The aim of this study was to establish the cause-effect relationship between renal tubular dysfunction and mortality. A 19-year cohort study was conducted in 900 men and 1313 women in 1993 or 1994 who lived in two cadmium non-polluted areas in Japan. Hazard ratio (HR) and 95% confidence interval (95% CI) of urinary β2-microglobulin (β2-MG) and N-acetyl-β-glucosaminidase (NAG) for mortality were calculated using a proportional hazard regression. Forward stepwise model selection was applied to the potential covariates such as age, body mass index, mean arterial pressure, various lifestyle factors and present illness. Simultaneously, the dose-effect relationship between renal tubular markers and urinary cadmium at baseline was evaluated using multiple regression analyses. In men, HR was significant for β2-MG (HR corresponding to an increase of 100 μg/g cre: 1.02) and NAG (HR corresponding to an increase of 1 IU/g cre: 1.05). In women, a significant HR was observed for β2-MG (HR corresponding to an increase of 100 μg/g cre: 1.01) and NAG (HR corresponding to an increase of 1 IU/g cre: 1.02). Dose-effect relationships were significant for urinary cadmium and all renal tubular markers in men and women. The present study indicated that renal tubular dysfunction was significantly related to mortality in the general population of cadmium non-polluted areas in Japan.