The purpose of this study was to add to our understanding of naturalistic decision making (NDM) in health care and how after-action reviews (AARs) can be utilized as a learning tool to reduce errors. The study focused on the implementation of a specific form of AAR, a post-fall huddle, to learn from errors and reduce patient falls. Utilizing 17 hospitals that participated in this effort, information was collected on 226 falls over a period of 16 months. The findings suggested that the use of self-guided post-fall huddles increased over the time of the project, indicating adoption of the process. Additionally, the results indicate that the types of errors identified as contributing to the patient fall changed, with a reduction in task and coordination errors over time. Finally, the proportion of falls with less adverse effects (such as non-injurious falls) increased during the project time period. The results of this study fill a void in the NDM and AAR literature, evaluating the role of NDM in health care specifically related to learning from errors. Over time, self-guided AARs can be useful for some aspects of learning from errors.