STATISTICAL METHODS FOR QUALITY-OF-LIFE OUTCOME RESEARCH

患者生活质量结果指标越来越重要， 在确定新的有效性和净效益方面的重要作用 和现有的药物疗法。 质量的解释- 用于治疗决策和政策规划的生命数据需要a） 关于测量和统计特性的有效假设 评估中使用的生活质量量表，B）代表性 将患者特异性生活质量结局与 有效性、效用、成本和总体净效益，以及3）适当 推理统计方法来测试治疗效果， 差异 虽然有大量的文献在讨论 各种模式、手段和工具的意义、效用和有效性， 生活质量的定义，很少有研究集中在这些 影响生活质量分析和解释的问题 用于确定相对风险和获益的结局数据 替代药物治疗方案。 本项目的主要目标是 建议a）评估和分析电流和电势测量 生活质量相关患者中使用的统计技术 通过对治疗和统计结果的详尽审查， 文献，B）改进现有分析和统计方法 因此，它适用于治疗科学和评估 药物治疗和c）开发数据分析，演示 在三个主要的生活质量分析领域内的项目， 现有的生活质量临床试验的选择方法 高血压（n = 2，500）、糖尿病（n = 300）和HIV（n = 1000）。 所有这些数据库都包含大量关于质量的前瞻性数据 在替代药物的随机多中心临床试验期间， 疗法 模型审查、发展和 示范将在以下几个主要方面进行： 信度、反应度、效度和敏感度 生活质量测量; 2）多变量，纵向模型（两者 参数和非参数），以评估生活质量治疗效果 在随机临床试验和3）全球模型的汇总估计 用于质量调整生存、成本效益、成本效用 和健康状态转移马尔可夫模型。最终的产品 研究将是生活质量分析的综合纲要， 模型，适用于生活质量的适当统计方法 相关治疗结果研究和实际数据分析 对临床研究者有用的示范项目，卫生政策 规划者和监管机构，以设计，评估和 解释药物治疗的生活质量研究。

Quality-of-life patient outcomes measures are playing an increasingly important role in determining the effectiveness and net benefit of new and existing pharmaceutical therapies. The interpretation of quality-of- life data for therapeutic decision making and policy planning requires a) valid assumptions concerning measurement and statistical properties of the quality-of-life scales used in assessment, b) representative analytical models relating patient-specific quality-of-life outcomes to effectiveness, utility, costs and overall net benefits and 3) appropriate inferential statistical methodology to test therapeutic effects and differences. While a large body of literature exists debating the meaning, usefulness and validity of various models, instruments and definitions of quality of life, little research has focused upon those issues impacting on the analysis and interpretation of quality-of-life outcome data used in determining the relative risks and benefits of alternative pharmaceutical regimens. The major objectives of this proposal are a) to evaluate and analyze current and potential measurement and statistical techniques used in quality-of-life related patient outcomes through an exhaustive review of the therapeutic and statistical literature, b) to refine existing analytical and statistical methodology so that it is appropriate for therapeutic science and the evaluation of pharmacologic therapies and c) to develop data analysis, demonstration projects within three major quality-of-life analytical areas by applying the selected methods to existing quality-of-life clinical trials databases of hypertension (n = 2,500) diabetes (n = 300) and HIV (n = 1000). All these databases contain extensive prospective data on quality of life during randomized multicenter clinical trials of alternative drug therapy. The three-step methodology of model review, development and demonstration will be employed in the following major areas: 1) reliability, responsiveness, validity and sensitivity of health-related quality-of-life measures; 2) multivariate, longitudinal models (both parametric and nonparametric) to assess quality-of-life treatment effects in randomized clinical trials and 3) global models for summary estimates for use in quality-adjusted survival, cost-effectiveness, cost-utility and health state transition Markov models . The final product of this research will be a comprehensive compendium of quality-of-life analytical models, appropriate statistical methods suitable for quality-of-life related therapeutic outcomes research, and practical data analysis demonstration projects useful to clinical investigators, health policy planners and regulatory agencies for designing, evaluating and interpreting quality-of-life studies of pharmacologic therapies.