Robust Clustering with Subpopulation-specific Deviations

TL;DR

This paper introduces Robust Profile Clustering, a new method for dietary pattern analysis that accounts for regional variability in large, heterogeneous populations, improving interpretability over traditional clustering methods.

Contribution

The paper proposes a novel two-level clustering approach combining global and local models to better capture regional dietary differences in large datasets.

Findings

Identified distinct dietary patterns across regions.

Improved interpretability of dietary clusters.

Effectively handled heterogeneity in large populations.

Abstract

The National Birth Defects Prevention Study (NBDPS) is a case-control study of birth defects conducted across 10 U.S. states. Researchers are interested in characterizing the etiologic role of maternal diet, collected using a food frequency questionnaire. Because diet is multi-dimensional, dimension reduction methods such as cluster analysis are often used to summarize dietary patterns. In a large, heterogeneous population, traditional clustering methods, such as latent class analysis, used to estimate dietary patterns can produce a large number of clusters due to a variety of factors, including study size and regional diversity. These factors result in a loss of interpretability of patterns that may differ due to minor consumption changes. Based on adaptation of the local partition process, we propose a new method, Robust Profile Clustering, to handle these data complexities. Here, participants may be clustered at two levels: (1) globally, where women are assigned to an overall population-level cluster via an overfitted finite mixture model, and (2) locally, where regional variations in diet are accommodated via a beta-Bernoulli process dependent on subpopulation differences. We use our method to analyze the NBDPS data, deriving pre-pregnancy dietary patterns for women in the NBDPS while accounting for regional variability.