A punitive approach to safety event reporting at healthcare organizations is counterproductive, a recent research article concludes.
Two decades after the dawn of the patient safety movement in healthcare with the publication of “To Err Is Human: Building a Safer Health System,” medical errors remain a vexing challenge at healthcare settings. Encouraging staff to participate in safety event reporting is a primary strategy in fixing systemic problems that jeopardize patient safety.
“Punitive reports have important implications for reporting systems because they may reflect a culture of blame and a failure to recognize system influences on behaviors. Nonpunitive wording better identifies factors contributing to safety concerns. Reporting systems should focus on patient outcomes and learning from systems issues, not blaming individuals,” the co-authors wrote.
The research, which was conducted at Richmond, Virginia-based VCU Health, examined more than 500 safety event reports from January to June 2019. The study includes several key data points.
- 25% of the safety event reports were designated as punitive and 68% of the reports were designated as nonpunitive
- Punitive safety event reports compared to nonpunitive reports were more likely focused on communication (41% vs. 13%), employee behavior (38% vs. 2%), and patient assessment (17% vs. 4%)
- Nonpunitive safety event reports compared to punitive reports were more likely focused on equipment (19% vs. 4%) and patient or family behavior (8% vs. 2%)
- More nonpunitive safety event reports involved patient harm than not (5% vs. 2%)
“A high frequency of punitive reports may reflect a culture of blame and retribution, rather than a just culture focused on learning and improvement,” the research article’s co-authors wrote.
They wrote that there are two primary strategies to promote a nonpunitive safety event reporting regime. First, training staff to use safety event reporting to focus on creating a high-reliability organization and a just culture. Second, using alternative resources or tools rather than safety event reporting to disclose problematic behavior by colleagues such as intimidation.
Crafting an effective safety event reporting regime
One of the co-authors of the research article—Robin Hemphill, MD, chief quality and safety officer at VCU Health—told HealthLeaders that an effective safety event reporting system has five components.
1. Leadership: A good safety reporting system is supported by leadership. There needs to be an expressed desire from leadership that they do not want to punish people—they want to understand vulnerabilities in the healthcare organization’s system.
2. Accessibility: A good safety reporting system is easy to access. If you must go searching on the intranet of your healthcare organization to find the patient safety reporting system, people are not going to use it. There should be a desktop icon that makes it easy to find—you log on to your computer and there it is.
3. Ease: Safety reporting should not be lengthy and onerous when people want to report. You may offer a lot of detail that people can report, but you must limit the computer system fields in a required reporting form to the bare elements. A busy healthcare worker should be able to get into the safety report and give enough information in the report so someone can understand the concern. The safety report should include the location of the safety event, a brief description of the event, and the perceived level of seriousness of the event.
4. Receptivity: Management needs to be responsive to safety event reports. If people put safety event reports into the system and it feels like a black hole, then few people will use the reporting system.
5. Training: Healthcare organizations need to train staff on how to use a safety reporting system and help them understand why the safety reporting system is important. Management should remind staff that safety reporting is about fixing systemic problems.
When to focus on individuals
There are circumstances when focusing on an individual’s actions related to a safety event is appropriate, Hemphill said.
“There are times when you must look at the individual. But what we try to do before we leap to whodunit is to try to find out why it happened. Most errors, adverse events, and near misses have an individual at the sharp end of the processes that lead up to that moment in time. So, it is uncommon that people are not involved in these events. But, if you fundamentally believe that people don’t go to work to hurt patients, then you need to understand everything around a healthcare worker that may have contributed to a bad choice or a wrong decision,” she said.
When a patient safety event occurs, managers should conduct a rigorous investigation before laying blame on an individual, Hemphill said. Four key questions should be addressed, she said.
- What is the action that we see? The patient safety event could be an innocent mistake.
- Was there a low or moderate level of risk involved? For example, a nurse could have given a medication without using the barcode scanner, but there could be extenuating circumstances. If the barcode scanner is wireless and unreliable, then it could be rational behavior to not use the device, particularly if management has been alerted to the problem and failed to fix it.
- Was there an unacceptable level of risk involved? To use the barcode example, if the nurse says the barcode scanner works fine but she just decided not to use it because she is smart enough to know which medication to administer, that would be an example of the kind of overly risky behavior that is inappropriate.
- What is the performance of the individual? A just culture cannot tolerate repeated errors. If someone is making the same error week after week, maybe they are just bad at their job and maybe they need to be fired or reassigned to a different area where they can be more successful. Eventually, the competency of the individual must be questioned, but it is not the first step in a just culture.
“Safety reporting that is designed to detect systems deficiencies is complicated and very difficult to conduct. Situations must be managed fairly. You must make sure that you have a complete picture of what is going on if someone is being singled out for a behavioral problem,” Hemphill said.
Christopher Cheney is the senior clinical care? editor at HealthLeaders.