Optimizing Brief Alcohol Interventions for Young Adults via Computational Methods

点击翻译按钮获取中文摘要

Project Summary The goal of this Independent Scientist Award (K02) is to seek “protected time” for a period of intensive research focus to enhance the Candidate’s career development and to conduct research on the comparative effectiveness of competing brief alcohol interventions and their mechanisms. The proposed research strategy represents an ongoing, NIAAA-supported independent research program (R01 AA019511: Estimating Comparative Effectiveness of Alcohol Interventions for Young Adults, received 3rd percentile score). This K02 extends the R01 by (1) including an additional 25 data sets of individual participant data from brief alcohol interventions conducted between 2013 and 2020, and (2) embracing machine learning algorithms and other state-of-the-art statistical methods to help identify mechanisms of behavior change and to integrate and synthesize them across heterogeneous studies. The proposed protected time will be critical to update and enrich my research program in this current, rapidly changing clinical research environment. With accumulating high-quality clinical data and real-world data in the context of emerging data science capabilities, there exist opportunities as well as challenges. I am well suited to tackle emerging challenges for unprecedented opportunities, given my work in statistical data integration and broad experience in alcohol research. I want to create a large database where one can simultaneously examine secular/cohort effects, intervention effects, and effect heterogeneity at the most granular data level by developing and using new tools. My career goal is to become “one of the major hubs,” connecting disparate networks of researchers into a large connected network, promoting fluid interactions. The synergy from the large, interdisciplinary network would be helpful toward creating a “value” proposition of Big Data – the sum of the knowledge and insights from the collaboration by investigators is greater than isolated pursuits by each. Toward this goal, career development activities will be to (1) train in computational machine learning algorithms and real-world large-dimensional data analysis to facilitate collaboration with addiction researchers and statisticians; (2) enhance my understanding of technology-assisted adaptive interventions and passively collected outcomes data; (3) update my knowledge in basic health sciences related to addiction; and (4) submit at least two publications and two grants per year using newly acquired knowledge and skills proposed under K02. I have identified six highly active, nationally acclaimed experts in their respective fields. I will have individual and group research meetings with all six collaborators (three new collaborators for this K02) to develop innovative projects. In addition, I will attend workshops and conferences for training, carry out online training, and have guided reading and discussion with collaborators. I will visit collaborators for in-depth discussions and problem-solving. Improving big data capabilities for biomedical and clinical research has been one of the major strategic plans of the NIH, and I am motivated and prepared to help meet that critical mission.