Self-Reported Learning (SRL), a voluntary incident reporting system experience within a large health care organization.

ABSTRACT Background In aviation, significant improvements in safety have been attributed to a system of voluntary reporting of errors and hazards by pilots and other frontline personnel. Such a system is lacking in health care. Methods A system to allow physicians to self-report their clinical care errors along with insights to prevent recurrence (“self-reported learning (SRL) system”) was established in three hospitals and used for four years in one center and for two years in two others. Clinicians were educated in how to use the system and encouraged to report deviations from standard care by secure email, a telephone hotline, or the institutional incident reporting system. We only included events in the SRL system if clinicians self-reported them prior to others doing so. Physicians were also surveyed for their perceptions about the SRL. After analysis, each case was placed in one of four categories: ‘get help’, ‘improve technique or knowledge’, ‘improve system’ and ‘be more careful’; the first three were considered to show secondary insight, defined as learning something that might drive personal or system safety improvements. A quality assurance physician reviewed and excluded errors that were subject to mandatory reporting or were committed while the physician was under the influence. The learnings were then shared hospital-wide. An analysis of SRL submissions was conducted by physician quality leaders familiar with clinical peer review. Submissions were analyzed for evidence of primary insight, recognition of error, and secondary insight. Physicians were surveyed afterwards about their attitudes towards clinical peer review, the physician's role in errors, and the SRL program. Results There were 117 SRL submissions, which accounted for less than 5% of clinical peer review cases; 87 had complete information available. Of these, evidence of secondary insight was present in 40%, and several were novel submissions that otherwise would have not been identified. The response rate to the physician survey was 18%. Of the respondents, 31% had never been involved in clinical peer review before, a majority had an overall favorable impression of clinical peer review, and 47% reported not having made “any mistakes worth reporting.” Conclusion A SRL system modeled after the aviation reporting system elicited a low number of reports and did not decrease the number of clinical peer review reports. In a high proportion of SRL reports reporting physicians demonstrated secondary insight. A SRL system has potential to improve patient safety and increase engagement of physicians. Further research is required to determine how to implement and sustain such a system.