We address 2 errors in the paper by Beaumont et al1 that reanalyzes the data of the late Dr. Zhang,2,3,4 which in 1995 we coauthored (S.L.) and helped him submit for publication (B.K., W.B., T.Y.).
First, among the 5 exposed villages, there is no consistent pattern of association of cancer rates with the gradient of exposure to Cr6+ as noted earlier by Dr. Zhang.2,5 The villages more distant from the alloy plant had higher, not lower, cancer rates while being exposed later and at lower concentrations than the nearer villages. Beaumont et al1 are incorrect in stating there was no gradient in exposure when data for the beginning (1965), middle (1972), and end (1979) of the episode support such a gradient. (The data6,7 for the 2 later time periods are not presented by Beaumont et al.) The geologic and recharge characteristics of the local aquifer caused the main contamination zone to be localized between the alloy plant and the near villages6,7 and the most contaminated groundwater to travel away from the more distant villages.
Second, Beaumont et al (Table 5)1 use incorrect cancer comparison groups for the exposed agricultural villages. The majority of the difference in the local stomach cancer rates is due to TangHeZi (14.5 per 100,000) having a substantially lower rate than the average both for the unexposed villages (27.6 per 100,000) and for the exposed villages (35.4 per 100,000) (Table 4).1 TangHeZi cannot represent the expected stomach cancer rate in the exposed villages “but for” the Cr6+ contaminated water when its stomach cancer rate is significantly lower by about the same magnitude compared with the unexposed villages. A comparison of the average stomach cancer rate in the exposed and the unexposed villages indicates little difference (RR = 1.29; 95% CI 0.75-2.21).8,9 Neither are the average Province cancer rates (Table 5, bottom)1 appropriate as they refer to a vast, heterogeneous geographic area and differ by about the same magnitude from the averages for the unexposed and exposed villages.
Gwiazda,10 an external reviewer selected by Cal-EPA for these analyses by Beaumont and colleagues, described as “misguided” the use of any comparison group that does not represent the expected cancer rates for the 5 exposed villages but for the exposure to Cr6+ contaminated water. This criticism applies equally to the inclusion of TangHeZi in the local comparison group and to the use of the cancer rates for the Province.
It is not necessary to know what specific risk factors are the source of the lower stomach cancer rates for TangHeZi and for the Province to recognize they are inappropriate comparison groups. TangHeZi is among those industrial towns created in the 1950s by the communist Chinese government and populated by skilled workers from the overcrowded major cities. This town has been shown to or is expected to differ from agricultural villages on demographics, urbanization, industrialization and occupation, socioeconomic status, diet, and access to medical care.9,11-12
Finally, Zhang's first manuscript2 reported the absence of a dose-response relationship, and his second3 cautioned that the higher cancer rates in the villages with groundwater contamination relative to the Province average “can only be considered to be possibly connected to Cr6+ contamination of the groundwater, soil and crops.” Zhang's fourth manuscript5 similarly cautioned that “Cr6+ contamination cannot be ruled out completely as the reason for the high cancer death rates in these villages,” and acknowledged that the lacking dose-response trends “do not support such a relationship.” The secondary analyses of Beaumont et al1 are questionable because, as explained above, they misuse the data and reverse the interpretations of the original researcher.9
Brent D. Kerger
Health Science Resource Integration
William J. Butler
Environmental Risk Analysis, Inc.
San Mateo, CA
BenXi Health and Anti-Epidemic Station Ben
Xi City, Liao-Ning Province, China
1. Beaumont JJ, Sedman RM, Reynolds SD, et al. Cancer mortality in a Chinese population exposed to hexavalent chromium in drinking water. Epidemiology
2. Zhang J, Li X. Study of the effect of environmental pollution in the Jinzhou area on residents' health: I Mortality analysis. Jinzhou Health and Anti-epidemic Station. 1980. Original document and certified translation available at http://www.chemrisk.com/public/Zhang
3. Zhang J, Li X. Chromium contamination in the city of Jinzhou. Jinzhou Health and Anti-epidemic Station; 1986. Original document and certified translation available at http://www.chemrisk.com/public/Zhang
4. Zhang J, Li X. Chromium pollution of soil and water in Jinzhou (Chinese language with English abstract). Zhonghua Yu Fang Yi Xue Za Zhi
(Chinese Journal of Preventive Medicine). 1987;21:262-264. Original document and certified translation available at http://www.chemrisk.com/public/Zhang
5. Zhang J, Li SK. Cancer mortality in a Chinese population exposed to hexavalent chromium in water. J Occup Environ Med
6. JinZhou Disease Control and Prevention Station. Survey report on groundwater contamination by waste water containing chromium by the Jinzhou alloy plant. 1979. Original document and certified translation available at http://www.chemrisk.com/public/Zhang
7. Kerger B. Memorandum to Joan Denton, Director of OEHHA re: Groundwater chromium 6+ levels associated with the 1970-78 mortality study by Zhang and Li (1997). April 6, 2006. Available at http://www.chemrisk.com/public/Zhang
8. Butler WJ, Ye T. Memorandum to Joan Denton, Director of OEHHA, re: Statistical details on the analysis of Zhang and Li (1997). February 16, 2006. Available at http://www.chemrisk.com/public/Zhang
9. Kerger B, Butler W, Paustenbach D, et al. Cancer mortality in Chinese populations surrounding an alloy plant with chromium smelting operations 2009. J Toxicol Environ Health A
11. Whyte MK, Parish WL. Urban Life in Contemporary China.
Chicago: University of Chicago Press; 1984.
12. Parish WL, Whyte MK. Village and Family in Contemporary China.
Chicago: University of Chicago Press; 1978.