UCSF Data Science Training to Advance Behavioral and Social Science Expertise for Health Research (DaTABASE) Program

摘要 行为和社会科学研究人员对于推进缩小健康差距的研究至关重要。 下一代这样的研究人员必须配备最强大的、当代的分析 工具。新获得的数据和方法有可能改变所提出的问题，研究 所使用的设计和应用的统计方法，以加深我们对 提高我们改善人口健康和缩小差距的能力。利用，利用 这些机会，我们需要经过跨学科培训的研究人员，使他们能够将 传统的行为和社会科学专业知识，以及快速发展的 计算健康科学家。加州大学旧金山分校数据科学培训促进行为科学和社会科学 健康研究专业知识(数据库)计划编织了以前操作的培训链 独立于加州大学旧金山分校。数据库的独特优势包括：1)植根于严格的、 为研究设计和统计分析方面的基础提供跨学科量化培训计划 提供关于健康的可操作证据；2)集成多个数据流，包括：组学、临床(例如， 来自加州大学旧金山分校独特临床的全加州大学范围、地理位置和纵向医疗记录、点击流数据 信息学基础设施)、数字(基于设备的数据流、社交媒体数据、支持技术的队列)、 人口(例如，旧金山公共卫生部将卫生和社会服务与 无家可归的个人或其他弱势群体)和政策数据集(例如，管理 加州社区的大麻零售网点)；3)围绕 消除健康差距。数据库学员将完成严格的方法论培训 流行病学和转化学博士。其他内容培训，由加州大学旧金山分校的教职员工监督 健康和社区中心，将提供理论框架和学科背景 和行为科学。其他定量方法培训将通过Bakar计算提供 健康科学研究所(BCHSI)。BCHSI强调机器学习工具来挖掘小说的数据源 洞察和发现，包括精准医学和人口健康。这项培训计划将为 毕业生将这些分析工具应用于新的数据集，以研究行为和社会过程 潜在的健康差距。该数据库计划将提供密集的指导，从内容和 方法专家。数据库校友将准备领导针对当前和新兴公众的研究 创新的计算和数据科学分析方法带来的健康挑战。

Summary Behavioral and social science researchers are essential for advancing research on reducing health disparities. The next generation of such researchers must be equipped with the most powerful, contemporary analytic tools. Newly available data and methods have the potential to transform the questions asked, the research designs used, and the statistical approaches applied to deepen our understanding of fundamental drivers of health and increase our capacity to improve population health and reduce disparities. To take advantage of these opportunities, we need researchers with interdisciplinary training that allows them to connect the traditional behavioral and social science expertise with the rapidly evolving technical repertoire of computational health scientists. The UCSF Data Science Training to Advance Behavioral and Social Science Expertise for Health Research (DaTABASE) program braids training strands previously operating independently within UCSF. Particular strengths of DaTABASE include: 1) Grounding in a rigorous, interdisciplinary quantitative training program for a foundation in study designs and statistical analyses to deliver actionable evidence on health; 2) Integration of multiple data streams, including: ‘omics, clinical (e.g., UC-wide, geolocated and longitudinal medical records, clickstream data from UCSF’s unique clinical informatics infrastructure), digital (device-based data flows, social media data, technology enabled cohorts), population (e.g., San Francisco Department of Public Health data linking health and social services provided to homeless individuals or other vulnerable populations) and policy data sets (e.g., local policies regulating cannabis retail outlets in California communities); 3) training and research activities organized around the elimination of health disparities. DaTABASE trainees will complete the rigorous methodological training for a PhD in Epidemiology and Translational Science. Additional content training, overseen by faculty in the UCSF Center for Health and Community, will provide theoretical frameworks and disciplinary background for social and behavioral sciences. Additional quantitative methods training will be delivered via the Bakar Computational Health Sciences Institute (BCHSI). BCHSI emphasizes machine learning tools to mine data sources for novel insights and discovery, including precision medicine and population health. This training initiative will prepare graduates to apply those analytic tools to novel data sets for research on behavioral and social processes underlying health disparities. The DaTABASE program will provide intensive mentoring from both content and methods experts. DaTABASE alumni will be prepared to lead research addressing current and emerging public health challenges with innovative computational and data science analytic approaches.