During the past decade occupational health and safety performance has increasingly been used as a measure of organizational excellence. Companies in all sectors and of all sizes are making improvements to safety programs not only because it is a matter of compliance, but also to improve productivity and reduce liability. These organizations have come to expect real and lasting value from improvements to their safety programs. Steel mills are among the most hazardous work environments in the industrial sector. The processes employed to manufacture steel in the mini-mill environment are not highly variable, yet there is evidence that some mills perform much better with respect to safety outcome measures, such as Occupational Safety and Health (OSHA) recordable injury rates and lost-time work days, than others. Since the process of making steel does not vary substantially from mill to mill, it stands to reason that other variables impact safety performance within the steel production environment. Previous research has shown predictive associations between a transformational leadership style and other organizational outcomes. The research undertaken to date has not, however, adequately evaluated the possible association between safety outcome metrics (e.g. OSHA recordable injury rates and OSHA severity rates) and a transformational leadership behaviors. The intent of this research is to evaluate the leadership behaviors of the positional leaders and correlate that to safety performance. Three non-union, Midwestern steel production facilities were asked to participate in the study. The transformational leadership behaviors of the leaders in the organization were evaluated by individuals who report to the person being evaluated. The transformational leadership behaviors displayed by the leader was correlated with safety performance. Data were analyzed to determine if there is a statistically significant association between leadership behaviors and safety performance. It was found transformational leadership has a significant association with OSHA injury rates but not with OSHA severity rates. Findings and implications for practice and research are discussed.