Editorial: Toxic Metals
Wilcox, Allen J.; Savitz, David A.; Samet, Jonathan M.
This issue includes an extraordinary pair of environmental papers.1,2 Both deal with the health effects of water highly contaminated by toxic metals. Both present data collected decades ago, and with inescapable limitations as a consequence. Both studies have publication histories distorted by regulatory pressures and (at least indirectly) by actions of industry. Both studies shed light on environmental pollutants and human health that remain relevant.
The better known of these 2 incidents took place at Minamata Bay. A Japanese factory discharged methylmercury into the bay from around 1953 to 1968. The bioconcentration of methylmercury in fish led to toxic exposures among the residents of nearby fishing villages. Methylmercury causes neurologic symptoms in adults and neurologic damage to the fetuses of pregnant women. Readers may recall Eugene Smith's iconic photograph of a mother cradling her severely-disabled adult daughter in the bath.
In 1971, Japanese investigators conducted the only population-based study of Minamata effects—a survey of neurologic symptoms among residents in 3 small villages around the Bay.1 The fact that these data have remained unpublished for 35 years suggests the difficulties of conducting science in such a politically-charged atmosphere.
In the second article,2 Beaumont and his colleagues at the California Environmental Protection Agency revisit the story of Dr. Jiandong Zhang, a city doctor who was banished to the countryside during China's Cultural Revolution. The villagers who came under Dr. Zhang's care were afflicted with mouth sores and gastrointestinal ailments. Zhang traced these to drinking water so heavily contaminated with chromium (a byproduct of local steel manufacturing) that the water had turned yellow. Zhang and his colleagues documented the contamination of wells with chromium, which led the factory to reduce and eventually stop active pollution. Zhang went on to gather mortality data that suggested an excess of stomach cancer in the exposed villages. Zhang's data were published in Chinese in 1987—a paper that won a prize in China and influenced environmental policies in the United States.
In 1997, Zhang appeared as the author of a paper in English that reinterpreted the cancer data as negative.3 It was subsequently discovered that the English-language publication had failed to disclose financial and intellectual input from consultants connected to the chromium industry. Zhang's death in 1999 made it impossible to clarify his own role in this paper. The editors of the journal publishing the paper retracted it in 2006.4 The report by Beaumont et al2 provides a further reanalysis of the Chinese data, and affirms elevated cancer risk in the exposed. Beaumont and his colleagues include a number of the original documents (many in Chinese) as electronic appendices. Further historical and scientific context for these 2 remarkable cases of environmental pollution is provided by Smith5 and Grandjean.6
These dramatic events and their human cost provide lessons on several levels. One lesson is regarding the practice of epidemiology. Even as our journal encourages high-quality epidemiologic methods, we recognize that imperfect methods can yield valuable etiologic information. The most compelling studies are not necessarily the most elegant. These data from Japan and China demonstrate the value of even limited studies of extreme exposures—high levels, with defined sources, affecting large areas. The findings from Japan fill a gap in one of the most tragic epidemics ever to be caused by an environmental toxicant. The findings from China provide grist for ongoing regulatory deliberations.
These studies are not the final word. They cannot define dose-response relationships, or measure latency from exposure to disease manifestation, or address cofactors in disease etiology. Nonetheless, once certainty is established at high doses, the subtler aspects of etiologic relationships can be pursued with more confidence, better focus, and even the possibility of elegance.
1. Yorifuji T, Tsuda T, Takao S, et al. Long-term exposure to methylmercury and neurological signs in Minamata and neighboring communities. Epidemiology. 2008;19:3–9.
2. Beaumont JJ, Sedman RM, Reynolds SD, et al. Cancer mortality in a Chinese population exposed to hexavalent chromium in drinking water. Epidemiology. 2008;19:12–23.
3. Zhang J, Li S. Cancer mortality in a Chinese population exposed to hexavalent chromium in water. J Occup Environ Med. 1997;39:315–319.
4. Brandt-Rauf P. Editorial retraction. J Occup Environ Med. 2006;48:749.
5. Smith A. Hexavalent chromium, yellow water and cancer: a convoluted saga. Epidemiology. 2008;19:24–26.
6. Grandjean P. The delayed appearance of a mercurial warning. Epidemiology. 2008;19:10–11.
© 2008 Lippincott Williams & Wilkins, Inc.