As I was writing my last post about interpreting medical data I was reminded of a legal case from earlier this year. A court in Missouri in the US ruled that a lady’s ovarian cancer was caused by her use of talcum powder produced by Johnson & Johnson. The lady herself had died due to her cancer by the time of her verdict but her family was awarded $72m (£51m). This result was slightly surprising as there is no evidence that talc can cause cancer. There has been a reasonable amount of research into this so let’s look at the evidence.
Up until the 1970s there were trace amounts of asbestos to be found in talc and so up till that time there may well have been an increased risk, however, since then talc has been asbestos free and so we need to consider those time periods separately.
Right off the bat we can conclusively say that if you are just applying talc on the outside of your body then there is no risk at all; no study has ever found a hint of any association between external talc use and ovarian cancer. This rules out all men from the equation, indeed the only people left to consider are women that use it on their genitals.
The second of those studies is a meta analysis of 12,000 people where they found a 33% increase in relative risk for those women who practiced what they called perineal talc use. Here it becomes very important that we understand the difference between relative and absolute risk. The background risk of ovarian cancer is 0.0121% and so a 33% increase in this raises the rate of getting ovarian cancer to 0.0161%; that’s an increase in absolute risk of 0.004%.
Perhaps more importantly, the study found no dose-response relationship. When you are trying to prove the harmful qualities of a substance you would normally hope to see an increase in risk with greater exposure to the substance. The more cigarettes you smoke the greater your chance of getting lung cancer; everyone except Nigel Farage knows that. The fact that increased use of talc did not produce an increased risk in ovarian cancer is a big red flag.
A large study from 2013 looking at nearly 20,000 individuals again found no dose response and a relative risk increase of only 24%. These tiny increases in an already small risk equate to just 3-4 extra cases of ovarian cases per million users of genital talc. This is well within the range of the background noise and is a long way from proving that this particular lady’s cancer was caused by Johnson & Johnson’s products.
A potentially significant problem with the observational studies is that of recall bias. The women who had developed cancer might have been more likely to recall that they had used talc compared to those who hadn’t; they may have also inadvertently exaggerated the amount or frequency with which they used talc.
In my opinion people and, yes, even multinational corporations are supposed to be innocent until proven guilty. For me, there has not been nearly enough evidence shown to prove that daily use of talc caused this person to get cancer. In this particular trial the judgment was made by a jury, should we be surprised that they sided with a grieving family over a faceless corporation? Did they truly understand the evidence brought before them? We’ll never know, but perhaps these cases should be decided upon by a judge in the future.