April 11, 2022

expert reaction to study looking at exposure to per- and polyfluoroalkyl substances (PFAS) and risk in middle-aged women of developing diabetes

A study published in Diabetologia looks at per- and polyfluoroalkyl substances (PFAS) and diabetes risk in midlife women.

Prof Oliver Jones, Associate Dean – Biosciences and Food Technology, RMIT University in Melbourne, Australia, said:

“PFAS is a deceptively short term for a large family of compounds which are so widely used in consumer products they can now be found almost anywhere an analytical chemist cares to look.  Indeed, they are so ubiquitous that reducing our potential exposure to these substances is actually quite difficult.

“The fact that certain pollutants might be involved in potentially causing diabetes (likely by interfering with glucose metabolism) has been around for over a decade and PFAS themselves have been implicated in this disease before.

“It should be noted however that, while the present study is detailed and well thought out, and does a good job accounting for potential cofounders, it only claims a potential association between high concentrations of certain PFAS and a higher likelihood of diabetes in middle aged women; and correlation is not causation.  In addition, as the authors of the present work themselves point out, other studies have found no association between PFAS and diabetes so there is clearly some work to be done to unravel what is going on.  In my view, other known diabetes risk factors such as lack of physical activity, and higher BMI likely have a bigger effect on diabetes risk than PFAS exposure.

“PFAS are often referred to as ‘forever chemicals’ because they are very resistant to degradation.  This means if they do cause an increased risk of diabetes there are likely going to be doing so for some time to come.  The health implications of this could, potentially, be substantial however, much more work is needed to determine if the association holds up for groups and exactly how much PFAS exposure over what time period would potentially be required to have this effect.  There is still a lot we don’t know about PFAS.

“PFAS were made somewhat (in)famous by the film Dark Waters.  The range of potentially effects they cause makes it likely we will keep hearing about them for some to come.”

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is a competent study in statistical terms, but care is needed in interpreting its findings.  It’s important that the press release and the research report don’t go further than saying that PFAS may increase diabetes risk in middle-aged women.  They, rightly, can’t be more definite than that on the basis of a study like this.

“For me, the main issues about interpretation stem from the fact that it’s an observational study.  The researchers measured the level of PFAS in the blood of over 1,200 middle-aged women who did not have diabetes, who were enrolled in an existing cohort study in the USA.  They then followed them up for around 17 years and recorded whether and when they developed diabetes.  They found that the risk for developing diabetes was greater in women who had higher levels of PFAS in their blood.  The snag is that this is a correlation, and correlation does not imply causation.  There were differences in many other factors between women with different levels of PFAS, apart from the level of PFAS.  Some of these other factors could, in whole or in part, be what causes the higher risk of diabetes in the women with more PFAS, and not the PFAS level itself.

“The researchers made various statistical adjustments to allow for some of these other differences (ethnicity, the study site where each woman was enrolled, education, smoking status, alcohol consumption, total energy intake, physical activity, menopausal status and BMI, and (in a different statistical way) age).  But one can never be certain that everything relevant has been allowed for.  For example, there must be some reasons why the women’s PFAS levels vary, perhaps to do with the sources of exposure to PFAS such as eating contaminated food, or using some of the many consumer products that contain PFAS.  Some of those reasons for exposure are probably related to the factors that the researchers made adjustments for, but others may well not be, and so won’t have been taken into account.  (In the jargon, there could have been residual confounding.)

“Because of this, the researchers point out the possibility of residual confounding as a limitation of their study, and that’s the main reason they can’t go further than saying that PFAS exposure may increase diabetes risk.  Yes, it may, but there are other possible cause-and-effect explanations that this study can’t rule out.  Another limitation is that these results are only for women, because no men were studied in this research, so they tell us nothing direct about risks in men.

“If a competent observational study like this can’t definitely establish what causes what, you might well ask how we can ever know that pollutant exposure can actually cause health risks.  The answer is that evidence has to be put together from different sources, not just a single study.  There have been other observational studies looking for associations between PFAS exposure and diabetes risk, but, as the researchers point out in their report, the results from those studies have pointed in different directions, with some showing an association and some not.  The chemical structure of PFAS, and the fact they can persist in human bodies for a very long time, are relevant, as are the results of some laboratory studies of the effects of PFAS on cells from humans and from mice.  But some doubt about cause still remains – yes, this study does show that PFAS may increase diabetes risk in middle-aged women, but it certainly can’t rule out other explanations for its findings.”

‘Per- and polyfluoroalkyl substances and incident diabetes in midlife women: the Study of Women’s Health Across the Nation (SWAN)’ by Sung Kyun Park et al. was published in Diabetologia at 23:01 UK time on Monday 11 April 2022.

DOI: 10.1007/s00125-022-05695-5

Declared interests

Prof Oliver Jones: “I conduct research on the fate and behaviour of PFAS in the environment.”

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”