Concern over non-stick chemical

“A common household chemical found in everything from sofas and carpets to pots and pans has been linked to an increased risk of thyroid disease,” The Guardian said.

This report is based on a study looking for a link between thyroid disease and levels of two chemicals in the blood. Blood samples and medical questionnaires from nearly 4,000 adults in the US, between 1999 and 2006, were analysed. This showed that people with the highest levels of perfluorooctanoic acid (PFOA) were more than twice as likely to report thyroid disease than people with the lowest levels.

However, just one blood test was taken at the time of the survey, so it cannot be concluded that thyroid disease followed exposure to the chemical. The association could be the other way around, with people with thyroid disease more likely to absorb and store this chemical.

The researchers caution that these results are not conclusive proof of a link. Other studies have not found this link, and such conflicting results suggests that further research is needed.

Where did the story come from?

This research was carried out by Dr David Melzer and colleagues from the Epidemiology and Public Health Group, and the Environment and Human Health Group at the Peninsula Medical School and University of Exeter. These institutions also funded the study. The peer-reviewed study was published online in the journal Environmental Health Perspectives.

Several newspapers reported this study, and named a variety of different household items, such as sofas, non-stick cookware or carpets, that contain perfluorinated chemicals (PFCs). All correctly acknowledge that this is early research, which requires further investigation.

What kind of research was this?

This was a cross-sectional analysis of a long-running survey to assess the health and nutritional status of adults and children in the US. The surveys began in the early 1960s, and involve a combination of interviews and physical examinations. It is now a continuous program that questions and examines a nationally representative sample of about 5,000 people each year.

With such rich data and many possible links to analyse, it is possible that some statistically significant findings may occur by chance. Other limitations include the fact that all measurements were made at the same time. In addition, the researchers asked the participants whether they were diagnosed with thyroid disease, rather than confirming it through blood tests.

What did the research involve?

Researchers analysed the data from three of the annual surveys of the National Health and Nutrition Examination Survey (NHANES), 1999-2000, 2003-04 and 2005-06. They included data from 3,974 adults who had levels of PFCs measured. The PFCs analysed were perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS).

These compounds are stable and have many industrial and consumer uses because the carbon-fluoride bonds in the chemicals repel water. They are used as stain and water-resistant coatings for items ranging from carpets and fabrics to fast-food packaging, fire-resistant foams, paints and hydraulic fluids. The authors note that the persistence of these chemicals in the environment and their toxicity in animal models has raised concern over the possible effects of low-level chronic exposure on human health.

In addition to the PFC measurements, the people were asked if they had any diseases that had been diagnosed by a doctor. For example, whether they had ever been told by a doctor or health professional that they had a thyroid problem (goitre and other thyroid conditions), and if they still had the condition. The researchers defined people with thyroid disease as those who said they currently had thyroid disease and were taking any thyroid-related medication. However, no details were available on specific thyroid disease diagnosis.

The researchers analysed the data using the recognised statistical technique of regression analysis, with models adjusted for age, sex, race/ethnicity, education, smoking status, body mass index and alcohol intake.

What were the basic results?

The researchers report that individuals with the highest levels of PFOA (above 5.7 nanogrammes per millilitre [ng/ml]), were more than twice as likely to report current thyroid disease than people with the lowest (below 4.0ng/ml).

When men and women were analysed separately, the adjusted rates of thyroid disease were 16.18% for women (292 women) and 3.06% for men (69 men).

In the models, women with 5.7ng/ml or higher of PFOA (the top quarter of the population) were more likely to report current treated thyroid disease than those in the lowest two quarters (odds ratio [OR] 2.24, 95% confidence interval [CI] 1.38 to 3.65, p=0.002).

For men there was a ‘trend’ towards a similar two-fold increase, but this was not statistically significant (OR 2.12, 95% CI 0.93 to 4.82, p=0.073).

For PFOS, a significant association between high levels of exposure and current thyroid disease was seen in men, but not in women.

How did the researchers interpret the results?

The researchers say that “higher concentrations of serum PFOA and PFOS are associated with current thyroid disease in the US general adult population”.

To exclude other possible explanations for this association, they call for further research “to establish the mechanisms involved”.

Conclusion

This analysis of population survey data has suggested a possible link between PFCs and thyroid disease that will need further investigation. Based on these findings alone, it is too soon to say that PFCs causes thyroid disease, because there are several limitations to this cross-sectional analysis:

• The exposure to PFCs was measured at the same time as the questions were asked about thyroid disease. It is therefore not possible to conclude that the higher exposure to PFCs came before the thyroid disease. It is also possible that people with thyroid disease are more prone to storing PFCs in their bodies, or slower at excreting it.
• The participants were asked to self report if they had been diagnosed with thyroid disease, but as this was not confirmed by a blood test it is not possible to say if they did have it or what type of thyroid disease they had.
• It is possible that when large amounts of data are collected and re-examined in multiple ways, some associations are made that may have occurred by chance. Ideally, studies should be designed to test specific theories before starting to collect data. In this case, a prospective study that followed the participants’ exposure to the chemical (and their outcomes over time) would be best.
• The authors note that some other studies have not found links between PFOA exposure and thyroid hormone imbalances, even though people in some of these studies had higher levels of PFOA exposure than in the current study. A systematic review of all such studies might suggest whether this link warrants further study.

The study was cross-sectional, therefore the researchers caution that the findings do not conclusively prove a link. Other studies have not found this link, and such conflicting results suggest that further study or a systematic review is needed.