Workforce Innovation in Data Science, Education, and Addiction Research (WISER)

As we work to build addiction data science literacy, the field of addiction research could benefit from an increased workforce with clinical and data fluency. The complexities of addiction as a clinical domain present challenges, including the intersection of mental health and chronic pain. Understanding how these factors influence data collection, oftentimes due to subjective reporting, differential health outcomes, and longstanding challenges to care, can impact data analytics and interpretation. Moreover, the separation between clinical experts and data expertise can create additional barriers to advancing addiction research. Aligning addiction knowledge with data science expertise could enhance the potential of emerging addiction researchers. Another impediment to progress is limited representation of addiction researchers with dual data and clinical proficiency in the workforce, which could be attributed in part to a lack of awareness of the field during training (undergraduate to postgraduate). As we look to analyze larger and more heterogeneous addiction datasets, a global concern is the risk of algorithmic bias.
Developing training for a broad workforce that understands challenges at the intersection of addiction medicine and data science will accelerate our understanding of addiction's complexity. The long-term goal of this Workforce Innovation in Data Science, Education, and Addiction Research (WISER) R25 application is to support the training of a broader workforce by building an addiction data science short course and scalable educational content with a focus on addiction data analytics through a comprehensive lens. The overall objective of this proposal is to provide a curated research framework and resources to support emerging investigators with varied data science addiction approaches. Our central hypothesis involves developing an addiction data science training program that will expand the research capacity of trainees and emerging investigators. We will achieve the goals of this proposal through the following aims: Aim 1: Develop and refine immersive, tailored addiction data science skills course that provides hands-on demonstrations, tutorials, and presentations on FAIR (Findable, Accessible, Interoperable, Reusable) data principles, computational analytical methods (AI and ML), systems modeling, natural language processing, and analysis and linking of addiction big data. Aim 2: Incorporate novel methods of program evaluation and dissemination using natural language processing to track participant outcomes and career pathways. At the successful completion of the proposed research, the expected outcome is a scalable and widely disseminated education intervention for addiction data science with enduring content to support emerging researchers removing many of the barriers to traditional pathways (e.g., asynchronous conceptual and project-based content that will be evergreen). This innovative research will make curated and refined addiction data science content freely available to educators from undergraduate to postgraduate training levels. This will provide a strong basis for the conceptual foundation needed to begin addiction data science research, without sustained effort from a limited pool of addiction data science expertise. This research aligns with the mission of NIH NIDA to train a broad workforce to perform addiction data science research at the highest quality levels.

Grant Number: 5R25DA061740-02
NIH Institute/Center: NIH
Principal Investigator: Amber Brooks