Abstract
A survey of background radiation and mortality from leukemia was conducted in the states of Maine, New Hampshire and Vermont. The radioactivity in bedrock underlying each of 1,178 minor civil divisions was estimated on the basis of radiogeological analysis. Values were found to range from 5 parts per million equivalent uranium in the limestone areas of Vermont to 45 parts per million equivalent uranium in the Conway granite region of New Hampshire.
Mortality records from the tri-state study area for the period 1925-54 were examined. One thousand nine hundred and seventy deaths from leukemia were identified and tabulated by minor civil division. An expected number of deaths by decade was calculated for each minor civil division using the tristate age specific rates for standardization. Minor civil divisions were grouped into four categories according to mean equivalent uranium concentration in underlying bedrock. The difference in annual dose rate from external background sources was estimated to be 14.65 mrads per year. No significant difference in exposure to internal emitters was noted.
The observed and expected numbers of deaths were compared within each of the four radiogeological categories. No statistically significant association was observed between mortality from leukemia and radiogeological category either for the entire 30-year study period or within successive decades.
These data do not allow rejection of the hypothesis that leukemia mortality varies with background radioactivity if the dose response relationship is of the order of that estimated from studies in humans at doses of 100 rads or greater. It seems unlikely that such a hypothesis could be adequately tested in the United States where geographic variation in background radiation is relatively small.
