A Hierarchical Multilevel Inference Framework for Structural Cardiovascular Risk Modeling: County-Scale Analysis of Cardiovascular Mortality in Ohio and Pennsylvania (1999-2020)

2026-07-07

AI Summary

Traditional epidemiological approaches struggle to disentangle hierarchical variation and structural disparities in cardiovascular disease (CVD) mortality across multiscale geographic units. This study proposes a reproducible multilevel statistical inference framework that integrates Gaussian, Poisson, and population-offset Poisson models to separate demographic effects from structural risk factors within a county-nested hierarchy. Fixed effects—including year, sex, race, PM2.5, and O₃—along with county-level random intercepts are incorporated, using data from Ohio and Pennsylvania between 1999 and 2020. Findings reveal that Pennsylvania experienced a steeper decline in age-standardized CVD mortality, though several CVD subtypes plateaued or rebounded after 2010. Black populations exhibited significantly elevated risk, and PM2.5 showed stronger associations with ischemic and hypertensive heart diseases. The inclusion of a population offset notably reduced unexplained variance, demonstrating the framework’s utility as a generalizable tool for environmental health assessment and health equity research.

Abstract

Cardiovascular mortality is shaped by interacting demographic, environmental, and structural processes operating across multiple spatial scales. Conventional epidemiologic analyses often rely on aggregate summaries or single-model formulations that obscure hierarchical variation and contextual heterogeneity. We present a reproducible multilevel statistical inference framework integrating Normal (age-adjusted), Poisson (count-based), and population-offset Poisson models to quantify cardiovascular mortality across nested geographic units while separating demographic effects from structural variation. The framework was applied to county-level mortality data from Ohio and Pennsylvania (1999-2020) using MLwiN hierarchical models for seven cardiovascular disease (CVD) subtypes. Fixed effects included year, sex, race, PM2.5, and O3, while county-level random intercepts captured spatial heterogeneity. Complete model equations are provided in the Supplementary Material. The framework reveals complementary perspectives on cardiovascular risk unavailable from a single model. Age-adjusted mortality declined more rapidly in Pennsylvania than Ohio, whereas Poisson models identified post-2010 stagnation or reversal for several CVD subtypes. Black populations experienced elevated mortality risks, males exhibited higher mortality than females, and PM2.5 showed stronger associations with ischemic and hypertensive mortality in Pennsylvania. Population-offset models reduced unexplained variance while preserving county-level structural disparities. Beyond cardiovascular epidemiology, this work introduces a generalizable hierarchical statistical framework for structurally nested health systems. The methodology provides a scalable foundation for disease surveillance, environmental health assessment, health equity research, reproducible statistical analysis, and AI-assisted scientific inference.

Problem

• cardiovascular mortality
• hierarchical variation
• contextual heterogeneity
• multilevel inference
• structural risk

Innovation

• multilevel modeling
• hierarchical inference
• cardiovascular mortality
• structural risk
• population-offset Poisson