Ever since To Err is Human advocated for reporting critical incidents to improve patient safety, incident reporting systems have proliferated across health care organizations (To Err Is Human: Building a Safer Health System. 2000). Over the past two decades, critical incident reporting systems have become recognized as one of the principal drivers for improving patient safety (Eur J Anaesthesiol 2010;27:592-7). Effective incident reporting systems enhance safety and foster a positive safety culture (PLoS One 2015;10:e0144107; Pract Radiat Oncol 2013;3:157-63). Incident reporting systems identify and address hazards and latent harms that may result in patient injury. However, widespread use of incident reporting systems has led to overwhelming numbers of reports, many focused on mundane events or specific individuals that have little to do with patient safety. To prioritize patient safety, we must explore pitfalls of incident reporting systems and identify paths for improvement.

Incident reporting systems are not a measure of patient safety. Using incident reporting systems as a measure of patient safety is problematic. First, the number of reports severely underestimates the true rate of adverse events (BMJ 2007;334:79; Jt Comm J Qual Improv 1995;21:541-8). Because data are self-reported, many events are forgotten or go unreported, often out of fear of punishment or blame, production pressure, etc. Second, fluctuations in certain incidents may reflect fluctuations in reporting behavior rather than true changes (Qual Saf Health Care 2008;17:400-2). Indeed, Amy Edmondson in The Fearless Organization found that nursing units that had a better patient and psychological safety culture were more likely to have a greater number of incidents reported (The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. 2019). Third, data from incident reporting systems do not track temporal trends in critical incidents. If anything, incident reporting systems better reflect an organization’s reporting culture rather than actual patient safety or clinician performance.

Physicians eschew incident reporting systems. Nurses are far more likely to report than physicians (Qual Saf Health Care 2008;17:416-23; Med J Aust 2004;181:36-9; Jt Comm J Qual Patient Saf 2009;35:139-45). However, physicians are more likely to report incidents with higher impacts on patients (asamonitor.pub/3vbgE58). The lack of physician engagement in incident reporting may be due to difficulty in accessing or navigating the system, production pressure, lack of time, and the burden of more paperwork. Physicians may feel reluctant to report if there will be no feedback or action from the organization. Finally, fear of retribution or unfair blame is a significant concern (Anesth Analg 2012;114:604-14).

Criteria for reporting are undefined. Clinicians are typically not trained on what should be reported. This leads to significant variation in types of reported events (for example, near-misses versus actual harm or an unexpected event versus a known complication). Should an unanticipated patient death be reported, or will that automatically be followed up without a report? Should an infiltrated I.V. discovered on induction of anesthesia be reported, or is that too trivial? Lack of clarity on what to report may lead to a large volume of reports on individual problems and mundane issues rather than bringing forward incidents that provide generalizable learning opportunities for the organization.

Define the purpose of the incident reporting system. Ideally, incident reporting systems reveal hazards or rare events that may lead to patient harm. The purpose is to identify opportunities for intervention, improvement, and generalizable learning. By their very nature, incident reports promote a reactive response to error. The reports become the first step in identifying systems issues that deserve thorough investigation and analysis. Institutional leaders should use reports to compare faulty processes to those that have consistently worked well.

This represents a shift from Safety I (investigation into why things go wrong) to Safety II (investigation into why things go right and adapting those successes to similar processes) (Safety-I and Safety-II: The Past and Future of Safety Management. 2014). Leaders must emphasize that the intent of incident reporting systems is to identify flawed systems rather than flawed individuals. Process-related reports (e.g., “the order set for patient transfer fails to reconcile medications”) may be more valuable than outcome-related reports (e.g., “Mr. Jones was injured because his heparin was stopped when he was transferred”) (Jt Comm J Qual Patient Saf 2009;35:139-45). The organization must educate the clinical workforce on which events should be reported (for example, a process-related issue that led to patient injury) and which events should be discussed via other means (for example, complaints about specific individual behaviors). The quantity of reports should be de-emphasized, and the quality of reports should be prioritized.

Redesign the incident reporting systems. To encourage engagement, especially from physicians, incident reporting systems must be accessible, intuitive, and quick. For example, incident reporting systems can be placed within the electronic medical record for easy access. There should be only a few categories, facilitating easy data entry. The reporter should not be expected to enter every detail regarding the event. For example, it doesn’t matter if the patient’s date of birth is missing, because the report is only intended to initiate an investigation of the incident. The report should take no longer than a minute or two to enter.

After reports are made, there must be timely investigation into the event and a standardized approach to analysis of root causes. Integration of the reporter into the investigation and analysis process can help boost engagement as well as provide insights from the front line into how processes can be improved. Feedback regarding next steps must be provided to everyone involved in the event. Confidentiality of the individuals involved must be maintained to avoid a culture of blame and shame. When severe harm occurs, the organization must have a peer support system available to provide resources to clinicians to mitigate the second victim effect.

Maintain a healthy culture of safety. Maintenance of a healthy culture of safety requires participation of everyone, physicians and nonphysicians alike, in understanding that patient safety issues cannot be addressed if they are not reported. Principles of a “Just Culture” (see article on page 22) should be emphasized. There must be no retribution for individuals who report patient safety problems. Conversely, if health care providers are reckless, they must be held accountable. Psychological safety must be ingrained within a healthy culture of safety. Good “catches” should be celebrated! Reporting bullets dodged encourages engagement in the patient safety process and sets examples for others to be vigilant.

To be effective, incident reporting systems should focus on safety and not be used to report mundane events, target individual behaviors, or track trends in patient safety. Incident reporting systems create opportunities for generalizable learning that can benefit the entire organization. The number of reports should be ignored, while the quality of and interventions from those reports should be celebrated. The patient must remain the core focus for the institution so that we are always making progress in advancing patient safety.