Bisphenol A: what the substance from the can does to your hormones
For four decades BPA was considered under control. Then a European authority lowered its own limit value by roughly a factor of 20,000, and another countered it by a factor of 1000. This story is worth your time, and it works without panic.
For me, bisphenol A is neither a single culprit nor a reason to panic. It is the best available evidence that toxicological safety is a moving target. A substance was considered well studied for forty years, then an authority moved its own assessment down by four orders of magnitude, and a second one disagreed. That is why, with hormonal symptoms, I look at the sum of everyday sources and not at the single tin can.
You are standing at the till. The receipt rolls out of the printer, you put it in your trouser pocket because you might still need it. At home you put two cans of tomatoes on the shelf and pour the leftovers into the old plastic container that has already gone slightly cloudy.
Three completely unremarkable movements. All three involve the same substance, or a very close relative of it.
Many women know the pattern that comes next. You read something somewhere about hormone poisons in plastic, you google for half an hour, and at the end you are left with a mixture of unease and helplessness. Panic is one wrong answer. Shrugging is the other.
So I would like to offer you something different: the story of how the scientific community itself wrestles with this substance. It is more interesting than any warning list, and in the end it explains better why I ask about environmental factors in my consultations when hormonal symptoms come up.
What awaits you here
- Why EFSA lowered the limit value roughly 20,000-fold in 2023
- Why the BfR countered with a thousand times that value
- Which receptors BPA docks onto in cell experiments
- Where it hides in everyday life, and what the EU ban has changed since 2025
- The half-life, and why it means two things at once
- What non-monotonic dose response curves are, in everyday language
- Three real human intervention studies, with their limits
- What observational data show on PCOS, weight and diabetes
- The null finding on fertility that almost nobody quotes
- Why BPA free is not a safety promise
- What measurably lowers exposure, without turning into control
Before we get to environmental factors, one paragraph belongs up front. There are hormonal and gynaecological complaints where a medical work-up comes first, not the search for a cause in the kitchen. These include:
- any bleeding after the menopause, meaning after the menstrual cycle has already been absent for a year
- very heavy bleeding or a period that has suddenly changed markedly
- lower abdominal pain together with fever
- acute, one-sided lower abdominal pain
- unintended weight loss
- visual disturbances or headaches together with milk discharge from the breast, because a prolactinoma can be behind this
- rapidly increasing virilisation, meaning heavy body hair, a deepening voice or hair loss on the head within a few months
These signs belong promptly in a medical or gynaecological consultation. The rest of this article describes the level that comes afterwards, not the one that comes before.
The day an authority lowered its own limit value 20,000-fold
Imagine someone tells you for years that one glass of water a day is harmless. And then the same body gets back to you and says: one drop.
That is exactly what happened in April 2023, in numbers rather than images.
The European Food Safety Authority, EFSA for short, had set a temporary tolerable daily intake for bisphenol A in 2015. It was 4 micrograms per kilogram of body weight per day. Such a value, called a TDI in technical language, is not a poison threshold. It is the amount that, according to the knowledge at the time, could be taken in daily over a lifetime without a health concern being expected.
The responsible panel for food contact materials adopted a new opinion on 6 December 2022. It was published in April 2023. The new value reads 0.2 nanograms per kilogram of body weight per day.
A nanogram is a thousandth of a microgram. From 4 micrograms to 0.2 nanograms is a difference of roughly a factor of 20,000.
Please read the third card again. It looks almost identical to the second, and it is a thousand times as large. This unit trap is the most common error in texts on this topic, and it is not a detail. It is the core of the whole dispute.
The EFSA CEP panel worked through the BPA literature according to a protocol defined in advance, with its own uncertainty analysis.
A change in Th17 cells in mice was selected as the critical endpoint, meaning a group of immune cells involved in inflammatory processes and in autoimmune events. From this the panel derived a reference point of 8.2 nanograms per kilogram of body weight per day as a human equivalent dose, applied a total uncertainty factor of 50 and arrived at 0.2 nanograms. The panel also recorded that the exposure estimates from the 2015 opinion exceed this new value in all age groups by two to three orders of magnitude.
For you this means: the new value did not arise because anyone had observed new poisonings. It arose because a different endpoint was considered decisive than before.
EFSA CEP Panel. Re-evaluation of the risks to public health related to the presence of bisphenol A (BPA) in foodstuffs. EFSA Journal. 2023;21(4):e06857. PMID: 37089179 · DOI: 10.2903/j.efsa.2023.6857 [Guideline]The political consequence came a year and a half later. With Regulation (EU) 2024/3190 of 19 December 2024, the European Commission banned the use of BPA in materials in contact with food. It has been in force since 20 January 2025. Affected are adhesives, plastics, printing inks, ion exchange resins, silicones as well as varnishes and coatings, expressly including the inner coating of tin cans.
What matters alongside this is realism. Staggered transition periods apply. The first general deadline for single-use products ends on 20 July 2026, and existing stocks may be sold off. A ban of this magnitude does not take effect overnight.
Most people read a piece of news like this as a scandal. Too high for decades, so we were deceived.
I read it differently. A limit value is not a constant of nature. It is a snapshot of the state of knowledge, assembled from the selection of studies, the choice of the critical endpoint and a safety margin for everything that is not known. If one of these three building blocks changes, the number changes.
And now you know why I rarely ask in the consultation whether a limit value is being met. I would rather ask how high the daily exposure actually is and where it comes from.
What bisphenol A is and which switches it docks onto
If you picture hormones as messages, then receptors are the letterboxes. Each message fits into a particular box, and only there does it trigger something.
Bisphenol A is a molecule made of two phenol rings connected by a small bridge. This shape resembles the body's own oestrogen closely enough to fit into the same letterbox. Not well, but well enough.
The substance has been produced industrially since the 1950s, as a building block for polycarbonate and for epoxy resins. Both are materials we can no longer imagine doing without: hard, clear, heat resistant, strongly adhesive. The hormonal similarity was a footnote for a long time.
The term for the whole field has existed since 1991. That year, twenty-one scientists at the Wingspread conference agreed on the expression endocrine disruptor. Anyone searching today for examples of endocrine disruptors will find, besides bisphenols, phthalates, certain pesticides, PFAS and fungal toxins such as zearalenone. I have written the overview of this group elsewhere, for women under Xenoestrogens in everyday life and for men under Xenoestrogens in men.
Why weak oestrogen is an incomplete description
Almost every text says BPA is a weak oestrogen. That is true for the classical pathway. Its binding strength at the nuclear oestrogen receptor is far below that of estradiol, the body's own hormone.
The sentence still falls short, for a simple reason: the letterbox is not the only one.
A review from 2018 collected all hormone receptors for which binding of BPA had been described.
In the cell and computer models evaluated, BPA binds to a considerable number of receptors: to several nuclear and membrane-bound oestrogen receptors, to the androgen receptor, to the thyroid hormone receptor, to the glucocorticoid receptor and to PPAR gamma, a switch of fat metabolism. The authors argue that precisely this diversity might explain several things about the behaviour of BPA that seem puzzling.
For you this means: talk of a weak oestrogen describes one door out of several. And very importantly: these are cell experiments and model calculations, not an observed process in your body.
MacKay H, Abizaid A. A plurality of molecular targets: The receptor ecosystem for bisphenol-A (BPA). Horm Behav. 2018;101:59-67. PMID: 29104009 · DOI: 10.1016/j.yhbeh.2017.11.001 [Mechanism Review, in vitro and in silico]The Endocrine Society, the largest endocrinological professional society in the world, published its second scientific statement on endocrine disruptors in 2015, EDC-2 for short. It covers seven topic areas that the author group considers most transferable to humans. I am only reporting the list here, and I derive from it no statement about diagnostics or treatment in my practice for any of these fields: obesity and diabetes, female reproduction, male reproduction, hormone-sensitive cancers in women, prostate, thyroid and neurodevelopment.
BPA is expressly highlighted in it, because the most literature exists on it. The author group records that non-monotonic dose response relationships, low-dose effects and the particular sensitivity of early developmental phases now belong to the understood endocrinological framework. And in the same document it expressly urges caution when inferring causality in humans.
Gore AC, Chappell VA, Fenton SE et al. EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocr Rev. 2015;36(6):E1-E150. PMID: 26544531 · DOI: 10.1210/er.2015-1010 [Guideline]
Many people expect an endocrine disruptor to behave like a poison: one amount, one harm, and below that nothing. But that is not how the hormonal system works.
Hormones are more like a dial with feedback. Small signals can trigger a lot, large signals can blunt the dial, and the body is constantly compensating. Anyone who has once grasped this difference also understands why toxicology and endocrinology have not agreed on BPA for over twenty years.
And now you know why I note several times in this article whether an observation comes from cell culture, from animal experiments or from humans.
Where bisphenol A really hides in everyday life
The most common search query about this substance is roughly: where is this stuff? The answer is less spectacular than many think, and that makes it more practical.
From the can to the water pipe
- Tins and drinks cans. The inner coating is classically an epoxy resin that uses BPA as a building block. It protects the metal from the acidity of the contents. That is exactly why this source is the most important in terms of quantity, and exactly why it is named explicitly in the EU ban. How often BPA was actually detectable in tins was examined by Stiftung Warentest in 2024. Anyone who wants the exact figures will find the test at Stiftung Warentest itself. I deliberately do not pass on a second-hand number here.
- Thermal paper. Till receipts, parking tickets, labels. Here the bisphenol is not a firmly built-in component but sits on the surface as a free developer. That is why it can transfer to the skin easily. In the EU, thermal paper with a BPA content of 0.02 percent by weight or more may no longer be placed on the market since 2 January 2020.
- Polycarbonate. Hard, clear plastic, often marked with recycling code 7. Older reusable bottles, storage containers, tops of water dispensers. Heat, alkali and mechanical stress, meaning dishwasher and microwave, can promote release. This figure comes from laboratory material experiments.
- Dental plastics. Composites and fissure sealants contain BPA derivatives. What is measurable is above all the first hour after placement.
- Renovated drinking water pipes. Instead of being replaced, old pipes are sometimes lined with epoxy resin. If curing is incomplete, bisphenol A can transfer into the water. This source affects few households, but it is a continuous one.
The order is not a ranking of your personal risk. It only sorts where the substance is technically built in.
A Portuguese working group searched PubMed, Cochrane and Embase for studies that had measured BPA in urine, saliva or blood after dental plastic treatments.
Of 5111 hits, 20 studies remained. Most showed a rise in BPA level about one hour after treatment. A week later the values had fallen again. The author group recommends practical precautions: rubber dam, immediate polishing of all plastics used, glycerine gel against incomplete curing of the topmost layer, mouth rinsing after treatment and as few fillings per session as possible.
For you this means: the rise is temporary, and there are dental steps that can make it smaller. These are questions for the conversation in the dental practice, not a reason to postpone necessary treatment. The author group considers these precautions especially important in children, adolescents and in pregnancy, and it names an upper limit of at most four fillings or sealants per appointment.
Paula AB, Toste D, Marinho A et al. Once Resin Composites and Dental Sealants Release Bisphenol-A, How Might This Affect Our Clinical Management? A Systematic Review. Int J Environ Res Public Health. 2019;16(9):1627. PMID: 31075949 · DOI: 10.3390/ijerph16091627 [Systematic Review, human measurements]What has regulation changed? Two things, and both are concrete.
First, the till receipts. The European Chemicals Agency ECHA has formally classified BPA in several steps. On 16 June 2017 the Member State Committee agreed unanimously to place the substance on the candidate list of substances of very high concern because of endocrine properties with probable serious effects on human health. The proposal came from France. The restriction for thermal paper is separate from this and has applied since January 2020. German receipts today as a rule no longer contain BPA. That is half the news. The other half: other developers have taken its place, often bisphenol S, and about its consequences in humans we know even less. The practical advice therefore stays the same as before, namely do not collect receipts. Old receipts in drawers and imported goods remain a residual issue.
Second, the cans. The EU ban since January 2025 applies to food contact materials as a whole. It will take time until the transition periods have expired and stocks have been sold off.
In toxicology, pregnancy, breastfeeding and the first years of life are considered particularly sensitive phases. The reason is developmental biology: in these windows, control circuits are laid down, and a signal at the wrong time could shift more than the same signal later on.
What matters is what follows from this and what does not. It follows that caution during this time is sensible: more fresh food, fewer cans, no hot food in old polycarbonate containers, no collecting of receipts.
It does not follow that harm has been proven. And it certainly does not follow that anyone should eat less in pregnancy or skip an examination. All questions about nutrition, check-ups and medication in pregnancy belong in medical or midwife-led care.
Most lists of this kind end with the feeling that there is something in everything. That is even true, and it is still the wrong conclusion.
Because these five sources are not equal in size. The can you open twice a month and the can that is on the table every day are two completely different stories. The next section explains why that is, and in doing so it answers the question that almost never gets answered.
How it gets into the body and how quickly it disappears again
This is the question that almost nowhere gets answered. Yet it puts the whole topic in order.
The main route is food. What you eat and drink brings in the largest part. Some comes through the skin, above all with frequent contact with thermal paper, and very little through the air you breathe.
Then something remarkable happens in the gut and the liver. BPA is very quickly coupled to glucuronic acid. This process is called glucuronidation and is a standard route of the second detoxification phase. According to current understanding it makes the molecule water soluble and in doing so takes away its ability to dock onto hormone receptors. After that it leaves via the urine. I have described in detail how these two detoxification phases work together in Liver, phase 1 and phase 2.
A team around Teeguarden gave ten adult male volunteers 30 micrograms of labelled BPA per kilogram of body weight in soup and then followed serum and urine for 24 hours.
Absorption followed a half-life of 0.45 hours. The highest serum concentration was reached after 1.6 hours and made up less than 0.3 percent of the total labelled BPA. The coupled glucuronide appeared measurably faster than the free BPA, 0.29 versus 0.45 hours. Excretion of the administered dose was complete after 24 hours, which argues against a tissue depot.
For you this means: what went into the can yesterday is no longer in you today. The body needs no detox programme for this, it needs a few hours.
Teeguarden JG, Twaddle NC, Churchwell MI et al. 24-hour human urine and serum profiles of bisphenol A: Evidence against sublingual absorption following ingestion in soup. Toxicol Appl Pharmacol. 2015;288(2):131-42. PMID: 25620055 · DOI: 10.1016/j.taap.2015.01.009 [Controlled human study, pharmacokinetics, n=10]And now comes the punchline, and it has two halves that are both true.
A substance with a half-life in the range of hours that is nevertheless regularly measurable in European biomonitoring surveys must be resupplied constantly, otherwise it would no longer be there. That is a conclusion, not a measurement, and I say it as such. The single can is not what counts. The daily resupply counts.
The first half is reassuring. Anyone who eats a can of chickpeas once a month has not built up a depot. With BPA there is nothing that accumulates in fat tissue over years, as is the case with some other environmental substances.
The second half is sobering. If a substance should have disappeared after 24 hours and is nevertheless regularly measurable, then the conclusion is close at hand: new material keeps arriving. Exposure is not an event, it is a state.
That is why avoidance strategies for BPA have to be thought about differently than for heavy metals. With mercury and cadmium it is about stores, about half-lives of years, about measurements in the right medium. How that works is described in Measuring heavy metals in blood or urine. With BPA it is about the inflow.
The exception through the skin
A small experiment with six male volunteers who simulated till work with thermal paper showed a third aspect. After handling the receipts, urinary BPA concentration was three times the baseline value, while triclosan as a reference substance remained stable. 48 hours after the end of the experiment, BPA values had not yet fallen back to baseline. By calculation, the skin route explained between 51.9 and 84 percent of urinary BPA, with a geometric mean of 70.9 percent.
Six participants is very few, and that has to be said. What remains interesting is the direction: what comes through the skin bypasses the rapid first pass through the liver and therefore behaves more sluggishly than what you eat. I am not resolving this tension with the oral kinetics study here, I am naming it.
Lv Y, Lu S, Dai Y et al. Higher dermal exposure of cashiers to BPA and its association with DNA oxidative damage. Environ Int. 2017;98:69-74. PMID: 27729163 · DOI: 10.1016/j.envint.2016.10.001 [Human experiment, single-arm, uncontrolled, n=6]
Many people look for the one dangerous thing they can cross off. One product, one material, one brand. Then the topic would be done with.
With a substance that has an hourly half-life, this way of thinking does not work. It is not about crossing off one thing but about lowering a background hum. That is less spectacular and considerably less exhausting, because no single mistake carries weight.
And now you know why in the consultation I ask about habits and not about sins.
The dispute about low doses, told honestly
There is a sentence that has been in every textbook since Paracelsus: the dose makes the poison. It is correct and, for hormones, still incomplete.
The reason lies in how the system is built. A poison damages more the more of it is present. A hormone, by contrast, works within a control circuit. There is saturation, there is counter-regulation, there are receptors that get turned down under constant bombardment.
When you dim a lamp, it gets brighter the further you turn. When you test a motion detector, by contrast, a tiny movement can trigger it and a very large one can overload it. The first is monotonic, the second is not.
What a non-monotonic dose response curve is
Technically it is defined as a non-linear relationship in which the sign of the slope changes within the dose range studied. In plain language: the effect first increases and then decreases again, or the other way round. A 2012 review compiled, by its own account, hundreds of such examples, from cell culture, animal experiments and epidemiology. The conclusion of the author group: where such curves occur, the effects of low doses cannot be predicted from the effects of high doses.
That has consequences, because classical toxicology does exactly that. It tests high doses, looks for the threshold, divides by safety factors and assumes that nothing happens below that.
And here is the honesty that belongs with it: this review comes from authors who have held a clear position in this debate for decades. It is one side of the dispute, not a consensus.
Vandenberg LN, Colborn T, Hayes TB et al. Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr Rev. 2012;33(3):378-455. PMID: 22419778 · DOI: 10.1210/er.2011-1050 [Mechanism Review, cell, animal and human data]
CLARITY-BPA was a joint programme of the American institutes NIEHS, NTP and FDA. It combined a classical regulatory toxicology study with numerous academic sub-studies in the same Sprague-Dawley rats.
A synthesis of the academic sub-studies reports effects across several organ systems and dose levels, with non-monotonic patterns. Remarkably consistent across most sub-studies including the core study were effects at the low doses of 2.5, 25 and 250 micrograms of BPA per kilogram of body weight.
For you this means: this is the most elaborate attempt so far to settle the question experimentally, and it has not settled it. All of this comes from rats, not from humans. And the authorship overlaps with the 2012 review, which I am not hiding here.
Prins GS, Patisaul HB, Belcher SM, Vandenberg LN. CLARITY-BPA academic laboratory studies identify consistent low-dose Bisphenol A effects on multiple organ systems. Basic Clin Pharmacol Toxicol. 2019;125 Suppl 3:14-31. PMID: 30207065 · DOI: 10.1111/bcpt.13125 [In vivo, rat]And now the other side, on equal footing and with its own reasoning.
Two authorities, the same data, a factor of 1000 apart
EFSA, April 2023
Critical endpoint: Th17 cells in mice. Reference point 8.2 ng/kg body weight per day as a human equivalent dose, total uncertainty factor 50.
Result: TDI 0.2 nanograms per kilogram of body weight per day. Conclusion of the panel: a health concern from BPA in food across all age groups.
BfR, 19 April 2023
Its own derivation in opinion No 018/2023. Points of criticism included the restriction of the EFSA literature search to studies from the years 2013 to 2018 and the focus on immune effects in young mice.
Result: health-based guidance value of 0.2 micrograms per kilogram of body weight per day. That is a thousand times as much. The EMA also raised objections to the EFSA derivation.
Added to this is a methodological critique from the journal Toxicological Sciences in 2024. It argues that the chosen endpoint is an immunotoxicological intermediate finding in mice that was not observed in other species, and that evidence is missing that it is a precursor event of a pathological outcome. The new value lies several orders of magnitude below all values established worldwide, including EFSA's own temporary value from 2015.
Transparency about this, because it belongs to the assessment: both authors of this critique work, according to their PubMed affiliation, for the consultancy Gradient. That does not devalue their argument. It still belongs in the picture.
Prueitt RL, Goodman JE. Evidence evaluated by European Food Safety Authority does not support lowering the temporary tolerable daily intake for bisphenol A. Toxicol Sci. 2024;198(2):185-190. PMID: 38265237 · DOI: 10.1093/toxsci/kfad136 [Methodological comment in a scientific journal]
German Federal Institute for Risk Assessment (BfR). Bisphenol A: BfR proposes health-based guidance value. Opinion No 018/2023 of 19 April 2023. [Guideline]
When two serious authorities read the same data and end up a factor of 1000 apart, the obvious reaction is: then nobody knows, so it does not matter.
I draw the opposite conclusion. If the assessment is that movable, then the number I orient myself by is not the limit value. It is the question of how much resupply is happening in my everyday life and how easily that can be lowered.
And now you know why I tell this dispute at length instead of condensing it into a headline.
What human intervention studies actually measure
Up to here it has been about cells, mice, rats and regulatory calculations. Now comes the part that interests me most clinically: what happens measurably when real people change their behaviour?
There are three studies that deserve the name intervention. All three share one thing, which I put up front so that it does not get lost: what was measured was exposure, not disease. None of these papers shows that anyone feels better. They show that concentrations in urine shift.
A team around Rudel at the Silent Spring Institute in Newton, Massachusetts, recruited 20 people from five families, selected because they reported eating a lot from cans and packaged foods.
The sequence: first the usual diet, then three days of fresh foods without cans and without plastic packaging, then the usual diet again. Evening urine was collected over eight days in January 2010. BPA in urine fell from a geometric mean of 3.7 nanograms per millilitre to 1.2 nanograms per millilitre, which corresponds to a reduction of 66 percent. Maximum values fell by 76 percent. For plasticiser metabolites the maxima fell by 93 to 96 percent.
For you this means: three days are enough for a marked change, because the half-life is short. And: 20 people is few, there was no control group, and the selection consisted of heavy users.
Rudel RA, Gray JM, Engel CL et al. Food packaging and bisphenol A and bis(2-ethyhexyl) phthalate exposure: findings from a dietary intervention. Environ Health Perspect. 2011;119(7):914-20. PMID: 21450549 · DOI: 10.1289/ehp.1003170 [Human intervention, before and after, uncontrolled, n=20]A working group around Carwile at the Harvard School of Public Health in Boston placed 84 volunteers into a randomised crossover design, and 75 completed it. Median age 27 years, 68 percent women.
Two five-day phases with two washout days in between, each with one portion of soup daily, once from a can, once freshly prepared. The specific-gravity adjusted geometric mean of urinary BPA was 1.1 micrograms per litre after fresh soup and 20.8 micrograms per litre after canned soup. The authors report the difference of 22.5 micrograms per litre as an increase of 1221 percent.
For you this means: it is the most memorable number of the whole topic and at the same time the one most easily misunderstood. Canned soup daily for five days in a row is not normal eating behaviour, and what was measured was excretion, not harm.
Carwile JL, Ye X, Zhou X, Calafat AM, Michels KB. Canned soup consumption and urinary bisphenol A: a randomized crossover trial. JAMA. 2011;306(20):2218-20. PMID: 22110104 · DOI: 10.1001/jama.2011.1721 [RCT, crossover, n=75]A team around Lee at Seoul National University followed 54 female cashiers from seven branches of a large South Korean supermarket chain.
For one week they worked without gloves, the following week with them. Urine was examined before and after the shift on the first two consecutive days of each week. Without gloves, urinary BPA concentration rose across the shift to roughly double. With gloves, no change was seen. In an additional observational part, higher urinary BPA values were linked with higher fasting insulin and stronger insulin resistance.
For you this means: a simple barrier changed uptake measurably. The insulin part is a cross-section and must not be read causally.
Lee I, Kim S, Kim KT et al. Bisphenol A exposure through receipt handling and its association with insulin resistance among female cashiers. Environ Int. 2018;117:268-275. PMID: 29778011 · DOI: 10.1016/j.envint.2018.05.013 [Human intervention in the workplace, n=54]A survey by the French institute INRS fits with this. It examined 90 cashiers compared with 44 controls who were not occupationally exposed. The median of total BPA in urine was 3.54 micrograms per litre in the controls and 8.92 micrograms per litre in the cashiers.
The most interesting finding of this paper is a negative one: no correlation was found with the number of receipts handled, and neither hand-washing frequency nor age, length of employment or smoking showed an effect. The single till receipt is therefore not the unit of account that many make of it.
Ndaw S, Remy A, Jargot D, Robert A. Occupational exposure of cashiers to Bisphenol A via thermal paper: urinary biomonitoring study. Int Arch Occup Environ Health. 2016;89(6):935-46. PMID: 27126703 · DOI: 10.1007/s00420-016-1132-8 [Biomonitoring, cross-sectional with control group, n=134]
The usual mistake in advice articles looks like this: three days of eating fresh, minus 66 percent, so the hormonal balance falls back into place. That sentence is tempting, and it is factually wrong.
What these studies show is something different and still valuable: exposure can be influenced, quickly and without effort. Whether a health benefit follows from that has not been studied to this day. That is not an argument against the measure, it is simply an honest label on it.
What observational data show about hormones, and where they stop
Now comes the part most people are looking for when they search for this article at all. And it is the part where the data become thinnest, even though the opposite is claimed online.
Everything that links BPA with a disease comes from observational studies. That means: concentrations in urine or blood are measured in many people, researchers look at who has which diagnosis, and calculate. What comes out is an association. It may reflect a cause. It may also reflect that sick and healthy people eat differently, live differently or metabolise differently.
Added to this is a methodological problem that weighs particularly heavily in this field. Almost all of these studies measure once. With a substance that has a half-life in the range of hours, a single measurement mainly reflects what was eaten in the last few hours. It is a shaky proxy for a lifetime of exposure.
| Statement | Where the data come from | What follows from it |
|---|---|---|
| BPA binds to several hormone receptors | Cell culture and computer models | Mechanistically plausible, no bodily process documented in humans |
| Non-monotonic curves, effects at very low doses | Rats and mice, cell culture | Observed repeatedly in animal experiments, contested in regulatory toxicology |
| Link with PCOS, insulin resistance, diabetes, weight | Cross-sections and meta-analyses in humans | Association, no documented cause, direction open |
| Canned food and receipt contact raise urinary values | Intervention studies in humans | Documented, but what was measured was exposure and not health |
| Link with female fertility | Prospective cohort at a fertility clinic | Here no link was found |
A working group around Kandaraki at a university hospital compared 71 women with PCOS by NIH criteria with 100 healthy women, matched for age and body weight.
BPA in blood was 1.05 in the PCOS group versus 0.72 nanograms per millilitre in the controls. The difference showed up in both slim and overweight participants. BPA was also linked with testosterone and with androstenedione, and in the PCOS group with insulin resistance as well.
For you this means: this is a cross-section, meaning a snapshot. Measurement was in serum rather than urine. And the author group itself describes an interaction in both directions, because androgens in turn may influence the breakdown of BPA. The direction of causality is therefore open.
Kandaraki E, Chatzigeorgiou A, Livadas S et al. Endocrine disruptors and polycystic ovary syndrome (PCOS): elevated serum levels of bisphenol A in women with PCOS. J Clin Endocrinol Metab. 2011;96(3):E480-4. PMID: 21193545 · DOI: 10.1210/jc.2010-1658 [Cross-sectional, case-control comparison, n=171]PCOS is a medical diagnosis with its own criteria and its own treatment. What lies behind it and how the work-up runs I have described in Understanding PCOS. The insulin aspect, which for very many women is the actual lever, is covered in Insulin resistance and female hormones. Nothing in this section suggests replacing a gynaecological work-up or changing an ongoing treatment on your own. If you are already being treated, for example with the pill, with metformin or with an antiandrogen, then please change nothing about it on the basis of this article. These are all prescription-only medicines that have benefits and risks and are therefore adjusted and monitored medically. Every adjustment, including stopping, belongs in that conversation, and the same applies to hormone replacement therapy or a thyroid medication.
How broad the observational data are overall is shown by an umbrella review from 2024. It included 52 systematic reviews, from which 759 meta-analyses came, with 78 percent of these coming from reviews of only moderate methodological quality. For BPA, associations are reported with type 2 diabetes in adults, insulin resistance in children and adults, polycystic ovary syndrome, obesity and high blood pressure as well as cardiovascular disease. For the substitutes there were no meta-analyses at all. Part of the author group is based at the Minderoo Foundation, a foundation with a declared agenda in this field.
Symeonides C, Aromataris E, Mulders Y et al. An Umbrella Review of Meta-Analyses Evaluating Associations between Human Health and Exposure to Major Classes of Plastic-Associated Chemicals. Ann Glob Health. 2024;90(1):52. PMID: 39183960 · DOI: 10.5334/aogh.4459 [Systematic Review, umbrella review]
Two numbers that are told wrongly online
The first comes from a cross-sectional analysis of the American health study NHANES with 1455 adults. Per one standard deviation increase in BPA concentration, the odds ratio was 1.39 for cardiovascular diagnoses and 1.39 for diabetes. What is almost never quoted alongside it: for the other common conditions examined, no link was found. Two liver values were conspicuous, however, gamma-GT with an odds ratio of 1.29 and alkaline phosphatase with 1.48. Both are laboratory values in a snapshot, not diagnoses. Both halves belong to the study.
Lang IA, Galloway TS, Scarlett A et al. Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. JAMA. 2008;300(11):1303-10. PMID: 18799442 · DOI: 10.1001/jama.300.11.1303 [Cross-sectional, NHANES, n=1455]
The second number is the famous one. A prospective cohort with 3883 adults from NHANES 2003 to 2008, linked with mortality data up to the end of 2015, found a hazard ratio of 1.49 for all-cause mortality for the highest versus the lowest tertile. On advice pages this regularly turns into a 49 percent higher mortality.
Hazard ratio 1.49, confidence interval 1.01 to 2.19
A hazard ratio is not a percentage. And the confidence interval, meaning the range in which the true value lies with 95 percent probability, runs here from 1.01 to 2.19. The lower bound sits practically at 1, which means as much as: no difference.
Added to this is that both cause-specific results were not significant: 1.46 with an interval of 0.67 to 3.15 for cardiovascular mortality and 0.98 with an interval of 0.40 to 2.39 for cancer mortality.
I still quote this study, because it exists and is properly done. I only quote it in full.
Bao W, Liu B, Rong S et al. Association Between Bisphenol A Exposure and Risk of All-Cause and Cause-Specific Mortality in US Adults. JAMA Netw Open. 2020;3(8):e2011620. PMID: 32804211 · DOI: 10.1001/jamanetworkopen.2020.11620 [Cohort, prospective, n=3883]
A working group around Minguez-Alarcon at the Harvard T.H. Chan School of Public Health and at Massachusetts General Hospital followed 256 women through 375 IVF cycles between 2004 and 2012 in the EARTH study, with a total of 673 urine samples before egg retrieval.
The adjusted geometric mean for BPA was 1.87 micrograms per litre, comparable with women in the American population sample NHANES 2011 to 2012. No link was found between urinary BPA and endometrial thickness, peak estradiol, the proportion of high-quality embryos or the fertilisation rate. Likewise no link with implantation, clinical pregnancy or live birth. The authors name one qualification themselves: in the younger women under 37 the endometrial lining was thicker across rising BPA quartiles, in the older women from 37 it was thinner. What that means nobody currently knows, and I do not claim otherwise. But it belongs to the study.
For you this means: precisely where fear is greatest, nothing was visible in this cohort. The author group nevertheless still calls the data on BPA and reproduction sparse, and I adopt that reticence.
Mínguez-Alarcón L, Gaskins AJ, Chiu YH et al. Urinary bisphenol A concentrations and association with in vitro fertilization outcomes among women from a fertility clinic. Hum Reprod. 2015;30(9):2120-8. PMID: 26209788 · DOI: 10.1093/humrep/dev183 [Cohort, prospective, n=256]Time is a real factor with this topic. Please do not read this section as a reason to postpone a reproductive medicine work-up or treatment in order to put the environment in order first. Environmental questions can be tackled in parallel, they are not a preliminary programme.
I am promising you nothing at this point. Nobody can hold out the prospect of a pregnancy to you from these data, least of all me. What I can say: exposure can be lowered quickly and without effort. Whether an advantage for fertility follows from that has not been studied. And if the rearranging starts to put you under pressure, the price is too high.
On the male side I have worked through the topic in Improving sperm quality.
If an article only shows the studies that found something, a picture emerges that has nothing to do with the literature. Fear emerges, and fear is a poor adviser for eating.
That is why the null finding stands here on equal footing with the association. What is well documented so far: exposure can be lowered. What is still open: whether a difference for health follows from that.
BPA free, and what is in there instead
This label has reassured many people. It is on drinking bottles, on lunch boxes, on baby products, and it sounds like an all-clear.
Strictly speaking, though, it says only one thing: one particular molecule is not present.
What is present instead, it does not say. And because a material is supposed to keep its technical properties, the replacement is often chemically close to the original. Regulatory research has its own term for this: regrettable substitution. One problematic substance is swapped for a similar one about which even less is known.
A systematic review from 2015 following the OHAT protocol collected everything available on the hormonal activity of the two most common substitutes.
32 studies were included, 25 of them purely in cell culture and seven in animal models. The majority found the hormonal potency of bisphenol S and bisphenol F to be in the same order of magnitude and with the same direction as that of BPA, meaning oestrogenic, anti-oestrogenic, androgenic and anti-androgenic. In membrane-mediated signalling pathways, BPS showed potencies that came close to those of estradiol.
For you this means: the statement holds for cell cultures and partly for animals. It does not hold as a statement about disease in humans, because there is hardly any data on that. According to their PubMed affiliation, the authors belong to The Endocrine Disruption Exchange, an organisation with a declared agenda in this field.
Rochester JR, Bolden AL. Bisphenol S and F: A Systematic Review and Comparison of the Hormonal Activity of Bisphenol A Substitutes. Environ Health Perspect. 2015;123(7):643-50. PMID: 25775505 · DOI: 10.1289/ehp.1408989 [Systematic Review, in vitro and animal]A third analogue is called bisphenol AF. Individual cell and animal data exist on it, but no robust human evidence. I mention it here only so that you know the name when you come across it. I make no claim about its significance for humans.
What is actually measured in Europe
The European initiative HBM4EU carried out harmonised biomonitoring in 23 countries between 2014 and 2021. 10,795 people were examined in three age groups: 3576 children aged 6 to 12, 3117 adolescents aged 12 to 18 and 4102 young adults aged 20 to 39.
For 15 biomarkers the available health-based guidance values were exceeded. In 11 to 14 percent of participants, bisphenol S among others was above the guidance value, as was cadmium. One side finding deserves its own sentence: in adolescents and adults, higher bisphenol values were linked with a lower level of education in the household. Exposure is also a social question.
Govarts E, Gilles L, Rodriguez Martin L et al. Harmonized human biomonitoring in European children, teenagers and adults: EU-wide exposure data of 11 chemical substance groups from the HBM4EU Aligned Studies (2014-2021). Int J Hyg Environ Health. 2023;249:114119. PMID: 36773580 · DOI: 10.1016/j.ijheh.2023.114119 [Biomonitoring, cross-sectional, n=10,795]
A Danish working group analysed 300 urine samples from young men in the general population, 100 each from the years 2009, 2013 and 2017.
The median values of BPA and of several plasticiser metabolites more than halved over these eight years. The substitutes for plasticisers, by contrast, rose markedly, in part by more than twentyfold. The bisphenol substitutes BPS and BPF also rose, but only slightly. Despite this shift, exposure to the old known substances including BPA was still higher in 2017 than exposure to the new ones.
For you this means: regulation changes exposure measurably, that is the good news. Substitution is real, that is the other. Neither of them amounts to an all-clear, and neither of them amounts to an alarm.
Frederiksen H, Nielsen O, Koch HM et al. Changes in urinary excretion of phthalates, phthalate substitutes, bisphenols and other polychlorinated and phenolic substances in young Danish men; 2009-2017. Int J Hyg Environ Health. 2020;223(1):93-105. PMID: 31669154 · DOI: 10.1016/j.ijheh.2019.10.002 [Trend analysis from cross-sections, n=300]The obvious reaction to this section is: so you cannot do anything right anyway.
I see it differently. The mistake lies in the expectation that a label will do the thinking for you. BPA free is a statement about a molecule, not about safety.
In practical terms this means that materials without bisphenols simply sidestep the question. Glass contains none. With stainless steel bottles a second look is worth it, because inner coatings, lids and drinking spouts can be made of plastic. I deliberately name no brands and no sources of supply, because this is not a shopping guide. And even more practically: how you use a container often counts for more than the material. Hot food does not belong in old, cloudy or scratched polycarbonate containers, and those belong neither in the dishwasher nor in the microwave.
What measurably reduces exposure, and what does not justify panic
After so much dispute and so many caveats, now comes the simplest part. It is short, and that is deliberate.
I mainly name measures for which there are measurement data in humans, and I say where a point rests on laboratory material experiments instead.
What lowered urinary values in studies
- More fresh food, fewer cans. In the three-day fresh food intervention, the geometric mean fell by 66 percent and the maxima by 76 percent. No giving up whole food groups, only a shift in packaging.
- Soups and ready meals from cans less often. In the randomised crossover, the difference between fresh and canned soup was 1.1 versus 20.8 micrograms per litre in urine. That is the single largest lever in the literature.
- Do not collect receipts, do not put them in your trouser pocket. For the general population a small point, since the EU restriction took effect in 2020. Relevant for people with continuous occupational contact, and there gloves changed uptake measurably in the Korean study.
- No hot food in old polycarbonate containers. No dishwasher, no microwave for cloudy or scratched containers of this kind. Heat and alkali can promote release, which comes from laboratory material experiments and not from measurements in humans.
- Ask questions at dental treatments. Rubber dam, immediate polishing, mouth rinsing afterwards. These steps are expressly recommended in the dental review and cost nothing.
And now what expressly does not belong here.
Elimination programmes against BPA
- What is on offer
- Binding agents, detox regimens, juice fasts, teas, programmes for the elimination of plastic chemicals.
- What speaks against it
- I am not aware of any study documenting the benefit of such a programme for BPA. And mechanistically it makes little sense: with a half-life in the range of hours and complete excretion of the administered dose within 24 hours in a human experiment, your metabolism is already doing this work itself.
- What counts instead
- Less resupply. That is unspectacular, free of charge and the only thing for which intervention data exist.
Why I am sparing with the detox concept in general is set out in Detox programme, juice fast, detox tea: a critical look. For substances with real stores, such as heavy metals, different considerations apply than for BPA. How to investigate that sensibly is in Measuring heavy metals in blood or urine.
I know the reaction to articles like this well. First one point on the list, then two, then the kitchen gets rearranged, then invitations get declined because you do not know what people cook with there.
Anyone who starts counting till receipts and avoiding food out of fear has not lowered their risk. They have created a new one. Restrictive eating behaviour justified by purity is a known route into an eating disorder, and it often begins with a good intention. If that sounds familiar or if something felt tight while you were reading, then that is a reason to talk to someone, and not a reason to implement the next point on the list. In Germany the BZgA eating disorder helpline is reachable anonymously and free of charge on 0221 892031, and the Telefonseelsorge around the clock on 0800 111 0 111. In an acute crisis, call 112. For understanding the background there is more here: Understanding eating disorders.
The rule of thumb I give in the consultation: if a measure makes you more relaxed about eating, it is right. If it makes you tense, it is not, no matter how well reasoned it sounds.
Why I ask about this in the consultation at all
Here comes the sentence that stands above everything else: it does not follow that every hormonal disorder has an environmental cause. The vast majority of menstrual cycle complaints, most thyroid problems and most metabolic disorders have other, often more mundane reasons. Sleep, stress, weight, genetics, an autoimmune disease, a pregnancy, a phase of life.
So why do I ask? Because the sum of everyday sources is one of the few factors that can be changed. Genes cannot be swapped, age cannot be stopped. What stands in the kitchen and what lands on the plate every day, by contrast, can be.
And because BPA does not stand alone. Hormonally active molecules reach us from several directions: from packaging, from fungal toxins such as zearalenone, which is a documented mycoestrogen and is described in detail in Zearalenone and hormones, and from metals such as cadmium, which is well studied as a metalloestrogen. Where mould in the home is added to the picture, Mould in the home is the right entry point, and the basics are in Mycotoxins: the basics.
Then there is the channel through which many of these threads run: silent inflammation and insulin. Both are described in Silent inflammation and weight and in Nutrition, blood sugar and hormones, and both are considerably better documented than any single environmental substance.
The environmental perspective comes on top. It does not come in place of the gynaecological work-up.
Shukri JarmoukliThis matters to me because it is easy to misunderstand. Environmental medicine questions are a field of their own with a literature of their own, and I have specialised in it professionally. The ultrasound, the hormone lab, the work-up of bleeding disorders and the treatment of endometriosis belong in gynaecology, and they are in good hands there. What I look at is a different level alongside it, not a better one. How the examination works in gynaecology and what endometriosis involves I have written up separately.
What I look at in addition is the environment in which a control circuit operates. Sometimes there is nothing to be found there. Sometimes it explains part of the picture. Which values make sense at all with hormonal complaints is set out in Testing hormones: which test and when, and how oestrogen is broken down via the liver in Lowering oestrogen naturally via the liver.
Bisphenol A is well studied and nevertheless contested. The mechanistic data are strong, the animal data are good and in part contradictory, the human data are predominantly observational, and the few intervention studies measure exposure rather than disease. Anyone who knows this can act without being afraid.
Frequently asked questions about bisphenol A and hormones
Are there typical symptoms of BPA exposure?
No. There is no symptom list that would let you recognise BPA exposure. What circulates online as a BPA symptom, meaning fatigue, menstrual cycle problems or weight gain, is unspecific and has a great many possible reasons. What does exist are observational data linking higher BPA levels in urine with diagnoses such as type 2 diabetes, insulin resistance, PCOS or obesity. An association in a population sample is something different from a symptom you can assign to yourself. Anyone with symptoms needs a proper medical work-up first.
Can BPA be measured in the lab, and is it worth doing?
It can be measured, usually as total BPA in urine. The catch is the kinetics. In a controlled human experiment with ten volunteers, absorption followed a half-life of 0.45 hours, and the administered dose had been fully excreted 24 hours later. A single urine value therefore mainly reflects the last few hours and not your exposure over the last few years. No gynaecological or endocrinological guideline calls for this measurement. I am not demonising the test, I am only saying that it answers less than most people expect.
How long does bisphenol A stay in the body?
After oral intake, a very short time. Ten volunteers received labelled BPA in soup, after which blood and urine were followed for 24 hours. Absorption had a half-life of 0.45 hours, the highest serum concentration was reached after 1.6 hours and made up less than 0.3 percent of the total labelled BPA, free BPA stayed below one percent of total BPA throughout, and excretion of the administered dose was complete after 24 hours. Through the skin, for example with constant contact with thermal paper, the process is slower. In a small experiment with six volunteers, levels were still above baseline 48 hours after the end of contact.
Why did EFSA lower the limit value so drastically in 2023?
Because the responsible panel chose a different critical endpoint than before. The decisive finding was a change in Th17 cells in a mouse model, meaning a group of immune cells involved in inflammatory processes. From this a reference point of 8.2 nanograms per kilogram of body weight per day was derived as a human equivalent dose, and a total uncertainty factor of 50 was applied. The result was a tolerable daily intake of 0.2 nanograms per kilogram of body weight. The temporary value from 2015 was 4 micrograms, which is a difference of roughly a factor of 20,000.
Why does the BfR disagree with EFSA when both have the same data?
Because they weight the same data differently. In April 2023, in its opinion No 018/2023, the BfR proposed a health-based guidance value of its own of 0.2 micrograms per kilogram of body weight per day, which is a thousand times the EFSA value. The reasons given included the restriction of the EFSA literature search to studies from the years 2013 to 2018 and the focus on immune effects in young mice. The EMA also raised objections. Please do not mix these up: 0.2 nanograms at EFSA, 0.2 micrograms at the BfR. That is the most common numerical error on this topic.
What does BPA free really mean?
It means that one particular molecule is not present. It does not automatically mean that nothing hormonally active is present. Bisphenol S, bisphenol F or bisphenol AF frequently replace the predecessor. A systematic review evaluated 32 studies, 25 of them purely in cell culture and seven in animal models, and found the hormonal potency of BPS and BPF to be in the same order of magnitude as that of BPA. Human data on disease consequences of these substitutes are largely missing. In the European HBM4EU survey, 11 to 14 percent of participants had a bisphenol S value above the health-based guidance value.
Are BPA free drinking bottles made of plastic or stainless steel the better choice?
The same applies to bottles as to all BPA free products: the label says something about one molecule, not about the entire chemistry of the material. I deliberately name no brands and no sources of supply here. What is sensible and takes no effort is the question of how you use a container. Old, scratched or cloudy polycarbonate containers do not belong in the dishwasher or in the microwave, and hot food is better not poured into them. Glass contains no bisphenols. With stainless steel a second look is worth it, because inner coatings, lids and drinking spouts can be made of plastic. So the same applies here: the material is a statement about the material, not about the whole product.
Is a can of tomatoes dangerous?
No, and this article deliberately does not suggest that conclusion. What is measurable is excretion: in a randomised crossover with 75 evaluable participants, the geometric mean of urinary BPA was 1.1 micrograms per litre after fresh soup and 20.8 micrograms per litre after canned soup. Participants ate one portion of canned soup daily for five days in a row, which is not normal eating behaviour. What was measured, moreover, was excretion and not health damage. What is relevant is continuous exposure over years, not the single can.
Do I really have to avoid till receipts?
For most people this is not a big issue. In the EU, thermal paper with a BPA content of 0.02 percent by weight or more may no longer be placed on the market since 2 January 2020. What remains relevant is occupational contact day after day. In a Korean study of 54 female cashiers, urinary concentration doubled across a shift without gloves, while with gloves no change was seen. A French survey of 90 cashiers compared with 44 controls found higher median values, but no correlation with the number of receipts handled. Not collecting receipts and not carrying them in your trouser pocket is enough in everyday life.
Is BPA in dental fillings and fissure sealants a problem?
A systematic review screened 5111 hits and included 20 studies that had measured BPA in urine, saliva or blood after dental treatment with composites or sealants. Most showed a rise about one hour after treatment, and a week later the values had fallen again. The author group names practical precautions: rubber dam, immediate polishing, glycerine gel against incomplete curing of the last layer, mouth rinsing afterwards and as few fillings per session as possible. These are questions for the conversation in the dental practice, not a reason to postpone necessary treatment. The author group considers these precautions especially important in children, adolescents and in pregnancy, and it names an upper limit of at most four fillings or sealants per appointment.
Can BPA be eliminated or detoxed?
There are no robust data for this, and mechanistically it makes little sense. With a half-life in the range of hours and complete excretion of the administered dose within 24 hours in a human experiment, your metabolism is already doing this work itself. That a detox programme achieves anything beyond that is not documented. What lowers exposure is less resupply, not a detox programme. I am not aware of any study showing a benefit of such a programme for BPA. If someone offers you one, the question about the study behind it is the simplest and fairest one you can ask.
What applies during pregnancy and breastfeeding?
Early developmental phases are considered particularly sensitive windows in toxicology, and that is the reason for the caution of the authorities, not proven harm in humans. What makes sense is therefore the calm option: more fresh food, fewer cans, no hot food in old polycarbonate containers, no collecting of receipts. What does not make sense: fear, going without adequate nutrition or postponing medical examinations. All questions about check-ups, medication and nutrition in pregnancy belong in medical or midwife-led care.
Does BPA have anything to do with PCOS or endometriosis?
For PCOS there are observational data. In a cross-sectional study with 71 women with PCOS and 100 controls, BPA in blood was 1.05 versus 0.72 nanograms per millilitre, with links to testosterone and androstenedione. An umbrella review from 2024 also finds an association. No cause follows from that. The author group of the PCOS paper itself describes a possible interaction in both directions, because androgens in turn may influence the breakdown of BPA. On endometriosis the data are thinner. Both are medical diagnoses that need a gynaecological work-up.
Since the EU ban in 2025: is the topic settled?
Not yet. Regulation (EU) 2024/3190 has been in force since 20 January 2025 and bans BPA in food contact materials, including the coating of tin cans. However, staggered transition periods apply, the first general deadline for single-use products ends on 20 July 2026, and existing stocks may be sold off. Then comes the second half of the answer: in a Danish time series from 2009 to 2017, median BPA values in urine more than halved, while BPS and BPF rose slightly. Regulation changes exposure measurably. It does not end the question.
Where this topic connects to the rest of your hormonal system
Bisphenol A is one piece of a puzzle. It only gets interesting next to the others. Here are the articles that make the most sense from here.
Xenoestrogens in everyday life
The whole group of endocrine disruptors at a glance, from plasticisers to pesticides. BPA is only one of many entries there.
If it is about menXenoestrogens in men
The same substances, different target measures: testosterone, aromatase, sperm quality. Worth reading with a partner.
If you want to start right at the beginningHormonal imbalance in women
The large overview from which this article deepens a single topic. There you will find all the control circuits placed in context.
If PCOS is on the tableUnderstanding PCOS
The diagnosis that comes up most often in the BPA context. Here is how it is made and what it actually means.
If weight and menstrual cycle are connectedInsulin resistance and female hormones
The control circuit that appears most often in the BPA observational data, and that can be influenced well independently of it.
If you want to know what to measureTesting hormones: which test and when
Which values make sense for which complaints, and why a BPA measurement in urine is not among them.
If the biochemistry interests youLiver, phase 1 and phase 2
Glucuronidation, through which BPA is excreted within hours, explained in the larger context.
If oestrogen dominance is your topicLowering oestrogen naturally via the liver
How the body's own oestrogen is broken down, and why the same route is also responsible for foreign substances.
If mould plays a roleZearalenone, a mycoestrogen
The parallel case from a completely different source: a fungal toxin that fits the same receptor. Much better documented than many think.
If you are thinking about stored substancesMeasuring heavy metals
Why cadmium and mercury have to be investigated quite differently from BPA, and which medium is suited to what.
If you are considering a detox programmeDetox programmes under scrutiny
What detox programmes can achieve and where they only cost money. For BPA the answer is particularly clear.
If avoidance turns into compulsionUnderstanding eating disorders
The protective link for this article. Purity thinking around food often begins with a good intention and does not end there.
If plasticisers are the questionPhthalates: the plasticisers
Why phthalates affect testosterone rather than oestrogen, and where they come from in everyday life.
If you want to place drinking water and cookwarePFAS: the forever chemicals
Why the forever chemicals stay in the body for years and where individual avoidance reaches its limit.
If you are looking at the bathroom shelfEndocrine disruptors in cosmetics
Parabens, UV filters and fragrances, sorted by what is documented and what is merely loud.
If you want the whole pictureWhy I look for environmental factors
Why I look for environmental factors in hormonal symptoms, along which grid, and when that search adds nothing.
Scientific sources
- EFSA Panel on Food Contact Materials, Enzymes and Processing Aids (CEP). Re-evaluation of the risks to public health related to the presence of bisphenol A (BPA) in foodstuffs. EFSA Journal. 2023;21(4):e06857. PMID: 37089179 · DOI: 10.2903/j.efsa.2023.6857 [Regulatory document, Guideline]
- German Federal Institute for Risk Assessment (BfR). Bisphenol A: BfR proposes health-based guidance value. Opinion No 018/2023 of 19 April 2023. bfr.bund.de [Regulatory document, Guideline]
- European Commission. Regulation (EU) 2024/3190 of 19 December 2024 on the use of bisphenol A and other bisphenols and bisphenol derivatives in materials and articles intended to come into contact with food. Official Journal of the EU, L series, 2024/3190, in force since 20 January 2025. eur-lex.europa.eu [Regulatory document, Guideline]
- European Chemicals Agency. Bisphenol A, candidate list of substances of very high concern (SVHC), decision of the Member State Committee of 16 June 2017 on endocrine properties. REACH restriction for thermal paper at 0.02 percent by weight or more, applicable since 2 January 2020. EC 201-245-8, CAS 80-05-7. echa.europa.eu [Regulatory document, Guideline]
- Gore AC, Chappell VA, Fenton SE, Flaws JA, Nadal A, Prins GS, Toppari J, Zoeller RT. EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocr Rev. 2015;36(6):E1-E150. PMID: 26544531 · DOI: 10.1210/er.2015-1010 [Guideline, consensus statement]
- MacKay H, Abizaid A. A plurality of molecular targets: The receptor ecosystem for bisphenol-A (BPA). Horm Behav. 2018;101:59-67. PMID: 29104009 · DOI: 10.1016/j.yhbeh.2017.11.001 [Mechanism Review, in vitro and in silico]
- Teeguarden JG, Twaddle NC, Churchwell MI, Yang X, Fisher JW, Seryak LM, Doerge DR. 24-hour human urine and serum profiles of bisphenol A: Evidence against sublingual absorption following ingestion in soup. Toxicol Appl Pharmacol. 2015;288(2):131-42. PMID: 25620055 · DOI: 10.1016/j.taap.2015.01.009 [Controlled human study, pharmacokinetics, n=10]
- Vandenberg LN, Colborn T, Hayes TB, Heindel JJ, Jacobs DR, Lee DH, Shioda T, Soto AM, vom Saal FS, Welshons WV, Zoeller RT, Myers JP. Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr Rev. 2012;33(3):378-455. PMID: 22419778 · DOI: 10.1210/er.2011-1050 [Mechanism Review, cell, animal and human data]
- Vandenberg LN, Maffini MV, Sonnenschein C, Rubin BS, Soto AM. Bisphenol-A and the great divide: a review of controversies in the field of endocrine disruption. Endocr Rev. 2009;30(1):75-95. PMID: 19074586 · DOI: 10.1210/er.2008-0021 [Mechanism Review, cell and animal data]
- Prins GS, Patisaul HB, Belcher SM, Vandenberg LN. CLARITY-BPA academic laboratory studies identify consistent low-dose Bisphenol A effects on multiple organ systems. Basic Clin Pharmacol Toxicol. 2019;125 Suppl 3:14-31. PMID: 30207065 · DOI: 10.1111/bcpt.13125 [In vivo, rat]
- Prueitt RL, Goodman JE. Evidence evaluated by European Food Safety Authority does not support lowering the temporary tolerable daily intake for bisphenol A. Toxicol Sci. 2024;198(2):185-190. PMID: 38265237 · DOI: 10.1093/toxsci/kfad136 [Methodological comment in a scientific journal]
- Rudel RA, Gray JM, Engel CL, Rawsthorne TW, Dodson RE, Ackerman JM, Rizzo J, Nudelman JL, Brody JG. Food packaging and bisphenol A and bis(2-ethyhexyl) phthalate exposure: findings from a dietary intervention. Environ Health Perspect. 2011;119(7):914-20. PMID: 21450549 · DOI: 10.1289/ehp.1003170 [Human intervention, before and after, uncontrolled, n=20]
- Carwile JL, Ye X, Zhou X, Calafat AM, Michels KB. Canned soup consumption and urinary bisphenol A: a randomized crossover trial. JAMA. 2011;306(20):2218-20. PMID: 22110104 · DOI: 10.1001/jama.2011.1721 [RCT, crossover, n=75]
- Lee I, Kim S, Kim KT, Kim S, Park S, Lee H, Jeong Y, Lim JE, Moon HB, Choi K. Bisphenol A exposure through receipt handling and its association with insulin resistance among female cashiers. Environ Int. 2018;117:268-275. PMID: 29778011 · DOI: 10.1016/j.envint.2018.05.013 [Human intervention in the workplace, n=54]
- Ndaw S, Remy A, Jargot D, Robert A. Occupational exposure of cashiers to Bisphenol A via thermal paper: urinary biomonitoring study. Int Arch Occup Environ Health. 2016;89(6):935-46. PMID: 27126703 · DOI: 10.1007/s00420-016-1132-8 [Biomonitoring, cross-sectional with control group, n=134]
- Lv Y, Lu S, Dai Y, Rui C, Wang Y, Zhou Y, Li Y, Pang Q, Fan R. Higher dermal exposure of cashiers to BPA and its association with DNA oxidative damage. Environ Int. 2017;98:69-74. PMID: 27729163 · DOI: 10.1016/j.envint.2016.10.001 [Human experiment, single-arm, uncontrolled, n=6]
- Kandaraki E, Chatzigeorgiou A, Livadas S, Palioura E, Economou F, Koutsilieris M, Palimeri S, Panidis D, Diamanti-Kandarakis E. Endocrine disruptors and polycystic ovary syndrome (PCOS): elevated serum levels of bisphenol A in women with PCOS. J Clin Endocrinol Metab. 2011;96(3):E480-4. PMID: 21193545 · DOI: 10.1210/jc.2010-1658 [Cross-sectional, case-control comparison, n=171]
- Symeonides C, Aromataris E, Mulders Y, Dizon J, Stern C, Barker TH, Whitehorn A, Pollock D, Marin T, Dunlop S. An Umbrella Review of Meta-Analyses Evaluating Associations between Human Health and Exposure to Major Classes of Plastic-Associated Chemicals. Ann Glob Health. 2024;90(1):52. PMID: 39183960 · DOI: 10.5334/aogh.4459 [Systematic Review, umbrella review]
- Lang IA, Galloway TS, Scarlett A, Henley WE, Depledge M, Wallace RB, Melzer D. Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. JAMA. 2008;300(11):1303-10. PMID: 18799442 · DOI: 10.1001/jama.300.11.1303 [Cross-sectional, NHANES, n=1455]
- Bao W, Liu B, Rong S, Dai SY, Trasande L, Lehmler HJ. Association Between Bisphenol A Exposure and Risk of All-Cause and Cause-Specific Mortality in US Adults. JAMA Netw Open. 2020;3(8):e2011620. PMID: 32804211 · DOI: 10.1001/jamanetworkopen.2020.11620 [Cohort, prospective, n=3883]
- Mínguez-Alarcón L, Gaskins AJ, Chiu YH, Williams PL, Ehrlich S, Chavarro JE, Petrozza JC, Ford JB, Calafat AM, Hauser R. Urinary bisphenol A concentrations and association with in vitro fertilization outcomes among women from a fertility clinic. Hum Reprod. 2015;30(9):2120-8. PMID: 26209788 · DOI: 10.1093/humrep/dev183 [Cohort, prospective, n=256]
- Rochester JR, Bolden AL. Bisphenol S and F: A Systematic Review and Comparison of the Hormonal Activity of Bisphenol A Substitutes. Environ Health Perspect. 2015;123(7):643-50. PMID: 25775505 · DOI: 10.1289/ehp.1408989 [Systematic Review, in vitro and animal]
- Govarts E, Gilles L, Rodriguez Martin L, Santonen T, Apel P et al. Harmonized human biomonitoring in European children, teenagers and adults: EU-wide exposure data of 11 chemical substance groups from the HBM4EU Aligned Studies (2014-2021). Int J Hyg Environ Health. 2023;249:114119. PMID: 36773580 · DOI: 10.1016/j.ijheh.2023.114119 [Biomonitoring, cross-sectional, n=10,795]
- Frederiksen H, Nielsen O, Koch HM, Skakkebaek NE, Juul A, Jørgensen N, Andersson AM. Changes in urinary excretion of phthalates, phthalate substitutes, bisphenols and other polychlorinated and phenolic substances in young Danish men; 2009-2017. Int J Hyg Environ Health. 2020;223(1):93-105. PMID: 31669154 · DOI: 10.1016/j.ijheh.2019.10.002 [Trend analysis from cross-sections, n=300]
- Paula AB, Toste D, Marinho A, Amaro I, Marto CM, Coelho A, Marques-Ferreira M, Carrilho E. Once Resin Composites and Dental Sealants Release Bisphenol-A, How Might This Affect Our Clinical Management? A Systematic Review. Int J Environ Res Public Health. 2019;16(9):1627. PMID: 31075949 · DOI: 10.3390/ijerph16091627 [Systematic Review, human measurements]
- The receptor ecosystem is cell culture. The binding of BPA to androgen, thyroid hormone and glucocorticoid receptors and to PPAR gamma comes from cell experiments and computer models. No sentence in this article claims that these bindings measurably trigger anything in your body.
- Non-monotonic curves are contested. They are mechanistically plausible and have been observed repeatedly in animal experiments. In regulatory toxicology they are assessed differently to this day, and the most important reviews on them come from author groups with a clear position in the debate.
- The low-dose effects from CLARITY-BPA are rat data. The dose figures of 2.5, 25 and 250 micrograms per kilogram of body weight apply to Sprague-Dawley rats and not to humans.
- The EFSA derivation is not a consensus. On 19 April 2023 the BfR proposed a guidance value a factor of 1000 above it, and the EMA raised objections. This article does not judge which authority is right, because I cannot decide that.
- PCOS and BPA: association, not cause. The cross-section cited measured in serum, was a snapshot, and the author group itself describes a possible interaction in both directions. No cause and no treatment recommendation follow from these data.
- The umbrella review rests on reviews of moderate quality. The authors state themselves that 78 percent of the meta-analyses included come from reviews of moderate methodological quality. For the substitutes, such meta-analyses do not exist at all.
- The mortality figure is a cohort with a wide interval. The hazard ratio of 1.49 has a confidence interval of 1.01 to 2.19, and the two cause-specific results were not significant. It is deliberately quoted with its interval in this article.
- The single measurement is a methodological problem. With a half-life in the range of hours, a single urine value mainly reflects the last few hours. This weakens almost all observational studies in this field, and it is rarely mentioned.
- The dermal calculation rests on six people. The share of 51.9 to 84 percent comes from a simulation study with six volunteers. The 48-hour finding there stands in tension with the controlled oral kinetics study. This tension is named in the text and not resolved.
- BPS and BPF: cell culture, hardly any human data. Of 32 studies in the systematic review, 25 were purely in vitro. On disease consequences of these substitutes in humans there is little available. On bisphenol AF there is even less, which is why only the name appears here.
- The reduction measures are measured against exposure, not against health. All three intervention studies have urinary concentrations as their endpoint. No sentence in this article claims that better hormone values or fewer symptoms follow from that.
- No guideline calls for BPA diagnostics. Neither the international PCOS guideline of 2023 nor the ESHRE endometriosis guideline of 2022 contains a measurement of bisphenols. The reason is understandable: an intervention study is missing that would show that lowering exposure improves a clinical endpoint.
- What deliberately does not appear here. No dosage recommendation, no treatment protocol, no product names, no source of supply for tests. And no advice to change, reduce or stop an existing medication, neither the pill nor hormone replacement therapy, a thyroid medication, metformin or an antiandrogen. Every adjustment belongs in medical care. From no section does it follow that a gynaecological work-up or an indicated operation should be postponed.
- On the topic of breast cancer, nothing appears here. A link is discussed mechanistically and in animal models, but in humans it is not robustly represented in the available meta-analyses. A hint here would have created more fear than the data can carry.