EllaDx
Sign in
Science

Most of modern medicine
was never studied
on women.

Until 1993, women were legally excluded from most U.S. clinical trials. The reference ranges your doctor uses today — for thyroid, for iron, for hormones — were largely built from data on men.

We're rebuilding the foundation. Female bodies, female cycles, female life stages — as the default, not the edge case.

01 — The gap

The numbers that made
us start this company.

01 / 04
23
%
Of cardiovascular trial participants are women
Despite heart disease being the leading cause of death for American women.
SourceJAMA Cardiology, 2022
02 / 04
9
yrs
Average delay to an endometriosis diagnosis
One in ten women will have it. Few will be diagnosed in time.
SourceAmerican Journal of Obstetrics & Gynecology, 2023
03 / 04
70
%
Of chronic pain patients are women
Yet 80% of pain research has been conducted on male subjects.
SourceNature Reviews Neuroscience, 2021
04 / 04
1 IN 3
Women over 40 have a thyroid condition
Most will go undiagnosed because standard TSH ranges miss them.
SourceJournal of Clinical Endocrinology, 2024
02 — Reference ranges

"In range"
for whom, exactly?

A reference range is a statistical statement: 95% of a population falls between these numbers. But which population matters enormously. We rebuild ranges stratified by sex, cycle phase, and life stage — from our own proprietary dataset of 40,000+ women.

TSH
mIU/L

Standard range was set on a mixed population. Optimal thyroid function in women — particularly those trying to conceive — sits much lower.

Standard lab range
Legacy — mixed-sex population
0.44.5
2.5
0.44.5
EllaDx female-specific
Phase/stage-aware reference
0.52.5
1.5
0.44.5
Ferritin
ng/mL

The textbook lower bound of 10 ng/mL is based on 'not anemic.' Women with ferritin below 50 consistently report fatigue, hair loss, and restless legs.

Standard lab range
Legacy — mixed-sex population
10200
50
10200
EllaDx female-specific
Phase/stage-aware reference
50150
90
10200
Vitamin D
nmol/L

Levels associated with bone health in post-menopausal women are meaningfully higher than the legacy sufficiency floor.

Standard lab range
Legacy — mixed-sex population
30125
60
30150
EllaDx female-specific
Phase/stage-aware reference
75150
100
30150
Estradiol (luteal)
pg/mL

Phase-specific ranges — the only way to actually read a cycling woman's labs. Generic 'female' ranges collapse three different hormonal states into one.

Standard lab range
Legacy — mixed-sex population
30400
100
30400
EllaDx female-specific
Phase/stage-aware reference
80250
150
30400
03 — Validation

How a biomarker
earns its place in a panel.

Six steps, from clinical question to quarterly revalidation. Every marker in every panel has been through all of them.

01
Clinical question first
Symptom → hypothesis

Every panel starts with a specific symptom pattern women report. We don't build panels from a price list of available assays — we build them from the questions a 34-year-old with fatigue actually needs answered.

02
Literature & meta-analysis
Evidence grading A–D

Our medical team reviews the published evidence. We look for assays with peer-reviewed support for female-specific clinical utility — and flag the ones where the data is still thin.

03
Assay selection & validation
CV < 8% across labs

We choose assays available in CLIA/CAP-accredited U.S. labs. For sensitive markers (estradiol, testosterone) we cross-validate across two independent labs to confirm agreement.

04
Female reference dataset
n = 40,000+ and growing

Every biomarker gets rebuilt against our proprietary dataset of 40,000+ U.S. women — stratified by age, cycle phase, and life stage. Ranges are updated quarterly as the dataset grows.

05
Physician panel sign-off
3-physician consensus

Before a panel goes live, three board-certified physicians sign off on its design: which biomarkers are included, which cycle day is used, and which reference ranges trigger a flag on the report.

06
Continuous revalidation
Outcome audit every 6 mo

Every six months we audit flagged results against downstream outcomes. Ranges that don't predict anything get retired. Markers that over-flag get retuned.

04 — The evidence

Every claim, sourced.

Papers that shaped how we think about women's diagnostics — and the ones we keep returning to when building new panels.

01
Vogel B, Acevedo M, Appelman Y, et al. (2021)
The Lancet women and cardiovascular disease Commission: reducing the global burden by 2030
The Lancet, 397(10292): 2385–2438
Cardiovascular
02
Westergaard D, Moseley P, Sørup FKH, et al. (2019)
Population-wide analysis of differences in disease progression patterns in men and women
Nature Communications, 10: 666
Diagnostic delay
03
Mogil JS. (2020)
Qualitative sex differences in pain processing: emerging evidence of a biased literature
Nature Reviews Neuroscience, 21: 353–365
Pain research
04
Razvi S, Bhana S, Mrabeti S. (2019)
Challenges in interpreting thyroid stimulating hormone results in the diagnosis of thyroid dysfunction
Journal of Thyroid Research, 4106816
Reference ranges
05
Zondervan KT, Becker CM, Missmer SA. (2020)
Endometriosis
New England Journal of Medicine, 382: 1244–1256
Diagnostic delay
06
Mirin AA. (2021)
Gender disparity in the funding of diseases by the U.S. National Institutes of Health
Journal of Women's Health, 30(7): 956–963
Research funding
07
Soldin OP, Mattison DR. (2009)
Sex differences in pharmacokinetics and pharmacodynamics
Clinical Pharmacokinetics, 48(3): 143–157
Sex differences
08
Merz AA, Cheng S. (2016)
Sex differences in cardiovascular ageing
Heart, 102(11): 825–831
Cardiovascular
09
Daitch V, Turjeman A, Poran I, et al. (2022)
Underrepresentation of women in randomized controlled trials: a systematic review
Trials, 23: 1038
Clinical trials

Full bibliography of 240+ papers available at elladx.com/research · Last updated Q1 2026

Be counted

The dataset gets better
every time you test.

Every woman who tests with us — anonymously, with consent — helps rebuild reference ranges for the next woman who does. This is how the gap closes.