Nature Aging (2026) · TruDiagnostic

Your biological age, from a routine blood draw

EMRAge reads 19 standard clinical values — the kind already on your Quest or LabCorp panel — and estimates how old your body actually is. It out-predicts both chronological age and PhenoAge for all-cause mortality and chronic-disease risk. On PeptideDx it's free with your account and sits right at the top of your report.

Calculate my EMRAge → See the science
Cox proportional-hazards model 19 routine clinical variables Works from any EMR / lab panel
4.53×
All-cause mortality HR
per 1 s.d. (95% CI 4.06–5.06)
31,264
Discovery cohort
Mass General Brigham Biobank
3
Validation cohorts
All of Us · TruDiagnostic · Gen. Scotland
ρ = 0.76
Correlation w/ chronological age
P < 0.001 in testing set

What is EMRAge?

Biological aging reflects cellular and biochemical processes that unfold at different rates in different people. EMRAge (Electronic Medical Record Age) was developed with a Cox proportional-hazards model fit on 19 routine clinical variables — data available in virtually any electronic medical record or standard lab panel.

Unlike epigenetic clocks that require specialized molecular assays, EMRAge is broadly reproducible across lab systems and shows strong associations with incident and prevalent chronic disease, out-performing PhenoAge in all-cause mortality prediction. It also serves as the foundation for DNAmEMRAge (a DNA-methylation surrogate; ICC = 0.995) and OMICmAge (a multi-omic biomarker integrating proteomics, metabolomics, and clinical data via epigenetic biomarker proxies).

How it stacks up

Hazard ratios per 1 standard-deviation change, adjusted for age, sex, race, BMI, smoking, and alcohol. Higher = stronger prediction of the outcome.

OutcomeEMRAgePhenoAgeChronological age
All-cause mortality4.533.952.61
Stroke2.001.901.72
Type 2 diabetes2.052.261.22
COPD2.212.141.59
Cancer2.221.731.62

Independently validated

10,769
All of Us Research Program
Primary validation
14,213
TruDiagnostic Biobank
DNAmEMRAge / OMICmAge
18,672
Generation Scotland
DNAmEMRAge / OMICmAge

The science

Development & model architecture
EMRAge was developed from 60,370 individuals in the Mass General Brigham (MGB) Biobank, filtered to 31,264 participants with complete data on 19 clinical variables. The cohort was split 70:30 into training and testing sets. A Cox proportional-hazards model estimated the weightings of the selected variables, and — in a manner analogous to the GrimAge approach — the linear combination of weights and predictor values was converted into an "age" metric. The Pearson correlation between EMRAge and chronological age was ρ = 0.76 (P < 0.001) in the testing set and 0.75 (P < 0.001) in training.
Robustness & temporal validation
The EMRAge predictors were validated by retraining the algorithm at four time points in 2-year increments (1 Jan 2008, 2010, 2012, 2014) and applying the derived equations to participants (N = 11,673) on 1 Jan 2016. The Pearson correlations among these estimates were essentially perfect, affirming robustness. EMRAge showed the strongest association with all-cause mortality in prospective analysis (HR = 2.40, P = 9.08 × 10⁻²²) versus PhenoAge (HR = 1.67) and chronological age (HR = 2.00) in the All of Us cohort.
From EMRAge to OMICmAge
EMRAge serves as the training target for two advanced biomarkers. DNAmEMRAge uses 1,097 CpG sites selected via elastic-net regression (α = 0.1) to predict EMRAge from DNA-methylation data (R² = 0.82 training, 0.83 testing; ICC = 0.995). OMICmAge integrates proteomic, metabolomic, and clinical data through epigenetic biomarker proxies (EBPs) — the final model retained 990 CpGs and 40 EBPs (16 protein, 14 metabolite, 10 clinical), achieving R² = 0.83 / 0.84 and ICC = 0.998. Both are available through the TruDiagnostic platform.
Lifestyle factor associations
Across the MGB-ABC and TruDiagnostic Biobank cohorts, OMICmAge showed significant negative associations with female sex, education level, and weekly exercise, and positive associations with obesity and tobacco smoking. Antioxidant intake (FDR = 8.4 × 10⁻⁴) and fish-oil / omega supplementation (FDR = 5.25 × 10⁻⁶) were significantly associated with lower biological age — highlighting the modifiability of EMRAge-based aging through lifestyle and, potentially, targeted interventions.
Citation & further information
Chen Q, Dwaraka VB, Carreras-Gallo N, et al. "OMICmAge quantifies biological age by integrating multi-omics with electronic medical records." Nature Aging (2026). DOI: 10.1038/s43587-026-01073-7.

See your biological age

Create a free PeptideDx account, enter your blood work and a few details about you, and EMRAge appears at the top of your report — alongside your personalized peptide guidance.

Start my report →
Needs 12 routine blood values (RDW, platelets, neutrophils, eosinophils, hematocrit, ALP, ALT, AST, albumin, BUN, glucose, triglycerides) plus your age, sex, BMI, blood pressure & comorbidities.
For research and educational use only. EMRAge is a research tool derived from population-level data. It is not a clinical diagnostic instrument and should not replace clinical judgment. Model results are based on the Cox proportional-hazards model published in Chen et al., Nature Aging (2026), and should be interpreted in the context of a complete clinical evaluation. PeptideDx computes EMRAge locally from the values you provide; results are estimates, not medical advice.