Comments on "protracted radiation exposure and cancer mortality in the Techa River cohort" by Krestinina et al. (Radiat. Res. 164, 602-611, 2005).
I read with great interest the recent article by Krestinina et al. (1), which reported the results of a cohort mortality study of the extended Techa River cohort. I was especially intrigued by the authors' conclusion that this study provided "strong evidence of long-term carcinogenic effects of protracted low-dose-rate exposures". If this claim were borne out, it would set this study apart from the numerous other studies of such exposures that have failed to detect carcinogenic effects. I would like to submit two observations for the authors' and readers' consideration. First, the authors attribute the apparent increase in cancer mortality they observed solely to radiation exposure. However, it is at least possible, if not probable, that the liquid effluents discharged into the Techa River from the Mayak complex contained other nonradioactive, toxic and/ or carcinogenic chemical components. If so, this would present a very real potential for confounding the results of this study. This possibility should be explored and discussed before attributing the putative increase in mortality to radiation exposure alone. But there is an even more serious issue: the absence of a quantitative analysis of uncertainty. In particular, while the authors caution that there are large uncertainties in the dose estimates, they refer the reader to a paper by Napier et al. (2) for details. The authors' warning seems, if anything, understated, since Napier et al. report ranges of uncertainty of about a factor of four to five for external doses and ranges for internal doses, reported as ratios of the 97.5th percentile to the 2.5th percentile, as a factor of 20-30 for the TRDS-2000 dosimetry system used in this study. The authors did not display error bars on the data points for either the ERR or the dose axes in Figs. 1 and 2 and did not offer quantitative details on the magnitude of the uncertainties in the text; therefore, the reader is left to assume that the uncertainties in the current paper are as large as those reported by Napier et al. If so, these uncertainties combined with the other limitations the authors acknowledge in their paper make the conclusion that the results of this study "provide clear evidence of elevated solid cancer and leukemia mortality risks and a strong dose-response relationship associated with exposure to radiation from the contaminated Techa River" seem far more definitive than the data warrant.