Computational methods using electronic health records and registry data to detect and predict clinical outcomes in rheumatic disease

使用电子健康记录和登记数据检测和预测风湿病临床结果的计算方法

• 批准号: 10349472
• 负责人: Milena Anne Gianfrancesco
• 金额: $ 1.26万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-04-10 至 2022-03-04
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10349472
• 关键词: Address; Adverse event; Age; Algorithms; Antirheumatic Agents; Area; Autoimmune Diseases; Biological; Biological Products; Biological Response Modifier Therapy; California; Categories; Cessation of life; Characteristics; Clinic; Clinical; Clinical Sciences; Clinics and Hospitals; Code; Combination Medication; Computing Methodologies; Data; Data Set; Databases; Demographic Factors; Development Plans; Disease; Disease-Modifying Second-Line Drugs; Electronic Health Record; Environment; Epidemiologist; Epidemiology; Ethnic Origin; Event; Future; Goals; Gold; Hospitalization; Individual; Infection; Informatics; Institutes; Interdisciplinary Study; K-Series Research Career Programs; Lead; Marketing; Mentored Research Scientist Development Award; Mentors; Mentorship; Methods; Modeling; Monitor; Morbidity - disease rate; NCI Scholars Program; Opportunistic Infections; Patient-Focused Outcomes; Patients; Pattern; Pharmaceutical Preparations; Pharmacotherapy; Population; Positioning Attribute; Probability; Quality of life; Race; Randomized Controlled Trials; Recommendation; Registries; Reporting; Research; Research Personnel; Rheumatism; Rheumatoid Arthritis; Rheumatology; Risk; Risk Assessment; Risk Factors; Safety; Sample Size; San Francisco; Sensitivity and Specificity; Serious Adverse Event; Severities; Structure; Subgroup; System; Systemic Lupus Erythematosus; Time; Training; Translational Research; Universities; University Hospitals; Validation; Work; adverse event risk; base; career development; comorbidity; data registry; disorder control; electronic structure; ethnic minority; experience; high risk; improved; individual patient; individualized medicine; infection rate; infection risk; large datasets; medical schools; medical specialties; patient safety; patient stratification; personalized medicine; population based; predict clinical outcome; predictive modeling; programs; racial and ethnic; research and development; risk stratification; safety outcomes; sex; skills; sociodemographics; statistics; structured data; symposium; text searching; unstructured data

PROJECT SUMMARY / ABSTRACT This is a new application for a K01 award for Dr. Milena Gianfrancesco, an epidemiologist at the University of California, San Francisco (UCSF) School of Medicine, who plans a research program focusing on understanding risk factors as they relate to rheumatic disease patient outcomes, such as adverse events. Combined with a training plan focused on computational text mining methods and advanced causal inference statistics, the goal of the current study is to use large electronic health record and national registry data that reflects real-world prescribing patterns to examine the risk of infection attributed to biologic disease-modifying anti-rheumatic drugs in individuals with rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE). While biologic medications have improved disease control and are associated with significant gains in patients’ quality of life, several studies have demonstrated that biologic use is associated with an increased risk of serious adverse events, such as infection. How this risk differs based on a variety of patient factors, such as age, race, and ethnicity, is currently unknown, leaving clinicians with insufficient information to predict the probability of an adverse event occurring in a given patient who is prescribed a particular biologic. This proposal will utilize established local electronic health record and national registry data to examine over 80,000 individuals with RA and SLE to address three specific aims. In Aim 1, Dr. Gianfrancesco will apply and validate a text mining system to identify incident clinical and opportunistic infections from clinical notes. In Aim 2, Dr. Gianfrancesco will use the same databases to determine the longitudinal causal effect of biologics on risk of infection. In Aim 3, a risk-assessment model to predict risk of infection will be developed and validated in a rheumatology clinic. Findings from this study will further elucidate factors associated with infectious risk for individuals prescribed biologics, thereby improving their safety in the ambulatory settings. Dr. Gianfrancesco has assembled an exceptional mentorship team with expertise in computational text mining methods, advanced causal inference statistics, rheumatology and patient safety outcomes, as well as experience using national registry data to address these questions. She will have access to a rich research environment and provided support for career development through programs such as the UCSF Clinical and Translational Science Institute K-scholars program. Formal coursework and mentoring will also be supplemented with attendance at national conferences related to rheumatology, epidemiology, and informatics. Completing the proposed research and career development plan will allow Dr. Gianfrancesco to gain experience in state-of-the-art computational methods using large datasets to better understand important patient outcomes, such as serious adverse events. This mentored career development award will provide the skills, mentorship, and experience necessary to propel her to independence and enable her to lead an independent multidisciplinary research program.

项目总结/摘要 这是一个新的申请K 01奖博士米莱娜Gianfrancesco，流行病学家在大学 加州，旧金山弗朗西斯科（UCSF）医学院，谁计划的研究计划，重点是 了解与风湿性疾病患者结局相关的风险因素，如不良事件。 结合一个培训计划，重点是计算文本挖掘方法和高级因果推理 统计，目前研究的目标是使用大型电子健康记录和国家登记数据， 反映了现实世界的处方模式，以检查归因于生物疾病修饰的感染风险 类风湿性关节炎（RA）和系统性红斑狼疮（SLE）患者的抗风湿药物。 虽然生物药物改善了疾病控制，并与患者的治疗显著增加有关， 生活质量，几项研究表明，生物制品的使用与生活质量的风险增加有关。 严重不良事件，例如感染。这种风险如何根据各种患者因素而有所不同，例如 年龄，种族和民族，目前尚不清楚，使临床医生没有足够的信息来预测 在处方特定生物制剂的给定患者中发生不良事件的概率。 该提案将利用已建立的当地电子健康记录和国家登记数据， 80，000名RA和SLE患者，以实现三个具体目标。在目标1中，Gianfrancesco博士将提出申请， 验证文本挖掘系统，以从临床笔记中识别偶发临床和机会性感染。在Aim中 2，Gianfrancesco博士将使用相同的数据库来确定生物制剂对 感染的风险。在目标3中，将制定一个风险评估模型，以预测感染风险，并在 风湿病诊所本研究的结果将进一步阐明与感染风险相关的因素 个人处方的生物制剂，从而提高他们在流动环境中的安全性。 博士Gianfrancesco组建了一个具有计算文本挖掘专业知识的杰出导师团队 方法，先进的因果推理统计，流变学和患者安全性结果，以及 利用国家登记册数据解决这些问题的经验。她将有机会接触到丰富的研究 环境，并通过诸如UCSF临床和 翻译科学研究所K学者计划。正式的课程和指导也将 并参加与风湿病学、流行病学和信息学有关的国家会议。 完成拟议的研究和职业发展计划将使Gianfrancesco博士获得 在使用大型数据集的最先进的计算方法，以更好地了解重要的经验 患者结局，如严重不良事件。这个指导职业发展奖将提供 技能，指导和经验，以推动她独立，使她能够领导一个 独立的多学科研究计划。