Aiding Effective Decision Making in Dental Research Using Network Meta-analysis

描述（由申请人提供）：牙科手术的比较有效性研究（CER）从根本上依赖于对治疗效果的准确评估，通常通过多个临床指标来衡量。为了对最佳治疗进行排名和识别，需要以统一的方式将这些多个临床指标结合起来。按照目前的做法，分别报告每种结果的排名和最佳治疗的概率，往往会导致相互矛盾的结果，这对牙医和患者做出治疗决策没有帮助。针对特定牙齿状况的治疗选择越来越多，以及其成本的快速上升，已经产生了越来越多的对牙科临床实践中的治疗程序进行科学严格的同步比较的需求。网络荟萃分析（NMA）也被称为混合或多重治疗Meta分析，通过同时分析随机对照试验中干预措施的直接比较和试验间的间接比较，扩展了传统成对荟萃分析的范围。与比较两种治疗的传统荟萃分析相比，NMA提出了许多额外的统计挑战。例如，试图比较和排名大量的治疗，比如说一打，使用随机试验的数据，单独比较两种（或几种）治疗，会导致大量的数据缺失，因为在特定试验中没有研究的治疗结果是设计缺失的。目前可用的统计方法，这是基于治疗对比，只侧重于相对治疗效果的估计，有几个严重的局限性。该提案的总体目标是开发尖端的统计方法，并将其整合到公开可用的易于使用的软件中，以增强牙科研究中的NMA。具体来说，我们将从缺失数据方法的角度开发多个混合结局的多变量贝叶斯分层模型，具体目标有以下三个：1）将联合收割机（例如，2）对邻间口腔卫生方法在降低炎症临床指标方面的作用进行系统性评价和网络荟萃分析; 3）制作用户友好的免费开源软件，以促进牙科研究和其他研究领域的NMA。我们建议通过重新分析牙科研究中的几个已发表的NMA，广泛的模拟和一个真实的病例研究，对所提出的方法的优点和缺点进行实证评估邻间口腔卫生方法在减少炎症的临床指标。这三个目标的实现将大大推进CER分析方法，以同时比较多个终点的多种治疗程序。

DESCRIPTION (provided by applicant): Comparative effectiveness research (CER) of dental procedures relies fundamentally on accurate assessment of treatment efficacy, which is commonly measured by multiple clinical indices. To rank and identify best treatments, those multiples clinical indices need to be combined in a unified manner. Reporting the rank and the probability of being best treatments separately for each outcome, as it is currently done, can often lead to conflicting results, which are not useful for dentists and patients when making treatment decisions. The growing number of treatment options for a given dental condition, as well as the rapid escalation in their costs, has generated an increasing need for scientifically rigorous simultaneous comparisons of treatment procedures in dental clinical practice. Also called mixed or multiple treatments meta- analysis, network meta-analysis (NMA) expands the scope of a conventional pairwise meta-analysis by simultaneously analyzing both direct comparisons of interventions within randomized controlled trials and indirect comparisons across trials. Compared to a traditional meta-analysis that compares two treatments, NMA presents many additional statistical challenges. For example, attempts to compare and rank a large number of treatments, say a dozen, using data from randomized trials that individually compare two (or a few) treatments results in a large amount of missing data since the outcomes for treatments not studied in a particular trial are missing by design. Currently available statistical methods, which are based on treatment contrasts, focus only on relative treatment effect estimates and have several serious limitations. The overall goal of this proposal is to develop cutting-edge statistical methods, and to integrate them into publicly available, easy-to-use software, to enhance NMA in dental research. Specifically, we will develop multivariate Bayesian hierarchical models for multiple mixed outcomes from the perspective of missing data methods with the following three specific aims: 1) to combine multiple mixed endpoints (e.g., binary, categorical and continuous responses) in a unified framework to rank and identify best treatments; 2) to conduct a systematic review and network meta-analysis of interproximal oral hygiene methods in the reduction of clinical indices of inflammation; 3) to produce user-friendly, free, open-source software to facilitate NMAs in dental research and other research areas. We propose to perform empirical assessment of the strengths and weaknesses of the proposed methods through reanalyzing several published NMAs in dental research, extensive simulations, and a real case study on interproximal oral hygiene methods in the reduction of clinical indices of inflammation. Completion of the three aims will substantially advance CER analytical methods for simultaneously comparing multiple treatment procedures across multiple endpoints.