Phthalates: Where They Hide and How to Avoid Them

So what are phthalates? They're a family of plasticisersChemicals added to rigid materials (usually PVC) to make them soft and flexible. Without plasticisers, your shower curtain would be as rigid as a drainpipe. found in everything from vinyl flooring to IV bags to the perfume counter — and they don't work the way most people assume. They don't mimic estrogen. They suppress testosterone. The mechanism is different, the affected pathways are different, and the most-studied phthalate — DEHPDi(2-ethylhexyl) phthalate — the highest-volume phthalate, used in PVC, food packaging, and medical devices. Listed as a Substance of Very High Concern under EU REACH. — may not be the one doing the most harm. Our endocrine disruptors guide covers the broader class of hormone-interfering chemicals. This article covers the family that hides in plain sight.

Phthalates are a family of synthetic chemicals — esters of phthalic acid — used primarily as plasticisers that make rigid PVCPolyvinyl chloride — the third most produced plastic globally. Rigid PVC is used in pipes and window frames. Flexible PVC — shower curtains, flooring, cable insulation — requires plasticisers, which is where phthalates come in. soft and flexible, and as carriers that help fragrance compounds bind to skin and last longer. Over 8 million tonnes of phthalates produced globally per year — roughly the weight of 40,000 blue whales are produced globally per year. They are not one chemical. Different phthalates have different uses, different exposure routes, and different risk profiles — which is why the research gets misread when reporters treat the family as a single substance.

The 'fragrance' loophole deserves its own paragraph. The IFRAInternational Fragrance Association — the fragrance industry's self-regulatory body. Publishes the Transparency List of ingredients used in commercial fragrances. Transparency List catalogues 3,691 fragrance ingredients in commercial use (2025 edition) — 3,312 fragrance + 379 functional ingredients in commercial use. Under US law, none of them require individual disclosure on a label — the word 'fragrance' or 'parfum' covers the lot. The EU historically required naming 26 allergens above threshold; Regulation (EU) 2023/1545 expanded this to over 80 substances, with compliance deadlines rolling through 2026 and 2028. The UK retains the original 26 under retained EU law. The NRDCNatural Resources Defense Council — a US environmental advocacy organisation. tested 14 common air fresheners in 2007 and found phthalates in 86% of air fresheners tested — including products labelled 'all-natural' and 'unscented' — including products labelled 'all-natural' and 'unscented.' None listed phthalates on the label. One product contained 7,300 ppm of DEP. The word 'fragrance' is a locked door with 3,691 chemicals behind it. The US doesn't require a key.

Phthalates suppress fetal testosterone production — 10.2× the odds of shortened anogenital distance in boys with highest prenatal MBP exposure (Swan 2005) higher odds of altered genital development in boys whose mothers had the highest exposure. They are anti-androgenicA substance that reduces the production or action of androgens (male sex hormones, primarily testosterone). Anti-androgens work differently from estrogen mimics: instead of adding a false signal, they remove a real one., not estrogenic, and the distinction matters. Most endocrine disruptors in the Eso World library — BPA, parabens, triclosan — mimic estrogen by fitting into estrogen receptor binding pockets. Phthalates don't touch the estrogen receptor. They work upstream: in fetal Leydig cellsThe cells in the testes responsible for producing testosterone. During fetal development, Leydig cell testosterone output directs the formation of male reproductive anatomy — if the signal is insufficient, development is altered., they interfere with the enzymes and transport proteins needed to synthesise testosterone in the first place. Less testosterone is produced during the masculinisation programming windowA critical period during fetal development (roughly weeks 8-14 in humans) when testosterone directs the formation of male reproductive anatomy. Insufficient testosterone during this window produces measurable changes in genital development., and the developmental consequences show up at birth.

The consequences were measured directly in 2005. Eighty-five pregnant women in the Study for Future Families provided urine samples during pregnancy. After their sons were born, researchers measured anogenital distanceThe distance between the anus and the genitals — a standard marker of how much testosterone signalling occurred during fetal development. Shorter distance in boys indicates reduced androgen exposure in the womb. Used routinely in animal toxicology and increasingly in human studies. — a developmental marker that reflects how much testosterone reached the tissue during the masculinisation window. Mothers in the highest quartile of MBPMono-n-butyl phthalate — the urinary metabolite of dibutyl phthalate (DBP). One of the phthalate metabolites most strongly associated with reproductive developmental effects. — a metabolite of DBP, the phthalate in nail polish and adhesives — had sons with 10.2 times the odds of a shorter-than-expected anogenital distance in boys with the highest prenatal MBP exposure the odds of a shorter-than-expected anogenital index Swan et al. 2005.

Here's the part that almost every summary of this study gets wrong. The phthalate everyone talks about — DEHP, the high-volume plasticiser in food packaging and PVC — showed no association whatsoever with reduced anogenital distance. Its metabolite MEHPMono(2-ethylhexyl) phthalate — the primary urinary metabolite of DEHP. Despite DEHP's regulatory prominence, MEHP showed no significant association with AGD in the Swan 2005 study (p = 0.833). came in at p = 0.833. The metabolites that did show significant effects were MBP (from DBP, p = 0.031), MiBPMonoisobutyl phthalate — urinary metabolite of diisobutyl phthalate (DiBP). Showed the strongest association with reduced AGD in Swan 2005 (p = 0.007). (from DiBP, p = 0.007 — the strongest signal in the study), and MEPMonoethyl phthalate — urinary metabolite of diethyl phthalate (DEP), the fragrance carrier phthalate. (from DEP — the fragrance carrier, p = 0.017). The most-regulated phthalate showed nothing. The ones in your shampoo showed everything.

A dose-response relationship between MBP and low sperm concentration emerged in 463 men attending a Boston fertility clinic — the odds of low concentration climbed with each exposure quartile Hauser et al. 2006. A meta-analysis pooling 12 epidemiological studies confirmed MBP and MBzP as the metabolites most consistently associated with reduced sperm concentration, with pooled odds ratios of 2.19 and 1.88 respectively Wang et al. 2023. The direction is consistent. The effect sizes are modest. And once again, DEHP metabolites were not the strongest signal.

The developmental evidence extends beyond reproductive anatomy — to the brain. Three hundred and twenty-eight pregnant women in the Columbia University cohort provided third-trimester urine samples. Their children were followed through age seven. At that point, researchers sat each child down for a standardised IQ test — the Wechsler — and matched the score against the phthalate metabolites measured in the mother's urine years earlier. Children born to mothers in the highest quartile of DnBP and DiBP exposure scored 6.6 to 7.6 IQ points lower at age 7 in children with highest prenatal phthalate exposure — roughly the difference between the 50th and 37th percentile lower than those in the lowest quartile — roughly the gap between the 50th and the 37th percentile Factor-Litvak et al. 2014. The same metabolites as Swan. The same consumer products. A different endpoint, pointing in the same direction.

Phthalate exposure during pregnancy is associated with roughly 33% higher odds of preterm birth per ln-unit increase in maternal urinary DEHP metabolites (Ferguson 2014) of preterm birth, with the strongest signal in spontaneous preterm birthPreterm birth (before 37 weeks gestation) that begins without medical induction or planned cesarean — typically driven by spontaneous labour or premature rupture of membranes. The endpoint researchers focus on because it most likely reflects a biological insult rather than clinical decision-making. subgroups. The same anti-androgenic class shows up in adult outcomes too — gestational diabetes, antral follicle count, pregnancy loss — and the metabolites driving the female-side signals are not always the same ones that drove the male AGD findings.

The preterm-birth study deserves walking through. Researchers at Brigham and Women's Hospital in Boston enrolled pregnant women into a cohort and collected urine samples up to three times during each pregnancy. Of 482 women followed to delivery, 130 went on to have preterm births — birth before 37 weeks — and 352 carried to term. The researchers measured phthalate metabolites in the stored urine and compared the two groups. Per ln-unit increase in ΣDEHPThe sum of four DEHP metabolites — MEHP, MEHHP, MEOHP, MECPP — used as a single exposure marker for di(2-ethylhexyl) phthalate. Standard biomarker in epidemiological studies because individual metabolite measurements track each other closely. (the sum of four DEHP metabolites), the odds of preterm birth rose 33% (OR 1.33, 95% CI 1.04-1.70). For the spontaneous preterm subset — births most likely driven by biology rather than clinical decision-making — the odds rose 63% (OR 1.63, 95% CI 1.15-2.31), with MBP showing a 49% increase (OR 1.49, 95% CI 1.08-2.06) Ferguson et al. 2014. A pregnancy that ends three weeks early is not subtle. The exposures that pushed it there were measured in nanograms.

Gestational diabetes follows the same pattern but with a different metabolite. Seven hundred and five pregnant women across four US sites in the TIDESThe Infant Development and Environment Study — a multi-site US prospective pregnancy cohort tracking environmental exposures and birth outcomes. study had urinary phthalate metabolites measured in early and late pregnancy and were tracked for gestational diabetesA condition in which a woman without prior diabetes develops high blood sugar during pregnancy. Linked to elevated risks of preeclampsia, macrosomia, neonatal hypoglycaemia, and later type 2 diabetes for the mother.. Per IQR increase in pregnancy-average MEP — the metabolite of DEP, the fragrance-carrier phthalate — the odds of gestational diabetes rose 61% (OR 1.61, 95% CI 1.10-2.36) Shaffer et al. 2019. The same metabolite the Swan 2005 boys' AGD signal pointed at. The same chemical that hides behind 'fragrance' on the label.

Pregnancy loss and ovarian function complete the picture. A prospective cohort of 3,220 Chinese women in Anhui province documented increased odds of clinical pregnancy loss with higher MEP, MBP and DEHP-metabolite concentrations in early-pregnancy urine, with MEP and the MEHHPMono(2-ethyl-5-hydroxyhexyl) phthalate — a DEHP metabolite. Used together with MEHP and MEOHP as ΣDEHP markers in biomonitoring./MEOHP DEHP metabolites particularly associated with embryonic loss before 10 weeks Gao et al. 2017. In the EARTH cohort of 215 women attending a Boston fertility clinic, those in the second quartile of ΣDEHP exposure had 24% lower antral follicle count than women with the lowest DEHP exposure (Messerlian 2016, n=215, EARTH cohort) — a marker of ovarian reserve than the lowest-exposure quartile (95% CI -32%, -16%) Messerlian et al. 2016. Antral follicle countAn ultrasound count of small follicles visible in the ovaries. The standard marker of ovarian reserve — how many eggs a woman has remaining and how she's likely to respond to fertility treatment. is one of the strongest single markers of ovarian reserve, and the difference here was measured at consumer-level exposure, not occupational.

Urinary phthalate metabolites are detected in nearly every American the CDC tests — MEP, MBP and MBzP show up in over 95% of NHANES participants aged 6 and over. The bigger story is what the trend looks like over twenty years of biomonitoring. Pooling 11,071 NHANES participants across the 2001-2002 through 2009-2010 cycles, Zota and colleagues tracked the geometric-mean concentrations of every phthalate metabolite the CDC measures. The phthalates that got regulatory attention went down — MEP fell 42%, MBzP (from BBP) fell 32%, ΣDEHP fell 37% — and the ones that didn't went up. MiBPMonoisobutyl phthalate — urinary metabolite of DiBP, the phthalate that replaced DBP in many products after CPSIA 2008. (the metabolite of US Less restricted than DBP 2008 DiBP, which replaced DBP in many adhesives and personal care products after the 2008 children's-toy ban) rose 206% over the same window Zota et al. 2014. Same family, same anti-androgenic profile, different name on the label.

The downstream cost is now measurable in mortality data. Trasande and colleagues linked phthalate metabolite measurements from 5,303 NHANES adults aged 20 and over to the National Death Index through 31 December 2015. Per IQR increase in high-molecular-weight phthalate metabolites, all-cause mortality rose 14% (HR 1.14, 95% CI 1.06-1.23). Adults in the highest tertile of HMW phthalate exposure had 48% higher all-cause mortality than adults in the lowest tertile, NHANES adults 20+ followed for ~10 years (Trasande 2022, n=5,303) than the lowest tertile (HR 1.48, 95% CI 1.19-1.86). The cardiovascular signal tracked the DEHP-metabolite MEOHP specifically — tertile-3 cardiovascular HR 1.74 (95% CI 1.05-2.88). The authors estimated 90,761 to 107,283 attributable deaths per year among Americans aged 55-64, costing $39.9-47.1 billion in lost economic productivity Trasande et al. 2022. A follow-up extrapolation to global mortality data attributed 356,238 cardiovascular deaths to DEHP exposure worldwide in 2018 — concentrated in South Asia and the Middle East, where production has shifted as Western jurisdictions have restricted use Hyman et al. 2025.

The EU moved first and furthest. REACHRegistration, Evaluation, Authorisation and Restriction of Chemicals — the EU's chemical safety framework. Annex XVII lists restricted substances; Annex XIV lists substances requiring authorisation. Annex XVII Entry 51, in force since July 2020, restricts DEHP, DBP, BBP and DIBPDiisobutyl phthalate — a lower-volume phthalate used in adhesives, printing inks, and some personal care products. Listed as SVHC under REACH alongside DEHP. to a combined maximum of 0.1% by weight in any plasticised material in any consumer article — the EU's class-level phthalate restriction by weight in the plasticised material of any consumer article — not just toys, not just food contact, any article. All four are listed as Substances of Very High ConcernUnder EU REACH, substances that meet criteria for carcinogenicity, mutagenicity, reproductive toxicity, persistence, or endocrine disruption. SVHC listing triggers authorisation requirements — manufacturers must apply for permission to continue using them. under Annex XIV.

The food-contact side moved next. EFSA's 2019 Panel on Food Contact Materials reassessed the four highest-volume phthalates as a group, on the basis that they share a plausible common mechanism — reduction in fetal testosterone, and set a single group-TDIGroup tolerable daily intake — a combined safe-dose limit applied to multiple chemicals that share a common mechanism of toxicity. EFSA's group-TDI for phthalates is expressed as DEHP equivalents. of 50 µg/kg bw per daythe EU group tolerable daily intake covering DBP, BBP, DEHP and DINP combined, expressed as DEHP equivalents (EFSA 2019) for DBP, BBP, DEHP and DINP combined, expressed as DEHP equivalents EFSA Panel CEP 2019. The grouping itself was the news — EU regulators stopped pretending the four chemicals could be assessed one at a time.

The US restricts phthalates in children's products only. The CPSIAConsumer Product Safety Improvement Act — US 2008 legislation that, expanded by CPSC final rule in 2017 (82 FR 49938), permanently banned eight phthalates from children's toys and childcare articles at concentrations above 0.1%. (2008, expanded 2017) permanently bans eight phthalates in children's toys and childcare articles at above 0.1%. No equivalent federal restriction exists for adult consumer products, cosmetics, or food contact materials. The FDAUS Food and Drug Administration US Denied food-contact ban 2022 a 2016 citizen petition from EDF, NRDC, Breast Cancer Prevention Partners and others to revoke food-contact authorisations for 28 ortho-phthalates in May 2022, and overruled subsequent objections in October 2024 (89 FR 86290). Phthalates remain authorised in food packaging in the US as of 2026.

California's Prop 65Proposition 65 — California Safe Drinking Water and Toxic Enforcement Act 1986. Requires warning labels on products sold in California that contain chemicals known to the state to cause cancer, birth defects or reproductive toxicity. list now covers six ortho-phthalates: DEHP (cancer + developmental + male reproductive, listed 1988/2003), BBP, DBP and DnHP (developmental and/or reproductive, listed December 2005), DIDP (developmental, listed 2007), and DINP (cancer, listed 2013). The bigger US shift came in January 2026: EPAEnvironmental Protection Agency published final TSCAToxic Substances Control Act — the US chemicals law. Section 6 risk evaluations determine whether a chemical poses unreasonable risk; an unreasonable-risk finding requires EPA to initiate risk-management rulemaking. risk evaluations finding that five phthalates — BBP, DBP, DCHP, DEHP and DIBP — pose unreasonable risk to human health (91 FR 373, 6 January 2026). Under TSCA, that finding requires EPA to initiate risk-management rulemaking — not authorise continued use, restrict it. The first US federal phthalate restrictions for adult products are now legally required, even if not yet drafted.

Maybe — and the honest answer is the safety data isn't there yet. As DEHP fell out of favour, manufacturers switched to two main replacements: DINCHHexamoll DINCH — di(isononyl)cyclohexane-1,2-dicarboxylate, a non-phthalate plasticiser introduced commercially by BASF in 2002. Used in toys, food contact materials, medical devices and personal care. (a non-aromatic cyclohexane diester now used in toys, food contact and medical devices) and DEHTPDi(2-ethylhexyl) terephthalate, also called DEHT or DOTP — a non-ortho terephthalate plasticiser used as a DEHP replacement in food packaging, vinyl flooring and medical tubing. (a terephthalate analogue of DEHP). DINCH has limited reproductive-toxicity testing — a 2016 review of twenty alternative plasticisers concluded that body burden of DINCH and DEHTP has been rising in human biomonitoring while The papers do not say the replacements are toxic. They say the data needed to know whether they're toxic doesn't exist yet — which is exactly the position we were in with DEHP and DBP forty years ago. the toxicology data needed to evaluate endocrine disruption is not available Bui et al. 2016. CDC's 2019 analysis of NHANES 2015-2016 was the first US population-level biomonitoring of DEHTP metabolites and documented that DEHTP exposure is already widespread in the US population aged three and older — the substitution-driven uptake that began as DEHP fell out of favour Silva et al. 2019.

The PVC industry didn't stop using plasticisers — it changed which plasticisers it uses. Whether the replacements turn out to be safer than the originals is the question that takes thirty years of human studies to answer, and we are eight to ten years into the DINCH and DEHTP eras. The cautious read: phthalate-free is better than phthalate-laden when the regulatory pressure is on the phthalate, but a 'phthalate-free' label tells you which class of chemical isn't in the product. It does not tell you whether the substitute has been tested for the same endpoints. Glass, stainless steel, and unflavoured uncoloured silicone avoid the question entirely.

Three days. That's how fast your phthalate levels respond to product changes — because unlike PFAS, phthalates don't accumulate. Their biological half-lives are measured in hours, which means your body burden reflects what you were exposed to yesterday, not last year. Five families who switched to fresh food for three days dropped DEHP metabolites by 53-56% in 3 days of eating fresh food instead of canned or plastic-packaged Rudel et al. 2011. A hundred Latina adolescents who switched to labelled-free personal care products for three days dropped MEP — the fragrance-carrier metabolite — by 27% in 3 days of switching personal care products in the HERMOSA study Harley et al. 2016. When both groups went back to their normal products, levels rebounded within days. You're not stuck with what's in your body. You're stuck with what's in your bathroom.

The highest-leverage single change is switching personal care products — not because DEHP in food packaging doesn't matter, but because the Swan 2005 data showed the strongest developmental signals came from the metabolites of DEP, DBP and DiBP, which are the phthalates in fragrance, nail polish and cosmetics. For DEHP specifically — the food-contact and medical-device exposure most people associate with the word 'phthalate' — the DEHP profile walks through the regrettable-substitution pattern that produced DINCH and DEHTP.

Go back to the bottle on the bathroom shelf. The word 'fragrance' is still there — still one word, still covering thousands of possible ingredients, still not required to tell you which ones. The research that should have changed the label was published in 2005. The phthalate metabolites that showed the strongest developmental effects in boys — MiBP, MEP, MBP — are the ones in personal care products, not the ones in food packaging that get most of the regulatory attention. The intervention studies show your levels drop in three days when you switch products. Three days, roughly the time it takes to finish the bottle you already opened and start the one that doesn't say 'fragrance' on the back.

The label won't change by Thursday. What's in your bathroom can.