Supervised Robust Profile Clustering

• Activity Number: 537 - SPEED: Infectious Disease, Environmental Epidemiology, and Diet
• Type: Contributed
• Date/Time: Wednesday, August 1, 2018 : 10:30 AM to 11:15 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Briana Stephenson* and Amy H Herring and Andrew Olshan
• Companies: University of North Carolina at Chapel Hill and Duke University Statistical Science and University of North Carolina at Chapel Hill
• Keywords: food frequency questionnaire; robust profile clustering ; clustering; orofacial clefts; maternal diet

Abstract

Maternal diet has been shown to be associated with perinatal outcomes. Robust Profile Clustering (RPC) provides a dual flexible clustering model, where participants may cluster at two levels: (1) globally, where participants are clustered according to behaviors shared across an overall population, and (2) locally, where individual behaviors can deviate and cluster according to their identified subpopulation. We link this clustering model generated from the RPC with a health outcome using a predictive clustering model known as a supervised RPC joint model. We generate the utility of this model for the application of deriving dietary patterns in the United States. Using dietary consumption data from the 1997-2009 National Birth Defects Prevention Study, a population-based case-control study, we determine how maternal diet profiles are associated with an orofacial cleft among their offspring.