An interpretable machine learning model of cross-sectional U.S. county-level obesity prevalence using explainable artificial intelligence - read online free

Aug 3, 2026

Upload by Noah

Geographic heterogeneity in U.S. county obesity prevalence motivates interpretable modeling rather than opaque prediction. This study applies explainable artificial intelligence methods to cross-sectional obesity data from 3,142 counties, using features spanning health outcomes, health behaviors, clinical care, social and economic factors, physical environment, demographics, and severe housing conditions. Random forest prediction is accompanied by feature importance, accumulated local effects, a global surrogate decision tree, and local interpretable model-agnostic explanations. The model explains 79% of variance, highlighting physical inactivity, diabetes, and smoking prevalence as key predictors, enabling actionable public health insight.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Background

Objective

Methods

Results

Conclusions

Introduction

Geographic heterogeneity and the obesity crisis

Explainable artificial intelligence and study aims

Methods

Data sources

PLOS ONE OPEN ACCESS Citation: Allen B (2023) An interpretable machine learning model of cross-sectional U.S. county-level obesity prevalence using explainable artificial intelligence. PLoS ONE 18(10): e0292341 . [https://](https://)[ ](https://)[doi.org/10.1371/journal.pone.0292341](doi.org/10.1371/journal.pone.0292341) [Editor:](Editor: Mujeeb Ur Rehman)[ Mujeeb Ur Rehman](Editor: Mujeeb Ur Rehman), [York St John](York St John) University, UNITED KINGDOM Received: May 2, 2023 Accepted: September 18, 2023 Published: October 5, 2023 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: [https://doi.org/10.1371/journal.pone.0292341](https://doi.org/10.1371/journal.pone.0292341) [Copyright:](Copyright:) © [2023](2023) Ben Allen. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data for this study are publicly available from the OSF repository ([https://osf.io/xtfjk/](https://osf.io/xtfjk/)) . Funding: The author(s) received no specific funding for this work. RESEARCH ARTICLE An interpretable machine learning model of cross-sectional U.S. county-level obesity prevalence using explainable artificial intelligence Ben Allen * Department of Psychology, University of Kansas, Lawrence, Kansas, United States of America * [benallen@ku.edu](benallen@ku.edu) Abstract Background There is considerable geographic heterogeneity in obesity prevalence across counties in the United States. Machine learning algorithms accurately predict geographic variation in obesity prevalence, but the models are often uninterpretable and viewed as a black-box. Objective The goal of this study is to extract knowledge from machine learning models for county-level variation in obesity prevalence. Methods This study shows the application of explainable artificial intelligence methods to machine learning models of cross-sectional obesity prevalence data collected from 3,142 counties in the United States. County-level features from 7 broad categories: health outcomes, health behaviors, clinical care, social and economic factors, physical environment, demographics, and severe housing conditions. Explainable methods applied to random forest prediction models include feature importance, accumulated local effects, global surrogate decision tree, and local interpretable model-agnostic explanations. Results The results show that machine learning models explained 79% of the variance in obesity prevalence, with physical inactivity, diabetes, and smoking prevalence being the most important factors in predicting obesity prevalence. Conclusions Interpretable machine learning models of health behaviors and outcomes provide substantial insight into obesity prevalence variation across counties in the United States. Competing interests: The authors have declared that no competing interests exist. 1. Introduction Identifying the principal factors that impact health is an important theme in obesity research [ 1–3] . Multiple health behaviors and environmental conditions contribute to the obesity crisis [4] . There is also substantial geographic heterogeneity in the prevalence of obesity across the United States [5–7] . Machine learning may be the most powerful approach to modeling variation in obesity prevalence across the United States, but machine learning models are often opaque and difficult to interpret [8] . To open the black box of machine learning models, the field of explainable artificial intelligence has emerged with the goal of extracting domain knowledge about th

FAQs

What is the main objective of the study?

To extract domain knowledge from machine learning models explaining cross-sectional variation in county-level obesity prevalence in the United States.

Which data and counties are analyzed?

The study uses cross-sectional obesity data from 3,142 counties, based on the 2022 County Health Rankings dataset.

How is the model made interpretable?

Explainable AI methods are applied to random forest predictions using feature importance, accumulated local effects, a global surrogate decision tree, and local interpretable model-agnostic explanations.

Aug 3, 2026

Upload by Noah