How to write a scare story. Or, how to mis-report science

Penis. There, I said it. Now that I have your attention, I'd like to write about, er, penises. More accurately, I want to point you in the direction of some stories which appeared in the press over the weekend and today (Monday, 1st September-not-April, 2008).

From the Daily Mail:

How perfumes and scented creams could make your unborn baby boy suffer infertility or even cancer

From the Metro (London edition of the freesheet):

Perfumes linked to infertility

From Scotland on Sunday:

Women warned not to wear perfume during pregnancy

Strong stuff, and all of the stories cite work carried out by Professor Richard Sharpe, a serious academic in a serious institution who publishes in serious academic journals. So it must be true - cosmetics cause infertility and cancer, right?


Perhaps not.

Reading Prof. Sharpe's recent publications, I found out many interesting things. Firstly, that he and his group measure penile length and testicle descent in rats. But let's get the sniggering out of the way (apologies to the Prof, 'cos as you'll soon see this is no laughing matter), because doing this enables him to pinpoint the time window when genital development is most sensitive to going awry. Embryology 101: male foetuses can develop abnormal genitalia (ranging from small to overtly female) if they're not exposed to enough testosterone (an androgen, or 'male-making-hormone) early on in pregnancy. If there's too little androgen around in this window, Sharpe's (and other labs') data reveal, many reproductive disorders can occur, and in some cases these are predictive of developing prostate cancer.

The latest studies, published in very respected peer-reviewed journals, show that blocking androgen signalling during pregnancy, in rats, causes testicular dysfunction, and the rats have very small penises. Fair enough. That is pretty much as far as his work goes, that is all he claims thus far. No mention of cosmetics. Not one. Now here's the intrigue.

Other studies have shown (and Prof. Sharpe's publications discuss the possibility) that compounds present in the air, largely as a result of chemical processes used in industrial production of plastics (such as some phthalates), can block androgen signalling. So, did we get a series of headlines decrying industrial plastic production? No. Instead, we got the above scaremongering about moisturisers and perfumes. There's no evidence (yet) to suggest that any chemical in creams and perfumes, applied topically, could get to a foetus in sufficient doses to cause the required damage. Still, to find out more, I did what many journalists that covered the story (Scotland on Sunday excepted) failed/couldn't be bothered to do - I got in touch with Prof. Sharpe.

He was not best pleased with the media's handling of his work. For a start, the stories make out that the findings are to be presented at a conference - whilst this is true, they are also peer-reviewed and published. More importantly, they really did get the wrong end of the stick. In Prof. Sharpe's own words (the following is taken from a statement he provided to me as an News-of-the-World stlye EXCLUSIVE!!!!!!111!!111ONE!!1ELEVEN1!)

Common environmental chemicals can affect the processes described above in experimental studies in animals, which raises concerns as to whether they might also ... contribute to these disorders in humans. At present we do not know whether or not this is true and obtaining conclusive evidence one way or another ... will take several years. This 'uncertainty' ... often begs the question, especially from pregnant mums, "but what can I do to avoid these chemicals, just in case...?" My answer is that for much of our environmental exposure (e.g. via air, food, fabrics of our house etc) there is not a lot one can do, but you can avoid exposing your baby to ... chemicals present in personal care products (cosmetics, body creams, perfumes) simply by stopping using them - at least for the first 3 months of pregnancy... I would not dream of 'warning' a woman to do this as we have no conclusive evidence that these chemicals harm the baby in any way. My suggestion is purely precautionary - it cannot do any the baby any harm and may do it good. I always place this advice in context - by far and away the most important thing is to not smoke, not to drink alcohol and to eat a sensible well balanced diet.

(Bold emphasis mine, italics original). Eminently sensible, I would have thought. But from data showing that there's an early-pregnancy window where male reproductive organs can be disrupted, the newspapers chose to speculate that, well, creams and sprays do the disrupting. I managed, with no credentials to back me up, to contact Prof. Sharpe and speak to him about his work, get him to explain in simple terms (I am no fertility/developmental biology expert) the central take-home messages from his work - literally in fact, as he gave me some key facts he'll be presenting at the conference. The journalists concerned did not go to that rather simple effort, so very good science gets reported very badly. Not just annoying for the academic in question - it could happen to anyone, as they say - but clearly not responsible reporting, as it causes unjustified alarm amongst prospective mothers and belittles and future finding of culpable causative agents or, more likely, combinations thereof. Or perhaps Prof. Sharpe can put it better:

I would like to distance myself from 'scare story' journalism such as that triggered by the headline in the article in Scotland on Sunday which lead to a succession of similar headlines in other newspapers. It does not accurately reflect the science on which it is based (which, as always, has uncertainties). It insults the commonsense of the public and does them a disservice because it desensitizes them ... so that, if and when we have a health recommendation to make based on strong scientific evidence, it may mean the public do not take it seriously when they actually need to.

Quite.