Analysis for Incomplete Data in Oral Health/Ventilator-Associated Pneumonia Study

描述(由申请人提供)：我们的建议集中于开发一类新的和新的非参数似然方法，用于统计推断，以处理最近的临床试验-口腔健康和呼吸机相关性肺炎-A阶段III随机单中心试验(授权号：1 R01DE14685-01A1)中遇到的问题。共有175名重症监护病房(ICU)的患者接受了洗必泰口腔冲洗或安慰剂对照组的治疗，每天一到两次，并跟踪治疗直到他们出院。结果变量包括目标微生物的口腔定植、牙菌斑指数评分和肺炎的诊断变量。需要进行统计调查的三个主要问题如下。1)数据集中存在诸如口腔内菌斑指数和细菌定植分数等变量的数据损耗问题。在纵向研究中，实验单元的磨损是一个主要问题，目前非参数似然方法还没有很好地解决这一问题。特别是，对于口腔健康研究，来自患者的多个结果和相应的相关结构进一步使数据分析复杂化。2)一般结果的测量受仪器灵敏度的影响，由于检测范围的限制而无法获得数值，并受到测量误差的影响。3)数据中存在系统性缺失，这是因为只有当一个变量满足某个条件(例如，CPIS和触发BAL集合)时，才能观察到另一个变量。这个项目建议开发统计推断方法，使用非参数似然方法在存在不完整数据或数据损耗的情况下测试多个组。一些现有的参数似然(PL)方法可以解决缺失数据的问题，然而，对于不完整的数据，这些参数假设不能使用标准的拟合优度检验来检验。我们将发展一系列与不完全数据结构相关的非参数似然方法，其中考虑了缺失模式。这些新方法将允许用户通过使用非参数方法来避免强烈的分布假设。我们将特别注意利用不完全数据模式中保留的最大信息。这种新的方法将提供更强大和准确的分析。这种方法对于一般口腔健康数据的分析也立即有用，因为常见的龋齿或牙周病数据集充满了类似的缺失数据问题。为了帮助方法的转移，我们计划开发用户友好的软件。调查人员还计划在这项提议的背景下，培训学生和医学调查人员，使他们掌握正确和有力的方法来应对给定的挑战。这些方法的应用将使在临床研究中进行灵活而强大的推理成为可能。如果完全成功，我们相信所提出的方法有很大的潜力被采纳为主要的统计工具，改变不同临床领域的研究计划的实践。 与公共卫生相关：虽然进行临床试验是开发有益的和可靠的疾病治疗方法的必要步骤，但方法/统计方面的限制往往导致研究设计和后续数据分析的灵活性较低。继续开发能够处理实际临床试验中更现实问题的统计方法是当务之急。这项建议侧重于开发新的不完整/缺失数据的统计方法，这是临床试验中的一个常见问题，涉及基于“口腔健康与呼吸机相关肺炎--第三阶段随机单中心试验”的口腔健康研究。

DESCRIPTION (provided by applicant): Our proposal focuses on the development of a class of new and novel nonparametric likelihood methods for statistical inference to handle problems encountered during a recent clinical trial, Oral Health and Ventilator- Associated Pneumonia-A Phase III Randomized Single Center Trial (Grant number: 1 R01DE14685 - 01A1). A total of 175 patients in an intensive care unit (ICU) were treated with chlorhexidine oral rinse once or twice per day or with a placebo control, and followed until they were discharged. Outcome variables include oral colonization by target micro-organisms, the dental Plaque Index score and diagnostic variables for pneumonia. Three major issues that warrant the statistical investigation were as follows. 1) A data attrition problem exists in the data sets with respect to variables such as the Plaque Index and bacterial colonization scores in the oral cavity. In longitudinal studies, the attrition of the experimental units is a major problem and currently nonparametric likelihood methods have not been well developed to solve the problem. In particular, for oral health research, multiple outcomes from a patient and the corresponding correlation structure further complicate the data analysis. 2) General outcome measurements are subject to instrument sensitivity where values are not available because of the limit of detection and subject to measurement errors. 3) There is systematic missingness in the data due to the fact that one variable is observed only if the other variable satisfies a certain condition such as exceeding a threshold (e.g., CPIS and triggering collections of BAL). This project proposes the development of statistical inference methods using the nonparametric likelihood approaches to test multiple groups in the presence of incomplete data or data attrition. Some available parametric likelihood (PL) approaches can address the missing data problem, however, for incomplete data, these parametric assumptions cannot be tested using standard goodness-of-fit tests. We will develop a series of nonparametric likelihood methods relevant to the structure of incomplete data, where missing patterns are taken into account. These new methods will allow the users to avoid strong distributional assumptions by using a nonparametric approach. We will pay special attention toward utilizing the maximum information retained in the pattern of incomplete data. This novel approach will provide more powerful and accurate analyses. This approach is also immediately useful for the analysis of oral health data in general, since common dental caries or periodontal disease datasets are riddled with similar missing data problems. To help transfer of the methodology, we plan to develop user-friendly software. The investigators also plan, in the context of this proposal, to train students and medical investigators with the correct and powerful approaches to the given challenges. Application of these methods will enable flexible and powerful inference in clinical investigations. If fully successful, we believe that the proposed methods have a great potential to be adopted as a primary statistical tool that changes the practice of study planning for various clinical areas. PUBLIC HEALTH RELEVANCE: Although performing clinical trials is a necessary step to develop beneficial and fool-proof treatments for diseases, methodological/statistical limitations often cause a less flexibility in study designs and subsequent data analysis. Continuing development of statistical methods that can handle more realistic problems from actual clinical trials is imperative. This proposal focuses on the development of the novel statistical methodology for incomplete/missing data, a common problem in clinical trials, with respect to the oral health research based on the study "Oral Health and Ventilator-Associated Pneumonia-A Phase III Randomized Single Center Trial".