Abstract Background Limited data exist on long-term outcomes among men with prostate cancer (PCa) from population-based cohorts incorporating information on clinical risk category. Objective To assess 15-yr mortality for men with PCa treated with noncurative intent according to clinical stage, Gleason score (GS), serum levels of prostate specific antigen (PSA), comorbidity, and age. Design, setting, and participants Register-based cohort study of 76 437 cases in the National Prostate Cancer Register (NPCR) of Sweden diagnosed from 1991 through 2009 and treated with noncurative intent. Each case was placed in one of five risk categories: (1) low risk: T1–T2 tumor, PSA level <10 ng/ml, and GS ≤6; (2) intermediate risk: T1–T2 tumor and PSA level 10–<20 ng/ml or GS 7; (3) high risk: T3 tumor or PSA level 20–<50 ng/ml or GS ≥8; (4) regional metastases: N1 or T4 tumor or PSA level 50–100 ng/ml; and (5) distant metastases: M1 tumor or PSA ≥100 ng/ml. Outcome measurements and statistical analysis Ten- and 15-yr cumulative risk of death after diagnosis from PCa, cardiovascular disease, and other causes. Results and limitations Among men with a Charlson Comorbidity Index (CCI) score of 0, no differences were found in observed versus expected all-cause mortality in the low-risk group. Observed mortality was only slightly greater in the intermediate-risk group, but men with high-risk localized PCa or more advanced disease had substantially higher mortality than expected. CCI was strongly associated with cumulative 10-yr mortality from causes other than PCa, especially for men <65 yr. Limitations include potential misclassification in risk category due to GS assignment. Conclusions PCa mortality rates vary 10-fold according to risk category. The risk of death from causes other than PCa is most strongly related to comorbidity status in younger men.