The Occupational Safety and Health Administration (OSHA) revised the Bloodborne Pathogen Standard and, on July 17, 2001, began enforcing the use of appropriate and effective sharps devices with engineered sharps-injury protection. OSHA requires employers to maintain a sharps-injury log that records, among other items, the type and brand of contaminated sharps device involved in each injury. Federal OSHA does not require needlestick injury rates to be calculated by brand or type of device. A sufficient sample size to show a valid comparison of safety devices, based on injury rates, is rarely feasible in a single facility outside of a formal research trial. Thus, calculations of injury rates should not be used by employers for product evaluations to compare the effectiveness of safety devices. This article provides examples of sample-size requirements for statistically valid comparisons, ranging from 100,000 to 4.5 million of each device, depending on study design, and expected reductions in needlestick injury rates.