Response to Nicolich

In a previous letter (1), I identified a significant correlation between the statewide prevalence of diabetes and air releases of toxicants reported in the Toxics Release Inventory (TRI) report (available from www.epa.gov/tri). Although I suggested a possible mechanism for this link, related to dioxins, I was careful to note that correlations do not prove cause-and-effect relationships. Dr. Nicolich expands on that point in his letter (2).

A coincidence is a “striking occurrence of two events at one time apparently by mere chance” (3). I would agree that it is a virtual certainty that the correlation between state names and diabetes is a coincidence. It is probably a coincidence that latitude is correlated with the prevalence of diabetes, although there are diseases, such as multiple sclerosis and malaria, where latitude is important. It is almost certainly not a coincidence that Dr. Nicolich’s authorship is related to his employment history with ExxonMobil Biomedical Sciences, as shown by institutional affiliation listings on another letter like this one and publications retrieved via MEDLINE (4).

Perusal of the ExxonMobil web pages (www.exxonmobil.com) and TRI data show why this company might seek to dissociate itself from consequences of TRI-reported air releases. Affiliates of ExxonMobil mine and sell coal to electrical utilities; according to the 2000 TRI, electric utilities released 787.8 million pounds of toxicants into the air—more than any other industry. ExxonMobil produces a variety of chemicals: the chemical industry ranked 2nd in total TRI-reported air releases, discharging 277.5 million pounds of toxicants. The petroleum industry ranked 10th, discharging 10.5 million pounds. Together, these three industries accounted for ∼57.6% of all TRI air releases in 2000. Dioxins released by electric utilities and chemical industries accounted for ∼46% of all reportable air discharges.

Air releases tell only a small part of the story of toxicants. The TRI also provides data on waste disposal, where there is also a potential for human exposure. Production-related toxic waste reported by all TRI industries increased by 7.77 billion pounds or 26% between 1999 and 2000. In 2000, reporting industries listed a total of 37.9 billion pounds of TRI chemicals that were managed (recycled, treated, burned for energy, etc.). The chemical industry led all others in off-site waste transfers of >923 million pounds and off-site transfers of dioxins of >40,000 g. When managed wastes are added to air and other releases, the total shows that U.S. industries reported ∼46 billion pounds of toxic waste in 2000 or ∼160 pounds per U.S. citizen.

Is it a coincidence that the prevalence of diabetes is also rising? I don’t know the answer. However, Dr. Nicolich’s attempt to discourage rigorous efforts to find an answer makes the search seem more important.