BPA in the urine is primarily a marker for consumption of packaged food as this is the primary source of BPA intake as acknowledged in the paper. In particular food in tins lined with BPA containing polymers and packaged drinks. To find an association between high consumers of packaged food and CVD is not much of a surprise. It would have been appropriate for the authors to have adjusted for diet in their analysis and to have explored this rather obvious potential etiology.

For diabetes a reasonable conclusion from the data provided is that the earlier findings were simply not reproducible. Combining the data from the 2 NHANES surveys to obtain a significant result is questionable in this case. For CVD this approach is more reasonable given similar point estimates for the ORs which suggests lack of significance is a power issue rather than a true lack of association.