Feb 25 2016

The Johnson and Johnson Talc Cancer Case

A jury has recently found for the plaintiff against the company Johnson and Johnson over the claim that their talc powder may have caused ovarian cancer in an Alabama woman who died of the cancer at 62. They awarded her family $72 million.

This story has had a great deal of attention because it raises two questions: what is the scientific evidence for a link between talc use and ovarian cancer, and how should the courts rule in such cases when the science is ambiguous?

Talc and Cancer

Concerns about the cancer causing effect of regular talc use stem from a time when talc contained asbestos. Since the 1970s, however, talc has been asbestos-free. Asbestos is clearly linked to cancer, but for the asbestos-free talc the link is not as clear.

There have been a number of large epidemiological studies looking at the association of talc use and risk of ovarian cancer, with some mixed results, but overall not impressive.

First, there is no association between use of talc but not on the genitals and ovarian cancer. Some studies do show a small association between regular talc use on the genital area and some types of ovarian cancer, but this association is uncertain.

A 2003 meta-analysis involving 11,933 subjects found a 33% relative risk increase of ovarian cancer for regular genital talc use. However since the baseline risk of ovarian cancer is 0.0121%. per year, that results in an increase to 0.0161%, so the absolute risk increase, if real, is tiny (0.004%).

The authors, however, concluded that this increased risk is likely not real because their data did not show any dose response relationship. Exposure risks generally follow a dose-response curve. The more years you smoke, the greater your increased risk of lung cancer. But this relationship was not seen for genital talc use and ovarian cancer.

Further, when they looked at the hospital-based studies there was no effect. The effect was only seen in the population-based studies. Since these are observational studies that raises the possibility of confounding factors. For example, it is likely that women who do develop ovarian cancer may report increased talc use in the past. That is how these studies are generally done – they rely upon memory for past use, leading to a recall bias.

A large 2013 case-control study, looking at 8,525 cases and 9,859 controls, found a 24% relative risk increase in genital talc users. However, they also found no dose-response.

A somewhat smaller 2015 case-control study looking at “2041 cases with epithelial ovarian cancer and 2100 age-and-residence-matched controls” also found a 33% relative risk increase for genital talc use. They also found a “trend” for increasing risk with increased talc-years – “trend” means not statistically significant.

What all of this evidence means is that the likely upper limit of increased risk for ovarian cancer from regular genital talc use is an absolute risk increase of 0.004%. That translates to 4 extra cases of ovarian cancer for every one million genital talc users.

Even this small risk is questionable due to the lack of a consistent dose-response effect seen in the clinical trials.

Legal Implications

As a society what should our stance be toward corporate responsibility in the face of ambiguous scientific evidence? Clearly corporations should follow regulations. They should do due diligence in ensuring their products are safe and effective. They should have transparency and not hide possible risks of their products.

Nothing, however, is risk free. At what point are possible risks so low that they are below the radar, in the background noise of life? What is the responsibility of corporations to disclose possible but not proven tiny risks?

This is why product labels are often full of disclosures and warnings. There is a downside, however, to setting the bar incredibly low (warning about every possible risk, and every possible misuse of a product), it creates “alarm fatigue.” Such warnings become part of the background, and consumers learn to ignore them.

The other issue is this – who should be favored when the science is preliminary or ambiguous, the defendant or plaintiff? If a consumer may have been harmed, should the courts give them millions of dollars? Or should corporations be innocent until proven guilty? I don’t think there is any simple answer here.

I do think, however, that the courts are perhaps biased a bit too much toward the plaintiff. Dow chemical, for example, had to file bankruptcy after a massive settlement against them for their silicone breast implants causing autoimmune disease based upon dubious scientific evidence. Eventually the scientific evidence worked itself out and it was determined there was no risk, but the damage had been done.

In the early 20th century there were numerous court awards based on the theory that minor trauma could result years later in cancer. The trauma-cancer link was eventually disproved, but decades of court cases had already awarded damages based on this flawed theory.

The courts generally take the position that alleged victims should not be penalized because the science has not yet caught up and proven they are victims.

It also seems that juries are more sympathetic toward victims than large corporations, and perhaps that is reasonable, but does it result in the most just outcome? Perhaps juries are not the best option in determining such cases, which depend heavily on expert interpretation of scientific data.


Ideally court decisions would be based on a reasonable interpretation of the best scientific evidence. This then has to be applied in individual cases, balancing the need to compensate genuine victims, but also not unduly and unjustly burden businesses with liability that is ultimately not real.

There is no one simple answer to this dilemma – courts just have to make the best balance in each individual case.

In the case of the Johnson and Johnson award, they are likely to appeal and the size of the judgment is not likely to stand.

It is an interesting question – how much is a possible (but not proven) 0.004% increased risk of ovarian cancer worth?

44 responses so far