While most people have traces of the plastics chemical in their bodies, the study found that children with the highest levels in their urine were twice as likely to be obese than those with the lowest.
There are other factors that could explain the results, and many reasons why children gain too much weight, the researchers said.
"Clearly unhealthy diet and poor physical activity are the leading factors contributing to obesity in the United States, especially in children," said lead author Dr. Leonardo Trasande of New York University.
But the study does hint that causes of childhood obesity may be more complicated, he added. He said it is the first national research to tie a chemical from the environment to childhood obesity, and seems to echo what some studies have seen in adults.
One puzzling result: Significant differences were only detected in white children. For black and Hispanic kids, obesity rates were similar for those with the lowest levels of BPA as those with the largest amount. The researchers couldn't explain that finding.
The study was published in Wednesday's Journal of the American Medical Association.
BPA, or bisphenol A, has been used since the 1960s and become so widespread that it's found in virtually all Americans. Government health officials have deemed low levels of it to be safe, but haven't been able to decide what amount of BPA - if any - would be a health concern.
BPA is used in hardened plastics, food packaging, to line metal cans and other consumer and medical goods. Environmental groups have worried it interferes with children's development, and this year the Food and Drug Administration announced the chemical would be phased out of baby bottles and sippy cups.
Some experimental studies on animals have found that BPA can aid obesity by disrupting metabolism and helping to make fat cells larger. And studies have suggested a possible tie between urinary BPA levels and obesity-related illnesses in adults, including diabetes and heart disease.
In the new study, Trasande and his colleagues used information from an annual federal health survey, which includes weighing and measuring participants and taking blood and urine samples.
Their study involved more than 2,800 children ages 6 through 19, who took the survey in the years 2003 through 2008. They compared BPA levels in their urine to their weight, and divided them into four groups based on BPA amounts.
The key finding: About 22 percent of the children with highest levels of BPA were obese, compared to just 10 percent of kids with the lowest levels.
Was the reverse true? Did the heaviest kids have more BPA in their urine, and the thinnest kids less? Yes, Trasande said. But he did not include those numbers in his study, and declined to provide them.
The study raised more questions than it answered:
- The body excretes the chemical in a matter of hours. It's possible that the study is simply indicating that heavier kids are more likely to have recently consumed something from a BPA container.
- Only one urine sample was taken from each child, and the youngest children in the study were 6. What isn't known is how much BPA they were exposed to when they were infants - the time in life when the chemical theoretically could have had the greatest effect in triggering weight gain.
All this means is that the study raises some interesting questions, but at this point it's impossible to say BPA causes childhood obesity, said Linda Birnbaum, director of the National Institute of Environmental Health Sciences, a federal agency involved in research on BPA.
"It's a hypothesis that needs further exploration," she said.
The American Chemistry Council, which represents chemical manufacturers, called the study speculative and noted lab animal studies that found no evidence that BPA causes obesity.
"Attempts to link our national obesity problem to minute exposures to chemicals found in common, everyday products are a distraction from the real efforts under way to address this important national health issue," the organization said in a statement.