Cosmetics and Fertility — What Does the Science Really Say About Phthalates and Parabens?

In 2023, The Lancet published a series of articles on declining fertility rates in industrialised countries. Headlines linking the trend to cosmetics appeared almost immediately. Can substances found in everyday cosmetic products really affect reproductive capacity? The answer requires working through a dense thicket of data — not shortcuts.

Phthalates in research on male fertility

The most frequently cited work in this field is Shanna Swan's 2005 study (Environmental Health Perspectives), which found a correlation between the concentration of phthalate metabolites (MEP, MBP, MBzP) in mothers' urine and a shortened anogenital distance (AGD) in male newborns. AGD is a marker of androgen exposure in foetal life — a shorter distance suggests an anti-androgenic effect.

This finding was confirmed in several subsequent cohorts (Marsee et al., 2006; Suzuki et al., 2012), but with important caveats. First, the correlation involved longer-chain phthalates (DEHP, DBP) — the very ones that are banned in cosmetics in the EU. Second, AGD is an indirect marker — a shortened AGD does not automatically mean fertility problems in adulthood. A prospective study by Eisenberg et al. (2021, Fertility and Sterility) suggested a link between AGD and semen quality, but with a small sample size (n=185). Check the details in our ingredient encyclopaedia.

DEP — the only phthalate widely used in cosmetics — shows much weaker anti-androgenic activity. A systematic review by Radke et al. (2019, Environment International) rated the evidence for an effect of DEP on semen parameters as "insufficient to draw conclusions".

Parabens — preservatives under the scientific microscope

Methylparaben and ethylparaben are among the most commonly used preservatives in cosmetics. Their estrogenic activity was first described by Routledge et al. in 1998 (Toxicology and Applied Pharmacology). The scale, however, is key: the estrogenic activity of methylparaben is about 1/2,500,000 that of estradiol. For butylparaben the ratio is roughly 1/10,000 — higher, but still extremely low.

In its 2013 opinion (SCCS/1514/13), the SCCS deemed methylparaben and ethylparaben safe at concentrations up to 0.4% (individually) or 0.8% (combined). For propyl- and butylparaben the permitted concentration was lowered to 0.14% (combined) due to their higher estrogenic activity and incomplete data on use in the nappy area in children under three years of age.

A study by Kolatorova et al. (2018, Environment International) measured paraben concentrations in the blood of 100 Czech women and showed that, after discontinuing paraben-containing cosmetics, concentrations fell to undetectable levels within 24-72 hours — parabens are rapidly metabolised and eliminated, and do not accumulate in the body.

Biomonitoring — what the presence of a substance in urine does not tell us

Biomonitoring programmes (NHANES in the USA, GerES in Germany, HBM4EU in Europe) routinely detect phthalate and paraben metabolites in the urine of the general population. These results are often presented as proof that the body is "contaminated". Yet the detectability of a substance in urine means only that exposure occurred — it says nothing about whether the dose was harmful.

An analogy: caffeine is detectable in the urine of practically every adult European. No one, however, concludes from this that the population is "poisoned" by caffeine. What matters is the concentration relative to the threshold of biological action.

Additional limitations of biomonitoring:

• A snapshot measurement — a single urine sample does not reflect chronic exposure; concentrations can fluctuate by as much as tenfold over the course of a day (Preau et al., 2010).
• Unknown source — a metabolite in urine does not indicate whether it comes from a cosmetic, food, household dust or some other source.
• No clinical context — a numerical result alone, without reference to a biological-effect threshold, carries no health significance.

The "cocktail effect" — a hypothesis that needs careful assessment

One of the most serious arguments in the EDC debate is the hypothesis of combined action — the so-called cocktail effect, or mixture toxicity. The concept holds that even if individual substances occur at safe concentrations, their combined action may exceed the safety threshold.

A study by Kortenkamp et al. (2007, Environmental Health Perspectives) showed under in vitro conditions that a mixture of 8 substances with estrogenic activity, each present at a concentration below the NOAEL, produced a measurable estrogenic effect. This is an important finding that has shaped European regulatory discourse.

Extrapolating from in vitro studies to real-world exposure conditions is difficult, however. The EDCMET project (2019-2023, funded by Horizon 2020) developed PBPK (physiologically-based pharmacokinetic) models to simulate combined exposure, but the results do not clearly indicate that typical cosmetic exposure exceeds safe thresholds, even in a mixture model.

Since 2012 the SCCS has taken aggregate exposure (combined exposure from various cosmetic products) into account in its opinions, which represents significant progress compared with the earlier assessment of individual products in isolation.

The precautionary principle vs. the evidence-based approach

Two paradigms clash in the debate over cosmetics and fertility. The precautionary principle states: if there is scientific evidence pointing to a possible hazard, protective measures should be taken even without full proof. The evidence-based approach requires the demonstration of causality before restrictions are introduced.

EU law strikes a balance between these approaches. The Cosmetics Regulation 1223/2009 bans CMR substances (carcinogenic, mutagenic, reprotoxic) of categories 1A and 1B, while permitting substances with potential endocrine activity provided an SCCS assessment confirms their safety under normal conditions of use.

As a consumer, you have the right to choose the strategy that matches your own risk profile. If you are planning a pregnancy and prefer to minimise exposure in line with the precautionary principle, that is a rational decision. If you trust SCCS assessments and use products that comply with the regulations, that too is a justified stance.

To assess the composition of your cosmetics consciously, it is worth learning to read the INCI ingredient list and to understand what the individual names mean.

Conclusions — what we know and what we don't

What we know: some substances found in cosmetics show endocrine activity in vitro. The most active phthalates (DEHP, DBP) and isopropyl-/isobutylparaben are banned or restricted in the EU. Cosmetic exposure to permitted substances falls within the SCCS safety margins.

What we don't know: exactly what combined exposure from all sources (cosmetics + food + environment) looks like. Whether there are sensitive developmental windows in which even low doses matter. How the cocktail effect translates into real-world conditions of human exposure.

What you can do: check the composition of your everyday products, identify the substances that raise your individual concerns, and make an informed decision — without panic, but also without naivety.

Check your cosmetic's ingredients →