What Are Endocrine Disruptors? The Evidence-Based Guide

They don't poison you. They impersonate the messengers your body uses to run itself. A synthetic molecule shaped just enough like estrogen fits the estrogen receptor. One shaped just enough like testosterone blocks the androgen receptor. The system reads the signal as real, because the system can't check credentials — it was never designed to. The receptor doesn't ask for ID.

Most of these chemicals were declared safe after testing in a single adult animal, in isolation, at doses far higher than anything circulating in placental blood. Conventional toxicology was built on a five-hundred-year-old principle: the dose makes the poison. Endocrine disruptors break every assumption that principle depends on.

The WHOWorld Health Organization estimates that close to 800 chemicals are known or suspected of interfering with hormone receptors, synthesis, or conversion — and only a small fraction have been tested in assays capable of detecting endocrine effects at all WHO/UNEP 2012. Our guide to evaluating chemical safety covers the framework for any product. The chemical class that framework was built to address is the one the testing model keeps getting wrong — by factors of twenty thousand, not twenty percent.

Your endocrine system is your body's long-distance messaging service — and it runs at concentrations so low that measuring them requires mass spectrometry. Circulating estradiol in a premenopausal woman sits between roughly 30–350 pg/mL the concentration range the body uses to regulate reproduction — parts per trillion. That is the scale endocrine disruptors exploit. You already know intuitively that tiny things can have outsized effects: a single text changes how you feel about someone, a single word in a doctor's sentence changes everything. Hormones are the body's version of that — but the messages are molecules, and the concentrations are measured in trillionths of a gram.

The nervous system works fast — electrical signals in milliseconds. The endocrine system works slowly and persistently. Glands release chemical messengers into the bloodstream; those messengers travel to target cells anywhere in the body, bind to specific receptors, and trigger responses that can last hours, days, or permanently alter development. Estrogen drives reproductive cycling and bone density. Testosterone directs sexual differentiation and sperm production. Thyroid hormones T3 and T4 set the metabolic rate and wire the fetal brain. Cortisol calibrates the stress response. Melatonin governs the sleep cycle. Each messenger fits a specific receptor — a protein with a binding pocket shaped to recognise exactly that molecule. The right key, the right lock, the right signal fires. Shape is everything.

What makes this system uniquely vulnerable is how precisely it operates. Your body is responding to amounts so small it strains belief — and that sensitivity is exactly what endocrine disruptors exploit. Any synthetic chemical shaped closely enough to fit a receptor's binding pocket can activate or block the same signalling cascade, at the same concentrations your own hormones use. The precision that makes the system work is the same precision that makes it hackable.

An endocrine disruptor is a chemical that interferes with your hormonal signalling — 800 suspected EDCs on the WHO list chemicals are known or suspected of doing so. The formal WHO definition: 'an exogenous substance or mixture that alters the function(s) of the endocrine system and consequently causes adverse health effects.' That's deliberately broad, because the mechanisms are varied. Most real-world EDCsEndocrine Disrupting Chemicals — the shorthand researchers use for this entire class of hormone-interfering compounds. combine more than one. Eight hundred on the list. Routine testing catches maybe a dozen.

Mimicry. The chemical is shaped enough like a natural hormone to fit the receptor's binding pocket and activate it — firing the same signalling cascade as if the real hormone had arrived. BPABisphenol A — a synthetic chemical used in polycarbonate plastics, can linings, and thermal receipts. (C₁₅H₁₆O₂) binds estrogen receptor beta with affinity thousands of times weaker than estradiol Kuiper et al. 1998. That sounds reassuringly weak — until you learn that BPA also activates a completely different estrogen receptor, GPR30 (a membrane receptor characterised years later), at picomolar concentrations that are below the detection threshold for the nuclear pathway regulatory assays test. The membrane pathway, operating at even lower doses, goes unmeasured by the tests regulators run.

Parabens, used as preservatives in cosmetics, fit the same estrogen receptors less tightly than BPA — but they bind. Darbre et al. 2004 detected intact parabens in human breast tissue, showing that dermal absorption delivers them into the body without full metabolic breakdown. In each case, the receptor can't tell the impostor from the real thing. It activates.

Blocking. The chemical occupies the receptor without activating it — a key that fits the lock but won't turn. The real hormone is circulating at normal levels, but it can't reach its target because something else is sitting in the binding pocket. Vinclozolin, a fungicide used on grapes, produces a metabolite only two-fold weaker than hydroxyflutamide — a pharmaceutical designed to block testosterone Wong et al. 1995. DDE, the persistent metabolite of DDT, blocks androgen receptors almost exclusively — it has nearly no ability to bind estrogen receptors, which means its reproductive effects come entirely through anti-androgen action Kelce et al. 1995. The chemical looks nothing like estrogen. It interferes with testosterone. That's the part the original tests missed.

Interference with synthesis, transport, or metabolism. Some EDCs don't touch receptors at all — they cut off the supply chain upstream. PerchlorateFound in rocket fuel, fireworks, and explosives. Contaminates drinking water and some produce. Disrupts the thyroid by blocking iodide uptake — the same iodide needed to make thyroid hormones. blocks the sodium-iodide symporter in the thyroid, suppressing hormone production before it starts. PCBsPolychlorinated biphenyls — industrial chemicals banned since the 1970s but still persistent in older buildings, the environment, and the food chain. and PBDEsPolybrominated diphenyl ethers — brominated flame retardants used in furniture, electronics, and textiles. Persistent in household dust and human tissue. displace thyroid hormones from their transport proteins in the blood, altering how much reaches target tissues. Atrazine — one of the most widely used herbicides globally — induces aromatase, the enzyme that converts testosterone to estrogen, shifting the hormonal balance without ever binding a receptor Sanderson et al. 2000.

Most real-world EDCs don't fit neatly into one mechanism. BPA mimics estrogen at nuclear receptors AND activates membrane receptors via a different pathway. Triclosan disrupts thyroid hormone transport AND shows estrogenic activity. The Endocrine Society's Second Scientific Statement Gore et al. 2015 — 150 pages, seven health domains — concluded that across obesity, diabetes, reproduction, cancers, thyroid, and neurodevelopment, the accumulated evidence strengthens the case for EDC effects in human health. The mechanisms are multiple. The pattern is consistent.

No single classification system covers all EDCs — the WHO list, EU SVHC list, and individual regulatory assessments apply different criteria. But the chemicals with the strongest human exposure and the most robust evidence fall into recognisable families. The table below covers the main ones you encounter in everyday products — not the complete WHO list of 800 suspects, but the ones the research on actual human exposure has studied most.

This table covers the known family heads. Within each family are dozens or hundreds of variants — the PFAS family alone contains more than 14,700 identified chemicals, of which regulators have set enforceable limits for fewer than 30. The gap between what exists in commerce and what has been evaluated for endocrine activity isn't a rounding error. It's the whole story. See individual profiles for BPA, PFAS, phthalates, and parabens.

Everywhere, and you already have the data for this whether you've looked it up or not. CDCCenters for Disease Control and Prevention — the US public health agency. biomonitoringMeasuring chemicals directly in human blood, urine, or tissue to determine actual exposure — rather than estimating it from environmental levels. The gold standard for knowing what's really circulating in people's bodies. through NHANESNational Health and Nutrition Examination Survey — the CDC's ongoing population health study, which tests blood and urine samples from thousands of Americans for hundreds of chemicals. detected BPA in 92.6% of Americans aged 6 and over Calafat et al. 2008. Phthalate metabolites in 75%+ of Americans tested for the four most common species Silva et al. 2004. PFASPer- and polyfluoroalkyl substances — a class of more than 14,700 identified 'forever chemicals'. Used for stain, water, and grease resistance. Do not biodegrade. in nearly all blood samples tested. These aren't occupational exposures in factory workers. They're the baseline chemical load of ordinary people living ordinary lives. The receipt by the door, the takeaway box, the sunscreen — all of it is in there.

The European Human Biomonitoring Initiative (HBM4EU), spanning 28 EU countries and running from 2017 to 2022, confirmed that EU populations carry measurable levels of phthalates, PFAS, bisphenols, and flame retardants, and concluded that exposures to several chemical classes 'are too high.' The major ones, and how long they stay:

The half-life column matters more than it looks. PFAS stays in your blood for 3.5–4.8 years geometric mean half-life in human serum Olsen et al. 2007 — every day's exposure adds to a burden that barely diminishes. BPA clears in hours, but you're re-exposed daily, so the body holds a steady-state level — you never reach zero. PBDEs accumulate in fat tissue and household dust, persisting for decades; penta-BDE was phased out in 2004 and is still measurable in older furniture and blood.

Phthalates deserve particular attention because they're the hardest to avoid. They hide behind the single word 'fragrance' on ingredient lists — one word representing any of thousands of possible chemicals, most never individually tested for endocrine-disrupting activity. They're in PVC flooring, shower curtains, food packaging, and car interiors. The new car smell? Partly phthalate off-gassing. Exposure comes through three routes simultaneously: dermal from cosmetics, inhalation from off-gassing, and ingestion from food contact materials.

In 2023, EFSA cut BPA's safe dose by a factor of 20,000× EFSA 2023. Not twenty percent. Not double. Twenty thousand. That revision is the clearest evidence that the five-hundred-year-old model of toxicology doesn't fit endocrine disruptors — so here's what the old model was, and exactly where it breaks.

Paracelsus, writing in the 1530s, established that the dose makes the poison: more exposure, more harm, and below some threshold, any substance is safe. The entire regulatory framework for chemical safety descends from that assumption. Test one chemical at a high dose. Find where harm stops. Add a safety margin. Declare it safe. Paracelsus was right about cyanide. He had no idea about estradiol, and it took us five centuries to notice the difference.

Endocrine disruptors broke that model in three ways that regulators are still catching up to.

First: the dose-response isn't linear. At low concentrations, an EDC fits into hormone receptors and activates signalling — it's at the right concentration to be 'heard.' At high concentrations, receptors saturate and desensitise. The result is a non-monotonic dose responseA dose-response curve that doesn't run in a straight line. For endocrine disruptors, effects often peak at low doses and decline at high ones — the 'inverted U' — opposite to what toxicology assumes. — maximum effect at low doses, declining effect at high ones. Vandenberg 2014 reviewed the BPA literature and found that more than 20% of BPA experiments showed inverted U-curves — roughly one in five, varying with how many dose points the experiment used. Standard regulatory testing uses three dose points at high concentrations. Detecting an inverted U-curve requires an average of 6.9 points. The peak biological activity sits in the range they never examined.

Second: the exposure route doesn't match. Regulatory testing assumes oral ingestion — the chemical passes through the gut, through the liver, which conjugates and inactivates much of it before anything reaches the bloodstream. But your moisturiser absorbs through your skin. The receipt transfers chemicals through your hands. Shower steam enters through your lungs. None of those routes pass through the liver first.

Ten volunteers received deuterium-labelled BPA — orally one day, dermally another. From the oral dose, 0.56% of serum BPA was the bioactive unconjugated parent. From the dermal dose, 8.81% of serum BPA was the bioactive unconjugated parent Sasso et al. 2020. Sixteen times more bioactive chemical reaching your tissues from skin contact than from swallowing the same dose. And a receipt, a face cream, and a shampoo all go through the skin. The safety number was calculated for the mouth. The actual exposure is through the hand.

Third: it tests one chemical at a time. Nobody is exposed to one chemical at a time. Your morning routine involves dozens of chemicals through three simultaneous routes. The regulatory model evaluates each in isolation. Your body processes them all at once. The consequences of that mismatch are covered in the next section.

The EFSA 2023 BPA revision confronted the first problem directly. They re-evaluated the full BPA evidence — focusing on immune effects, where low concentrations shifted Th17 immune cell ratios toward allergic inflammation — and cut the tolerable daily intake from 4 micrograms per kilogram per day to 0.2 nanogramsOne billionth of a gram — a thousand times smaller than a microgram. per kilogram. A factor of 20,000×. The previous number had stood for over a decade. At the new TDI, dietary exposure alone exceeds the safe dose for every age group in the population.

Mothers in the highest quartile of prenatal monobutyl phthalate had sons with 10.2× the odds of reduced anogenital distance versus mothers in the lowest quartile (Swan et al. 2005, n=134). The exposures weren't extreme — ordinary consumer ranges, the kind shared by millions of women. The same phthalate at the same dose in an adult male: no detectable effect. That single finding captures the timing problem. If endocrine disruptors worked like conventional poisons, the dose that harms an adult would harm a fetus proportionally. But EDCs work through signalling rather than damage — and when the signal arrives matters as much as how loud it is.

During fetal development, hormones aren't maintaining a stable system — they're building one from scratch. Estrogen and testosterone direct the formation of reproductive organs, brain structures, immune architecture, and metabolic programming. The signals that determine whether tissue becomes ovarian or testicular operate at parts-per-trillion concentrations. A synthetic chemical at those same concentrations doesn't just interfere — it can redirect developmental architecture in ways that become permanent. The fetus has no 'adult dose' to compare against. It's calibrating on whatever is circulating.

Researchers tracked prenatal phthalate metabolites in the urine of 134 pregnant women, then examined the boys after birth. Mothers in the highest quartile of monobutyl phthalate — not factory workers, the ordinary products-and-packaging range — had sons with significantly reduced anogenital distanceThe distance between the anus and the genitals. In male infants, this is a developmental marker of how much testosterone signalling occurred in the womb — shorter distance indicates insufficient masculinisation during fetal development., a marker of androgen insufficiency during masculinisation Swan et al. 2005. The fetus responded to what the adult couldn't detect.

The principle held in actual human blood. Braun et al. (2024) extracted the real-world chemical mixture — 294 chemicals — from the blood of 624 pregnant women in the German LiNA cohort and applied it to developing nerve cells. At concentrations where individual chemicals showed no effect, the mixture damaged neural development Braun et al. 2024. Not a laboratory simulation: the actual chemical cocktail the mothers were carrying on an ordinary day.

The critical windows don't end at birth. Puberty is another active hormonal transition, and EDC exposure during it has been linked to early sexual maturation and altered growth trajectories. The population-level signal is already visible: sperm concentrations have fallen 51.6% globally since 1973, from a meta-analysis of 223 studies Levine et al. 2023. The rate of decline has doubled since 2000. Something in the modern environment is exerting selection pressure on human reproductive biology, and endocrine disruptors are the leading hypothesis for what. See our guide to chemicals and fertility for the full evidence.

Think about this morning. Coffee from a paper cup with a plastic lid. Moisturiser from a pump bottle. Toothpaste. The sofa you sat on. The food packaging from last night still in the bin. None of those products was tested in combination. None of the chemicals they contributed to your body burden was evaluated alongside the chemicals from the others. The mixture in your blood right now was never formulated, never approved, never reviewed by any regulatory body. It accumulated, one daily routine at a time.

In 2002, a team at the University of London proved this mattered. They took eight xenoestrogensSynthetic chemicals that mimic estrogen by binding to estrogen receptors — 'xeno' means foreign. Found in plastics, preservatives, UV filters, and pesticides. — a bisphenol from can linings, a hydroxylated PCB from old electrical equipment, a UV filter from sunscreen, an alkylphenol from detergent residues, a couple of chemicals leaching from common plastics. Nothing exotic; each chemical is in products people use daily. For each one, they established the no-effect dose in isolation. Then they mixed all eight at those individually inert concentrations. The mixture produced a substantial estrogenic response Silva et al. 2002.

The scale of the problem is visible in the data. The CDC's Fourth National Report on Human Exposure measures more than 350 chemicals in the blood and urine of the US population. There is no unexposed control group left in industrialised nations. Every study comparing 'exposed' to 'unexposed' populations is comparing 'more exposed' to 'less exposed.'

The EU-wide health burden attributable to EDC exposure runs to an estimated €157 billion/year 1.23% of EU GDP — calculated chemical by chemical, without accounting for mixture effects Trasande et al. 2015. The true figure is higher. We just can't calculate how much higher yet.

Regulation has moved — slowly, partially, and one chemical at a time, which is precisely the approach the cocktail-effect research says is insufficient. The movement is real. The table below traces the major actions since the science established the problem.

The pattern in that table is clear: each action targets one chemical or one family, after years of evidence accumulation. The regulatory system is not designed to evaluate a mixture of 294 chemicals simultaneously — the same mixture Braun et al. (2024) extracted from pregnant women's blood and showed was neurotoxic at real-world concentrations. The gap isn't negligence. It's architecture: a system built for one chemical at a time will always be decades behind a problem that is, by definition, about mixtures.

The EFSA 2023 BPA revision is the clearest illustration of what happens when the testing model finally catches up. The previous TDI of 4 µg/kg/day had stood since 2006. The revision, which examined immune cell shifts at low doses rather than the high-dose endpoints regulators had tested for decades, landed at 0.2 ng/kg/day — a factor of 20,000. At the new TDI, dietary exposure alone exceeds the safe dose for every age group. The old number was wrong for seventeen years. Nobody was lying. The testing model missed what was happening at the low end.

You already know the instinct: check what's in things, avoid plastic where you can, buy products with shorter ingredient lists. What you might not know is how fast it works. A three-day switch to fresh food dropped urinary BPA by 66% (Rudel et al. 2011). Three days. The body burden isn't permanent — it's maintained by daily re-exposure, and it drops when the exposure stops. Your body burden drops as soon as you stop topping it up.

Rudel et al. 2011 ran a straightforward experiment: five families ate their normal diet for three days, switched to fresh food — nothing from cans or plastic packaging — for three days, then went back to normal. During the fresh-food phase, urinary BPA dropped 66% in 3 days and DEHP metabolites dropped 53–56% in 3 days. When they returned to their normal diet, levels rebounded within days. The chemical load is maintained by daily re-exposure, and it drops when the exposure stops.

Harley et al. 2016 ran the same principle with personal care products. In the HERMOSA study, 100 Latina adolescent girls switched to low-chemical cosmetics for three days. Urinary methyl paraben dropped 44% in 3 days. Propyl paraben dropped 45% in 3 days. Triclosan dropped 36% in 3 days. MEPMonoethyl phthalate — the urinary metabolite of diethyl phthalate, the phthalate most commonly used as a fragrance carrier in personal care products. dropped 27% in 3 days. The products you apply every morning are a significant, measurable, and modifiable source of endocrine disruptor exposure.

The practical hierarchy follows the science: reduce what touches your body longest, through the most absorptive route. For short-lived chemicals — BPA, phthalates, parabens — switching products produces rapid results because the body clears them quickly once re-exposure stops. For persistent chemicals — PFAS, PBDEs — the focus is preventing further accumulation, since existing body burden diminishes slowly over years.

One trap worth flagging: 'BPA-free' doesn't mean safe. Rochester and Bolden's 2015 systematic review of 32 studies found BPS and BPF show comparable hormonal activity to BPA Rochester and Bolden 2015, and Yang et al. (2011) found that almost all commercial plastics — including BPA-free ones — released chemicals with measurable estrogenic activity Yang et al. 2011. Glass and stainless steel avoid the question entirely.

You don't need to replace everything at once. The intervention studies show that switching even a few products — the moisturiser that sits on your skin, the containers that hold your food — produces measurable reductions in days, not months. Start with what touches you longest. Work outward from there. The Eso-Friendly approach applies that lens to any product, including ones we don't sell.

The intervention studies give you a priority order: start with what sits on your skin the longest and what goes in your mouth most frequently. Those two categories — personal care and food contact — account for the majority of the measurable reduction in the Rudel and Harley studies.

For oral care: a boar bristle toothbrush removes nylon bristles held by epoxy resin adhesive. A copper tongue scraper removes the chlorhexidine exposure from plastic alternatives. The oral care chemicals guide covers every product in the routine in detail.

For water: activated charcoal filter sticks reduce chlorine and some organic compounds. A water distiller removes the broadest spectrum — including heavy metals and some PFAS. The tap water guide compares all filtration methods against what they actually remove.

For sleep: a silk sleep mask replaces polyester options that off-gas synthetic fibres. Silicone ear plugs avoid the biocide-treated foam alternatives.

These aren't premium-market products. They're affordable swaps for everyday items you already use. The goal isn't a perfect chemical-free home. It's reducing the highest-exposure sources — starting with what touches your body longest — and doing it in a way that doesn't require overhauling your life.

Go back to the pregnancy test on the Tuesday morning. Over the next nine months, the fetus that begins there will build a brain, a reproductive system, and an entire endocrine architecture — calibrated at concentrations so precise that twenty parts per trillion of estradiol at the wrong moment shapes it for life. Through the placenta, into that same narrow range, flows whatever the mother absorbed that day from her moisturiser, her tap water, her lunch, and the receipt from the chemist.

Most of those chemicals weren't designed to interfere with hormones. They were designed to preserve cosmetics, line cans, coat cookware, make plastics flexible. The disruption is a side effect of structural similarity — a molecule shaped enough like estrogen will be read as estrogen by a system that can't check credentials. The testing model that approved them was built for cyanide, not for estradiol. Endocrine disruptors follow different rules: maximum effect at low doses, different outcomes depending on timing, and biological activity that changes with the route of exposure.

The framework hasn't caught up. The research has. The fetus doesn't get to choose what's in the blood that arrives. You do.

Check your most-used personal care products tonight. If any list 'fragrance,' 'parfum,' or parabens in the first five ingredients — those are the easiest swaps to make, and the evidence says the difference shows up within days.