OBJECTIVE A prostate specific antigen velocity threshold of 0.75 ng/ml per year has commonly been used to distinguish men with prostate cancer from those with benign prostate conditions. In addition, a prostate specific antigen velocity greater than 2 ng/ml per year has been linked to an increased prostate cancer specific mortality rate after radical prostatectomy and after radiation therapy. However, both of these frequently cited thresholds were determined largely in groups of men with a prostate specific antigen greater than 4 ng/ml. METHODS Of approximately 26,000 men who participated in a prostate cancer screening study 22,019 had a prostate specific antigen of 4 ng/ml or less. Of these men 501 were diagnosed with prostate cancer and had sufficient data for a prostate specific antigen velocity calculation. We performed univariate and multivariate analyses to compare cancer detection rates and performance characteristics using various prostate specific antigen velocity thresholds in these men. RESULTS In men with a prostate specific antigen less than 4 ng/ml, a prostate specific antigen velocity threshold of 0.4 ng/ml per year was most useful for recommending prostate biopsy. Overall prostate cancer was diagnosed in 223 (2%) men with a prostate specific antigen velocity less than 0.4 ng/ml per year compared to 278 (13%) men with a prostate specific antigen velocity greater than 0.4 ng/ml per year (p <0.0001). On multivariate analysis a prostate specific antigen velocity greater than 0.4 ng/ml per year was a stronger independent predictor of prostate cancer diagnosis than age, race or a family history of prostate cancer. CONCLUSIONS The traditional prostate specific antigen threshold of 0.75 ng/ml per year was determined largely in men with a total prostate specific antigen of 4 to 10 ng/ml. Prostate specific antigen velocity thresholds in the range of 0.4 ng/ml per year should be used to help guide the need for biopsy in men with a total prostate specific antigen less than 4 ng/ml.