Integrating Data Science with Social and Behavioral Sciences Training to Tackle Complex Public Health Issues: An Example of an Interdisciplinary Course Sequence for Preparing the Next Generation of Researchers

Abstract

Despite recent advances in data science analytics, there are notable gaps in interdisciplinary training opportunities to prepare public health graduate students to effectively leverage modern data capacities to study and address pressing public health problems. We describe the development, implementation, and evaluation of two innovative courses on the data science and social behavioral science skills needed to address drug-related harms. A multidisciplinary team of instructors developed modules within the two courses on how to apply specific data science methods (e.g., administrative data analysis, geospatial methods, and machine learning) to study drug-related harms. The instructors integrated cross-cutting themes of ethics, internal and external validity, sampling and measurement, and health equity. The Intersectional Risk Environment Model provided theoretical underpinning to all modules. We share our model and lessons learned to encourage other training programs to consider how interdisciplinary training can help prepare trainees to thrive in rapidly changing public health and data landscapes.