Estimating substance use disparities across intersectional social positions using machine learning: An application of group-lasso interaction network.

Journal: Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors
PMID:

Abstract

OBJECTIVE: An aim of quantitative intersectional research is to model the joint impact of multiple social positions on health risk behaviors. Although moderated multiple regression is frequently used to pursue intersectional research hypotheses, such parametric approaches may produce unreliable effect estimates due to data sparsity and high dimensionality. Machine learning provides viable alternatives, offering greater flexibility in evaluating many candidate interactions amid sparse data conditions, yet remains rarely employed. This study introduces group-lasso interaction network (glinternet), a novel machine learning approach involving hierarchical regularization, to assess intersectional differences in substance use prevalence.

Authors

  • Connor J McCabe
    Department of Psychiatry, University of Washington.
  • Jonathan L Helm
    Department of Psychology, San Diego State University.
  • Max A Halvorson
    Department of Psychiatry, University of Washington.
  • Kieran J Blaikie
    Department of Epidemiology, University of Washington.
  • Christine M Lee
    Department of Psychiatry, University of Washington.
  • Isaac C Rhew
    Department of Psychiatry, University of Washington.