Background: Perfiuorooctanoate (PFOA) is a synthetic chemical widely detectable in blood of nonoccupationally exposed persons. Its human health effects are not well-characterized. Methods: We conducted a mortality study in a cohort of 3993 employees of an ammonium perfiuorooctanoate (APFO) manufacturing facility. APFO rapidly dissociates to PFOA in blood. We estimated standardized mortality ratios (SMRs) compared with the general population and fit time-dependent Cox regression models to estimate the risks using an internal-cohort referent population. A priori diseases of interest were liver, pancreatic, prostate and testicular cancer; cirrhosis of the liver and cerebrovascular disease. Results: APFO exposure was not associated with liver, pancreatic or testicular cancer or with cirrhosis of the liver. SMRs (95% CI) for prostate cancer with no, probable and definite exposure strata were 0.4 (0.1-0.9), 0.9 (0.4-1.8), and 2.1 (0.4-6.1), respectively and for cerebrovascular disease 0.5 (0.3-0.8), 0.7 (0.4-1.1), and 1.6 (0.5-3.7), respectively. The diabetes SMR for probable exposure was 2.0 (1.0-3.2). Compared with an internal referent population of nonexposed workers, moderate or high exposures to ammonium perfiuorooctanoate were positively associated with prostate cancer (HR = 3.0 [10.9-9.7] and 6.6 [1.1-37.7], respectively) and with cerebrovascular disease (1.8 [0.9-3.1] and 4.6 [1.3-17.0], respectively). Diabetes was associated with moderate exposure 3.7 (1.4-10.1); no deaths from diabetes occurred in workers with high exposure. Conclusion: We did not observe ammonium perfluorooctanoate exposure to be associated with liver, pancreatic and testicular cancer or cirrhosis of the liver. Exposure was associated (albeit inconsistently) with prostate cancer, cerebrovascular disease and diabetes.