The employee health service of a Boston hospital wanted a method to prioritize the risk of occupational injury or illness among its employees as the first step in developing a comprehensive ergonomics program. Data from the safety office and workers' compensation third-party administrator (TPA) was combined with hospital payroll data to create rates that compared all work areas based on the common denominator of 100 full-time equivalents (FTE). Rates for four different aspects of injury experience were calculated: incidence of total reported injuries, incidence of serious injuries, level of severity of injuries, and cost. The use of these simple rates alone was inadequate to accurately prioritize risk. Because most work areas ranked differently from one rate scale to the next, it was unclear which, if any, single rate most accurately defined risk. Composite statistics that combined all of the rates were needed. The Composite Risk Indicator (CRI), the Average Relative Risk (ARR), and the Justified Average Relative Risk (JARR) were developed and examined for their utility. The JARR emerged as the best choice in this setting because it captured all available information about injury or illness experience and provided a meaningful single indicator of risk that could be followed over time.