OBJECTIVE: To determine age-related variations in prostate cancer (PC) patients' utilities. PC is a disease of older men. Decisions regarding its management are sensitive to preferences for outcomes. METHODS: We derived a population-based sample of PC patients diagnosed in 1992, 1997 or 2002 from the Ontario Cancer Registry. Patients (n=1531). Consenting survivors were mailed questionnaires, includingdemographics, Health Utilities Index (HUI2/3), and UCLA Prostate Cancer Index (PCI). We computed Pearson's correlations between HUI3 utility and PCI scores in all patients and between age quartiles. We used regression to test for interactions, namely linear trends in HUI3-PCI slopes across age quartiles. RESULTS: The first 289 patients returned questionnaires in 2004. Mean age was 71.6 years. Primary treatments, received 1–11 years prior, were prostatectomy (46%), radiation (32%), hormones only (12%), and watchful waiting (10%). Mean HUI3 utility was 0.79, SD=0.24. HUI3 and sexual function scores decreased with age (p<0.001), but urinary function improved (p=0.01), with no changes in bother scores. For HUI3 and sexual function, r=0.43 (p<0.001) for 44-65 yr olds, and r=0.14 (p=0.28.) for 78-92 yr olds, p=0.35 for interaction. For HUI3 and sexual bother, r=0.41 for 44-65 yr olds, and r=0.09 for 78-92 yr olds, p=0.06 for interaction. Correlations between HUI3 and urinary function increased with age (from 0.24 to 0.43, p=0.04 for interaction). Similar age effects were observed for urinary bother. There were no age-related trends for correlations between HUI3 and bowel function or bother. Sexual and urinary function scores became less strongly related to their corresponding bother scores as age increased (p<0.002). CONCLUSIONS: Although sexual dysfunction increased with age, it became less bothersome and had less impact on utility. However, older patients' quality of life was more affected by urinary dysfunction. Modelers and policy-makers may need to adjust utilities for PC health states for patient age.