What Is BPA? Where It's Found and What the Research Shows

Bisphenol A is a synthetic chemical first tested as a hormone drug in the 1930s, shelved when better options came along, and pulled back out in 1958 when Bayer began commercial production of polycarbonate plastic. It's now produced at roughly 10 million tonnes of BPA produced globally per year a year — in can linings, polycarbonate water bottles, thermal receipt paper, dental sealants, and a long tail of food packaging. CDC biomonitoring picked it up in 92.6% of Americans had detectable BPA in CDC NHANES 2003-2004 of the US population in 2003–2004 (Calafat et al. 2008), and comparable detection rates have held in every NHANES cycle since, even as the absolute concentrations have dropped 40–60% from their peak. The failed hormone drug is in roughly nine out of ten people you meet.

BPABisphenol A — a synthetic organic compound used as the monomer for polycarbonate plastics and epoxy resins. is two phenol rings joined by a propane bridge — the compound C₁₅H₁₆O₂. That's the whole story at a structural level, and it's the part that matters biologically. The two rings happen to carry hydroxyl groups at almost exactly the same spacing as the hydroxyls on estradiol, the body's primary natural estrogen. Estrogen receptors evolved to recognise that specific spacing. They can't tell the difference between estradiol and BPA on shape alone.

That similarity has a twist. When Kuiper and colleagues measured BPA's binding affinity at estrogen receptor beta in 1998, it came in at thousands of times weaker than estradiol itself Kuiper et al. 1998. Which sounds reassuring, and industry has been citing it for twenty-five years. The next part gets cited less. Subsequent work showed BPA also acts through membrane-initiated estrogen signalling — a separate pathway that responds at far lower concentrations than the nuclear receptor, and that conventional toxicology testing wasn't designed to look at.

The 'weak estrogen' story is true of one pathway and misleading about the other.

BPA lives wherever polycarbonate, epoxy or thermal printing lives — which is more places than you'd guess from any single label. The big sources:

Thermal receipts deserve a special mention. On polycarbonate, BPA is chemically bonded into the polymer chain — it has to leach to reach you. On a receipt there's no bond at all. The BPA sits on top of the paper as a free, transferable coating that rubs off on whatever touches it.

The experiment that quantified this was deceptively simple. Participants rubbed their hands with ordinary alcohol hand sanitiser — the kind at every supermarket till — then picked up a thermal receipt, held it for a few seconds, and ate a portion of French fries with their fingers. Every step unremarkable. The dermal BPA absorbed through their skin jumped up to 185× the dry-handed baseline Hormann et al. 2014. The alcohol had stripped the skin's lipid barrier, and the BPA on the paper flooded through.

It isn't an exposure event you'd think to monitor. It is one anyway.

The dermal route is worse than the oral one in a way regulatory testing doesn't capture. Sasso and colleagues dosed volunteers with deuterium-labelled BPA and measured what actually reached circulation in its bioactive, unconjugated form. After dermal exposure, 8.81% of serum BPA was bioactive after dermal exposure of the serum BPA stayed bioactive. After swallowing the same dose, only 0.56% of serum BPA was bioactive after oral exposure did Sasso et al. 2020.

Sixteen times more active chemical through the skin than the gut.

Standard safety testing assumes the oral route — ingestion, then first-pass liver metabolism, then conjugation into a mostly inactive form. Receipts and water bottles don't honour that assumption.

BPA is the most-studied endocrine disruptor in the scientific literature — thousands of peer-reviewed papers across four decades. None of the individual associations are conclusive in the way regulators traditionally demand. The pattern across them is what makes the field take BPA seriously. The strongest signals cluster around dose-response anomalies, reproductive outcomes, metabolic effects, and developmental programming.

Start with the dose. Standard regulatory testing uses three dose points, typically at high concentrations — the assumption being that if you see no effect at the top of the range you're safe at the bottom. Vandenberg's 2014 review of the BPA literature found that 23.6% of BPA experiments showed inverted U-curve dose responses of experiments showed non-monotonic responses, peaking at low doses and declining at high ones Vandenberg 2014. Detecting a curve that bends like that needs an average of 6.9 dose points. Three can't find it.

EFSA confronted exactly this in 2023. After an extensive review weighing immune-system effects heavily, the agency cut BPA's tolerable daily intake from 4 micrograms per kilogram per day to 0.2 nanograms.

Twenty thousand times lower. Not twenty percent. Not double. The previous number had been the official safe dose for more than a decade, and millions of products were approved under it.

The reproductive signals are consistent across both sexes. In men attending a fertility clinic, an interquartile-range increase in urinary BPA tracked with 23% lower sperm concentration, 13% lower normal morphology, and 10% higher sperm DNA damage Meeker et al. 2010 — a pattern confirmed a decade later by a pooled analysis of 18 epidemiological studies Castellini et al. 2024. Women with polycystic ovary syndrome show notably elevated serum BPA compared to controls, correlating with testosterone and insulin resistance markers (Kandaraki 2011, Akın 2015) Kandaraki et al. 2011.

In pregnancy the picture narrows but sharpens. Ehrlich and colleagues found a dose-response relationship between urinary BPA and implantation failure during IVF. Lathi and colleagues followed women through early pregnancy and found those in the top quartile of conjugated serum BPA had 83% greater miscarriage risk in the highest BPA quartile greater miscarriage risk than the lowest quartile Lathi et al. 2014. Our guide to chemicals and fertility goes deeper.

On cancer specifically, the picture is genuinely unsettled. BPA has not been formally evaluated by IARCInternational Agency for Research on Cancer — the WHO body that classifies substances as Group 1 (carcinogenic), 2A (probable), 2B (possible), or 3 (not classifiable). — the WHO reviewed its carcinogenicity in 2010 and found the evidence insufficient at the time, and the substance now sits on the IARC Monographs 2025–2029 priority list waiting for its turn. Its endocrine-disrupting status, on the other hand, is settled: ECHAEuropean Chemicals Agency — the EU agency responsible for chemical regulation under REACH., the Endocrine Society and EFSA all classify BPA as a reproductive toxicant and endocrine disruptor.

Canada was first. In October 2008 it declared BPA a 'toxic substance' and banned it from baby bottles — the first country in the world to do either. The EU followed in 2011 (Directive 2011/8/EU, baby bottles only) and extended the ban to food packaging for children under three in 2018. The US FDA banned BPA from baby bottles and sippy cups in 2012 and from infant formula packaging in 2013 — both decisions essentially codified moves industry had already made quietly to avoid bad press.

The agency that moved the ground underneath the numbers was EFSAEuropean Food Safety Authority — the EU agency responsible for assessing risks in the food chain.. Its 2023 review — the 20,000-fold cut already mentioned — had a quieter follow-on finding that tends to get buried under the headline. With the new limit in place, EFSA found that every age group, from infants to adults, now exceeds the new safe dose through diet alone, by two to three orders of magnitude.

The most recent move is comprehensive. Regulation (EU) 2024/3190 was adopted in December 2024 and entered force in January 2025, with phased compliance running through 2026–2029. It bans BPA from food contact materials — and, crucially, the substitutes too. BPS, BPAF, TBBPA, phenolphthalein: all captured under the same class-based logic. The entire 'BPA-free' marketing premise had been built on the idea that the substitutes were okay. The EU now disagrees. The UK hasn't adopted equivalent restrictions post-Brexit, and the US has nothing comparable beyond the baby-bottle and infant-formula scopes.

Here's the practically useful part. BPA's half-life in the human body is roughly 4–6 hours — short enough that your body burden reflects what you've been exposed to in the last day, not the last year. Rudel and colleagues tested that the simplest way possible: five families switched from their normal diet to a fresh-food diet (nothing from cans, nothing in plastic packaging) for three days. Urinary BPA dropped by 66% drop in urinary BPA after 3 days on a fresh food diet in three days Rudel et al. 2011. When they went back to normal food, levels rebounded within days.

It's a daily input. What you buy this week lowers it next week.

'BPA-free' is the single biggest trap in this whole category. Yang et al. (2011) tested commercial plastic products — including those marketed as BPA-free — and found that almost all of them released chemicals with measurable estrogenic activity Yang et al. 2011. Rochester and Bolden's 2015 systematic review of 32 studies then showed the specific substitutes — BPS, BPF — have estrogenic, anti-estrogenic, androgenic and anti-androgenic activity comparable to BPA itself Rochester and Bolden 2015. Reininger and Oehlmann (2024) tested eleven common BPA analogues in an in-vitro battery under the title 'Regrettable substitution?' — the majority showed estrogenic or anti-androgenic effects comparable to, or greater than, BPA itself Reininger and Oehlmann 2024.

The label means they removed the one chemical you've heard of. It does not mean what replaced it was tested.

BPA is the textbook case of a chemical handled in slow motion. Estrogenic activity identified in 1936. Mass industrial production in 1958. First serious peer-reviewed concerns in the late 1990s. Canada's baby-bottle ban in 2008. EFSA's 20,000-fold revision in 2023 — sixty-five years after Bayer started scaling production. The EU's comprehensive food contact ban in force from 2025.

The research is settled enough that ECHA classifies BPA as a reproductive toxicant and an endocrine disruptor for both human health and the environment. The marketing is still catching up. The ingredient lists are on the cans in your cupboard, the studies are on PubMed, and the half-life is six hours.

Which means the tin you opened at the start of this article isn't a permanent fixture of your chemistry. It's an input. Change what you buy this week, and your body follows within days.