(LB04-06-26) Elucidating the Effects of Unprocessed Red Meat on Health Through Longitudinal ‘Omics Data

Tuesday, July 28, 2026

11:00 AM - 11:12 AM ET

Location: Maryland Ballroom D

Presenting Author(s)

Matthew Darmadi, B.S. (he/him/his)

Medical Student
Baylor College of Medicine
Houston, Texas, United States

Disclosure(s): Beef Checkoff: Grant/Research Support (Ongoing)
National Institutes of Health (NIH): Grant/Research Support (Ongoing)
USDA-ARS: Grant/Research Support (Ongoing)

Co-Author(s)

AW

Alexis C. Wood, PhD, FAHA (she/her/hers)

Associate Professor
Children's Nutrition Research Center, Baylor College of Medicine
Houston, Texas, United States

Co-Author(s)

ZC

Zsu Zsu Chen

Beth Israel Deaconess Medical Center, United States
CC

Clary B. Clish, PhD

Broad Institute of MIT and Harvard
Cambridge, Massachusetts, United States
RG

Robert E. Gerszten

Beth Israel Deaconess Medical Center, Harvard Medical School
MG

Mark O. Goodarzi

Cedars-Sinai Medical Center
Los Angeles, California, United States
MM

Michael Y. Mi

Beth Israel Deaconess Medical Center, United States
KT

Kent D. Taylor

The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, United States
JR

Jerome I. Rotter

The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center

Disclosure(s):

Matthew Darmadi, B.S.: Beef Checkoff: Grant/Research Support (Ongoing)
National Institutes of Health (NIH): Grant/Research Support (Ongoing)
USDA-ARS: Grant/Research Support (Ongoing)

Alexis C. Wood, PhD, FAHA: No relevant financial relationship(s) with ineligible companies to disclose.

Objectives: Associations of unprocessed red meat intake with poorer health contrast with those from feeding trials and are puzzling given meat’s nutritional profile. We aimed to examine whether plasma metabolites associated with unprocessed red meat intake have different associations with health when they represent “within-person” effects - representing covariation between red meat changes and the metabolome within a person - and “contextual” effects - the population-level association after accounting for within-person covariation.

Methods: We analyzed self-reported dietary intake, untargeted metabolomic compounds, and fasting insulin (FI) and glucose (FG) from the Multi-Ethnic Study of Atherosclerosis (N=4,567). Penalized regression within a multilevel framework identified a parsimonious subset of metabolites that independently predicted unprocessed red meat intake when modeled jointly. Person-mean centering decomposed associations for all metabolites with non-zero coefficients into within-person and contextual effects. Metabolite scores were created as weighted averages of all metabolites reaching P< .05 after an FDR correction for each of within-person and context effects. Associations between each score with FI and FG were estimated. All analyses were adjusted for socio-demographics, kidney function, and BMI.

Results: 47 metabolites were selected by the penalized LASSO model as a “core” signature of red meat; 5 showed significant within-person effects, and 33 showed significant context effects. Red meat metabolite scores constructed from within-person effects associated with lower FG (β=-0.12 ± 0.02, P=6.7*10^-11) and FI (β=-0.15 ± 0.02, P=5.4*10^-12). Red meat metabolite scores constructed from contextual effects associated with higher FG (β=0.24 ± 0.02, P=6.8*10^-5) and FI (β=0.28 ± 0.02, P=9.0*10^-31).

Conclusions: Metabolites that covary with a person’s changes in red meat intake are associated with better glucose homeostasis, unlike those reflecting contextual effects. As contextual effects can include sociodemographic, lifestyle and genetic factors, these findings align with prior work suggesting nutritional epidemiology can be biased by confounding and highlight the importance of distinguishing true diet-health covariation.

Funding Sources: Beef Checkoff; National Institutes of Health