领域参数特征，如Effect Size (ES)、 standardized response mean (SRM)、 standard error of measure(SEM)、standard error of the difference (MDC)、RCI等指标群， 结合受试者工作曲线分析等方法建立量表的临床意义解释体系。然后，为进一步提高量表测评效率，基于多维计算机自适应测试的自动化测试平台研发，将量表的解释度体系构建与计算机技术应用相结合，建立移动智能疾病测评APP及网络管理平台，实现临床病例自动化采集、单节点多维度评价以及多节点阶段性变化评价，并通过数据可视化技术，为临床诊疗及个人健康管理提供直观的辅助工具。. 本研究各阶段研究依据研究计划顺利开展，实现了中医脾胃系量表的临床意义解释体系构建的方法学创新，促进中医量表从科研到临床实践的转化，为临床量表的应用与推广提供了良好的研究范例。