A reply to Dan Strom's comments on the Nuclear Shipyard WorkersStudy (NS

I agree that the NNW have a high SMR for all causes and an even higher SMR for cancer (1.12). This doesn't mean that they were not representative of all NNW. It seems very unlikely that picking 32,500 NNW could have a very large error in their SMRs. I think it is important to find out why the NNW are so unhealthy. Do you suppose there is an "unhealthy worker effect" to help balance the healthy worker effect? Until we have data on other NNW we must assume that being a shipyard worker is unhealthy. We shouldn't assume that the study was flawed. I believe that the high dose cohort were protected by their low doses of radiation. The study should have been published. It is surprising that the PI did not comment on the unusually high SMRs of the NNW in the final report. Have you looked at the 100 years of British radiologists.." I can send you a pdf file of it and also a pdf of my letter to the editor of BRH that shows that the 1955-1979 radiologistshad an even greater SMR for all causes (0.68) (p<0.001) compared to other male MDs in England. That is even better than the nuclear shipyard workers. I don't see how selection bias can explain it.

We should both seek the correct explanation for the unusual NNW data. However, there can be no doubt that the cohort had a very low death from all causes, much lower than can be explained by "selection bias". It doesn't make sense for you (i.e., Dan) to blame the good health of the cohort on exceptionally poor health of the controls. I would bet that other unexposed shipyard workers are also unhealthy and someone should find out why.

Gerontology 1998;44(5):272-6 Effect of a continuous gamma irradiation at a very low dose on the life span of mice. Caratero A, Courtade M, Bonnet L, Planel H, Caratero C. Laboratoire d'Histologie-Embryologie-Cytogenetique, Faculte de Medecine Toulouse-Rangueil, Toulouse, France. caratero@cict.fr BACKGROUND: There is epidemiological evidence that suggests there are beneficial effects of ionizing radiation at low doses. Some experimental studies confirmed this hormetic effect with doses of about 1 cGy/day, but no data concerning very low dose rates are available. OBJECTIVE: The aim of this study was to determine the life span of mice exposed to very low doses of ionizing radiation. METHODS: Six hundred female C57BL/6 mice, 1 month old, were exposed to chronic gamma irradiation at very low dose rates of 7 or 14 cGy/year. These doses are about 25 or 50 times higher than background, but much lower than the doses of about 1 cGy/day used in previous experiments. Three hundred mice living in the same room were used as controls. RESULTS: The life span, after the beginning of the experiment, determined by the survival time of 50% of each population, is increased in irradiated mice: 549 days in controls, 673 days in both irradiated groups. The differences are significant between the control and the irradiation mice. Differences between mice irradiated with 7 or 14 cGy are not significant. CONCLUSIONS: These results confirm the possibility of a nonharmful effect (hormesis) of ionizing radiation. They demonstrate that the paradigm, which states that low-dose effects can be predicted high-dose effects, cannot be systematically applied in radiation biology in general and gerontology in particular. PMID: 9693258

A reply to Dan Strom's comments on the Nuclear Shipyard WorkersStudy (NSWS)