# Structured nonparametric methods for mixtures of exposures

> **NIH NIH R01** · DUKE UNIVERSITY · 2020 · $438,776

## Abstract

The parent R01 focuses on developing reliable and interpretable statistical methods for the
assessment of simultaneous health effects of multiple chemicals. This is challenging due to the
statistical curse of dimensionality, to moderate to high correlation in levels of exposure to
different chemicals, and to missing data and limit of detection issues. Current statistical
methods for nonparametric regression fail to adequately address these challenges, and can
produce uninterpretable dose response surfaces and high error rates in detecting interactions.
The parent R01 is developing transformative methods that incorporate mechanistic constraints
on response surfaces, allow for the complications inherent in epidemiology studies of mixtures,
produce interpretable results including for interactions, and borrow information across different
data sources. This R01 has already produced new statistical tools that clearly improve upon the
state-of-the-art, and that can be implemented routinely by epidemiologists using publicly-
available software packages (e.g., Ferrari and Dunson, 2020a,b).
This proposal describes a competitive revision of the parent R01 to provide a transformative
statistical toolbox for epidemiologists studying risk factors for COVID-19 infection,
morbidity and mortality. This toolbox builds on the Bayesian modeling frameworks developed
by the parent R01, while crucially accounting for the types of large spatially and temporally
structured datasets that are now being collected as part of the COVID-19 monitoring effort. A
new class of computational algorithms is proposed for rapid analysis of massive and
complex spatial-temporal data, these algorithms are used to develop statistical tools for
epidemiologists studying COVID-19 including an R package, and the approach is applied to
study interactions between environmental exposures, age, and other comorbidities with
COVID-19 mortality.

## Key facts

- **NIH application ID:** 10156375
- **Project number:** 3R01ES028804-03S1
- **Recipient organization:** DUKE UNIVERSITY
- **Principal Investigator:** David Brian Dunson
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $438,776
- **Award type:** 3
- **Project period:** 2018-03-01 → 2022-02-28

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10156375

## Citation

> US National Institutes of Health, RePORTER application 10156375, Structured nonparametric methods for mixtures of exposures (3R01ES028804-03S1). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10156375. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*