This paper applies Chance Discovery to the analysis of accident or incident reports to find hidden factors associated with these accidents or incidents. Recently, it has been recognized that medical risk management is very important both for hospitals and hospital patients. Consequently, risk-management experts check accident or incident reports. In addition, data mining methods have been applied to these reports. However, they can only find generalized reasons for frequently occurring accidents or incidents. Finding reasons for rare accidents or incidents is more important because they tend to be missed by experts. We have, therefore, developed an analysis method for such cases by using the concept of Chance Discovery.