A B S T R A C T Purpose Although it is the most powerful predictor of early prostate cancer treatment-related complications and quality-of-life (QOL) outcomes, most studies do not stratify results by baseline function. Further, reporting functional outcomes as averaged numerical results may obscure informatively disparate courses. Using levels of treatment-related dysfunction, we address these problems and present the final QOL outcomes of our prospective cohort study of patients with early

more »

... tate cancer. Methods We created categories for sexual, bowel, and urinary function, measured using numerical scores of the validated Prostate Cancer Symptom Indices and stratified into "normal," "intermediate" and "poor" levels of function by incorporating patient-reported symptom and distress information. We present QOL outcomes for 409 patients 36 months after radical prostatectomy, external-beam radiation therapy, and brachytherapy. Results Different levels of baseline sexual, bowel, and urinary function produced distinctive treatmentrelated changes from baseline to 36 months. In general, the average scale increases in dysfunction were greatest among patients with normal baseline function, although patients with normal and intermediate baseline function had similar increases in sexual dysfunction. For patients whose baseline urinary obstruction/irritation was poor, both average scale scores and most patients' level of function improved after treatment, particularly after surgery. Conclusion The use of functional levels to stratify treatment-related outcomes by pretreatment functional status and to display the proportions of patients with improved, stable, or worsened function after treatment provides information that more specifically conveys the expected impact of treatment to patients choosing among localized prostate cancer treatments.