An international research team has found evidence linking Bisphenol A, commonly used in food and drink containers, to diabetes and heart disease in adults.

The research produced from the Peninsula Medical School, the University of Exeter and the University of Plymouth, all in the United Kingdom, and the University of Iowa is to be published in the Journal of the American Medical Association on Wednesday.

It is the first time that evidence has emerged of the association between higher BPA levels and disease in adults.

BPA is a controversial chemical commonly used in food and drink containers. It has previously caused concerns over health risks to babies, as it is present in some baby bottles.

BPA is used in polycarbonate plastic products such as refillable drinks containers, compact disks, some plastic eating utensils and many other products in everyday use. It is one of the world's highest production volume chemicals, with more than 6.4 billion pounds produced in 2003, with an annual growth in demand of between 6 and 10 percent each year.

Many previous studies in laboratory animals have suggested that BPA is safe, but some laboratory studies have raised doubts.

Experiments in which mice and rats were exposed to BPA have shown that higher doses of the chemical can lead to liver damage, insulin resistance, diabetes and obesity. The laboratory animal evidence is complicated and controversial. Some scientists believe that BPA can disrupt the work done by hormones, especially oestrogen, but the full biological effects of BPA in humans is far from clear.

The research team analyzed information from the National Health and Nutrition Examination Survey (NHANES) 2003-2004, the only large-scale data available on BPA concentrations excreted in urine. The research team analyzed the results for the 1,455 adults aged between 18 and 74 years old for whom measures were available. The study group is representative of the general population in America.

The analysis found that the 25 percent of the population with the highest BPA levels were more than twice as likely to have heart disease and/or diabetes, compared to the 25 percent with the lowest BPA levels. Higher BPA levels were also associated with clinically abnormal liver enzyme concentrations.

Researchers cautioned that more work is needed to exclude the small possibility that the association is due to some other unstudied factor, or that people with these diseases somehow become more exposed to BPA. It is also unclear whether the liver enzyme changes are linked to liver damage.

"At the moment we can't be absolutely sure that BPA is the direct cause of the extra cases of heart disease and diabetes," warned David Melzer, professor of epidemiology and public health at the Peninsula Medical School in Exeter, U.K., who led the team. "If it is, some cases of these serious conditions could be prevented by reducing BPA exposure. This is therefore an exciting finding, but it is also just the first step in understanding the role of BPA."