Novel Informatics Approaches for Ascertainment of PAD Status and Adverse Outcomes

 描述（由申请人提供）：申请人和主要研究者（PI）是一名委员会认证的心脏病专家，医学助理教授和马约诊所心血管部门的高级副顾问。[The PI的长期目标是成为翻译信息学领域的独立临床医生-研究者，专注于开发、验证和部署电子工具，以改善心血管疾病患者的结局。PI计划利用EHR进行基于人群的研究，使用包括NLP在内的电子算法，并通过应用最先进的信息学方法在床旁提供知识，以支持外周动脉疾病（PAD）患者以患者为中心的决策。所开发的电子工具将可移植到不同的机构和其他卫生条件。PI将 作为大型学术卫生系统内的心脏病专家，在EHR和CDS的接口处使用新型NLP技术，以快速将基于社区的流行病学调查中获得的知识转化为患者护理。PI提出了一项研究职业发展计划，以获得翻译信息学方面的专业知识，包括电子表型分析，以及使用与临床决策支持（CDS）集成的新型应用程序自动计算和可视化显示预后评分。PI将完成信息科学的综合课程，并获得额外的以患者为导向的流行病学研究经验。主要导师将是医学博士伊夫蒂哈尔·库洛，心脏病专家和马约诊所电子医学记录和基因组学（eMERGE）网络的PI，以及PAD基因组研究电子表型的开发和应用的领导者。指导团队还包括克里斯托弗·丘特博士，医学博士，马约诊所医学信息学负责人，信息学和流行病学专家，马约eMERGE基金的共同主要研究者和罗切斯特流行病学项目的共同研究者; Hongfang Liu，博士，自然语言处理（NLP）及其在基于EHR的表型分析中的应用方面的专家;和肯特贝利博士，具有制定预后风险评分专业知识的高级生物统计学家。该提案的具体目标是：目标1 -应用包括临床记录NLP的表型分型算法，[a]识别1998-2011年社区中的PAD病例和无PAD的对照;[B]确定截至2013年12月的不良心血管和肢体结局。目的2 -创建多变量预后风险模型和社区PAD病例不良结局评分。我们将评估PAD患者与无PAD对照组相比的相对死亡风险。目的3 -开发和评价马约EHR中一种新型电子应用程序的有用性，以检索相关数据元素，计算和显示个性化预后评分。在完成拟议的调查后，可交付成果将是新的知识和PAD患者的电子健康诊断工具，将传播给利益相关者，包括临床医生、患者和研究人员，便于其他机构和其他心血管疾病。

 DESCRIPTION (provided by applicant): The applicant and principal investigator (PI) is a board-certified cardiologist, Assistant Professor of Medicine and Senior Associate Consultant in the Mayo Clinic Cardiovascular Division. [The long-term goal of the PI is to become an independent clinician-investigator in translational informatics focused on the development, validation and deployment of electronic tools to the point-of-care to improve outcomes for patients with cardiovascular disease. The PI intends to leverage the EHR to conduct population-based studies using electronic algorithms that include NLP and by application of state-of-the-art informatics approaches to deliver the knowledge at the point-of-care in support of patient-centered decision-making for patients with peripheral arterial disease (PAD). The electronic tools developed will be portable to different institutions and other health conditions. The PI will be uniquely positioned as a cardiologist within a large academic health system working with novel NLP techniques at the interface of the EHR and CDS to rapidly translate knowledge acquired in community-based epidemiologic investigations to patient care.] The PI has proposed a research career development plan to acquire expertise in translational informatics including electronic phenotyping, and automated calculation and visual display of prognostic scores using a novel application integrated with clinical decision support (CDS). The PI will complete comprehensive coursework in informatics science and gain additional patient-oriented and epidemiologic research experience. The primary mentor will be Iftikhar Kullo, M.D., cardiologist and PI of the Mayo Clinic electronic MEdical Records and GEnomics (eMERGE) network and a leader in the development and application of electronic phenotyping for genomic studies of PAD. The mentoring team also includes Dr. Christopher Chute, M.D., Head of Medical Informatics at Mayo Clinic, an expert in informatics and epidemiology, and co-PI of the Mayo eMERGE grant and co-investigator of the Rochester Epidemiology Project; Hongfang Liu, Ph.D., an expert in natural language processing (NLP) and its application for EHR-based phenotyping; and Kent Bailey, Ph.D., senior biostatistician with expertise in development of prognostic risk scores. The specific aims of the proposal are: Aim 1 - Apply phenotyping algorithms that include NLP of clinical notes to [a] identify PAD cases and controls without PAD in the community from 1998-2011; and [b] ascertain adverse cardiovascular and limb outcomes through December 2013. Aim 2 - Create multivariable prognostic risk models and scores for adverse outcomes in PAD cases from the community. We will estimate the relative risk of death in PAD patients compared to controls without PAD. Aim 3 - Develop and evaluate usefulness of a novel electronic application within the Mayo EHR to retrieve relevant data elements, calculate, and display individualized prognostic scores. On completion of the proposed investigations the deliverables will be new knowledge and an e-health prognostication tool for PAD patients to be disseminated to stakeholders including clinicians, patients, and researchers portable to other institutions and other cardiovascular diseases.