The term Adverse Safety Events (ASEs) refers to harm from medical errors and, if considered a disease, would rank third among the leading causes of death in the U.S. Patient safety research has traditionally performed manual review of patient care records (PCR) to characterize ASEs and develop prevention strategies. Manual review is labor-intensive, which prohibits continuous monitoring and improvement. Automating detection of ASEs would overcome these barriers and allow the healthcare community to address safety issues at a population scale. Our objective is to demonstrate the feasibility of automatically detecting safety events in PCRs.