Salt and geographical mortality of gastric cancer and stroke in Japan.

Age- and sex-adjusted mortality rates of gastric cancer and stroke in the years 1974-6 for 46 prefectures and 12 regions in Japan were related to regional differences of average per capita daily intake of salt during the period 1966-70 to test Joossen's hypothesis that salt is a common cause of both gastric cancer and stroke. While mortality rates of stroke were strongly correlated with salt intake in 12 regions (r = 0.85, p less than 0.001), mortality rates of gastric cancer showed hardly any correlation with salt intake (r = 0.02). There was no strong correlation between the two diseases in perfectural or regional mortality rates (r = 0.20). It was concluded that geographical mortality of gastric cancer and stroke in Japan did not support Joossens's hypothesis, and other evidence against the salt hypothesis has also been referred to.