I was going to avoid blogging on this topic, but seeing as the story made the Australian with the headline “Chemicals in lipstick and cleaning products linked to early menopause”, I feel I have to weigh in a bit to avoid undue panic and the inevitable dangers of people hurling their lipsticks out the window at great speed. Also, there are issues of science communication and “the dose makes the poison”

To set the scene, there are a number of chemicals that are persistent pollutants in the environment such as polychlorinated biphenyl’s (PCB’s) and phthalates. These can weakly mimic estrogen or testosterone. As they are (relatively) easily absorbed and can accumulate in the body (Phthalates much less than PCBs), these chemicals may accumulate to levels that have adverse effects on human health (although there is no strong evidence they actually do).

Against this background the Australians' headline (and others like it) were generated by this paper just published in PLOS One, “Persistent Organic Pollutants and Early Menopause in U.S. Women”

This study looks at the occurrence of early menopause in a sample of US women with levels of a variety of persistent organic pollutants that can mimic estrogens’ effects. The study found that women who had high levels of PCB’s or phthalates in their urine, higher than 90% of women in the general community, went through menopause between 6 months to four years earlier than women in the general community (not 2-4 years as reported).

The amount menopause was shifted by, and the statistical strength of the association, varied quite a bit even in the same chemical class.

There are a couple of limitations to the paper. One is that I could not find if they had adjusted their statistical analysis for multiple comparisons, which will exaggerate the strength of the association. As well, it is not clear that for different chemicals were found in different urine samples, or if say, people with high PCB levels also had high Phthalates, which could cause spurious associations. Another is that the urine samples may not accurately reflect exposure when the women were going through menopause.

The levels of the chemical were in many cases measured years after menopause occurred.

When the researchers tried to control for the length of time the women had been exposed to these chemicals, the association disappeared for phthalates and some pesticides, and remained for PCB’s. While the associations for PCBs are suggestive, correlation studies suffer from the problem that other factors may be involved.

For example, during the 1952 polio outbreaks in the US, there was a strong correlation between polio and icecream sales, leading some to suggest the banning of ice cream. In fact it was simply that in summer more people were outside interacting.

In this case, there is some reasonable evidence that high exposure to PCBs could have a an effect on menopause, so an actual link is plausible. For phthalates, the link is less likely given the concentrations that cause estrogen mimicking effects are relatively high, and that the link goes away when you control for length of exposure.

However, the headlines and stories concentrated on things like lipstick and makeup, cleaning products and food containers. Things that contribute to phthalate exposure, not PCB exposure. Remember that the association with phthalates went away when corrected for length of exposure to the compounds.

Also remember that the associations were seen in people who had concentrations higher than 90% of the US populations.

So what does this mean for Australians, should we be hurling away our lipstick and eye liner with great force? Not eating food wrapped in cling film?

Food Standards Australia and New Zealand surveyed foods for a variety of contaminants, and did not detect phthalates (or a number of other persistent organic pollutants). So food contributions to Phthalate exposure in Australia will be low.

What about make up? There are a variety of phthalates at low levels in makeup. Different phthalates are used as plasticizers in nail polish and hair sprays, and as a solvent in perfumes and such. Studies looking at typical make up and cosmetic product use show that these phthalates are absorbed and excreted, but that the levels of exposure to these compounds is around 10 fold lower than the recommended total daily intake of the compounds.

The amount absorbed depends on how much you use, but even using 11-12 personal care products (lotion, perfumes, lipstick etc.) only doubles the amount of metabolites excreted, suggesting exposure is still well under the recommended levels.

Now, there is a LOT of variability in how people absorb and break down phthalates, but all the data we have suggests that standard use of cosmetics and food consumption is not the source of the phthalates for that 10% of women with the highest phthalate levels. And again remember that the phthalate association went away when length of exposure was corrected for.

Yet this is the angle most newspaper went after, and the PCBs, which were the chemicals with the most consistent association after correcting for exposure, were largely ignored. PCB’s are banned in Australia, but people may still be exposed to PCBs through contaminated land sites, industrial exposure (working with old electrical equipment that contains PCBs, and through food that has been contaminated though bioaccumulation.

Fish is the most likely source of PCB exposure, although in Australia the levels are generally low.

Given the association of PCBs with early menopause these results should be carefully considered with a view to reducing PCB exposure in people with the highest levels of PCB’s. This is already occurring to some extent, the latest US National Health and Nutrition Examination Survey shows that exposures to a variety of persistent organic pollutants have fallen (see http://www.cdc.gov/exposurereport/pdf/fourthreport_updatedtables_aug2014.pdf#page=7&zoom=auto,-265,378 WARNING big file).

Including phthalates, levels of phthalates have fallen by around 45% from the levels examined in this study (they looked at menopause and urinary chemical levels from 1999-2008, levels of phthalates have dropped dramatically since then).

So what about lipstick then? While the newspaper articles gave the impression that ordinary personal care product use may put you at risk, the earlier menopause was seen only in women with the highest levels of personal care products, and the association for phthalates, the ones that actually occur in personal care products, went away when corrected for exposure puts a different complexion on things.

Obviously you do not need to hurl your lipstick (or other personal care products) away with great force. But still, you may wish to consider using fewer (12 is probably a bit much). And eating fresh food, especially fresh fruit and vegetables, is always a good idea.

Ian Musgrave does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations. This article was originally published on The Conversation. Read the original article.