Community hospital's incident learning system reduces errors, near misses

A small community hospital in the Quad Cities area of Illinois and Iowa implemented an incident learning system in its radiation oncology department, realizing 35 to 40 percent decreases in errors and near misses. All it took was a simple reporting program along with the time and dedication of a safety committee including physicians, physicists, dosimetrists, therapists, nurses and office coordinators.

The group of physicians and technologists from UnityPoint Health Trinity Cancer Center published their research in the Journal of the American College of Radiology.

Traditional error reporting systems in a radiation oncology department rely on dosimetry data, but research has shown errors are most commonly caused by faulty systems and processes, rather than individual mistakes. Armed with the framework laid down by the 2005 Patient Safety and Quality Act, the authors set out to create a simple yet effective reporting system.

“We have implemented a simple incident learning system (ILS) on the basis of the following steps: catch an incident, near miss, or unsafe condition; write a report; analyze what process step needs to be changed; create checklists, encounters, and workflow modifications, establish safety barriers, and write meaningful policies and procedures; and monitor progress,” wrote lead author and lead radiation physicist Murshed Hossain, PhD, et al. “Emphasis was placed on the paradigm shift in voluntary event reporting and the importance of creating and maintaining a nonpunitive and positive environment.”

The reporting form is a lean program built with Microsoft Excel and Visual Basic, and shortcuts were placed on every desktop computer ion the department. Simplicity is the name of the game—there are only four fields; name of the person reporting, process step when the event is discovered, patients medical record number, and free-text box for describing the incidents.

Hossain and his co-authors recorded 299 events by the end of the 354-day reporting period, with therapists reporting the most events with 178. Physicists and dosimetrists reported 92 and 17 events, respectively, while physicians reported just four events over the course of the year.

Most events happened during the first month of reporting, amid the planning stage of treatment. In response, the authors began using a dosimetry checklist featuring items such as correct patient orientation, appropriate calculation algorithm used, break pointed added, etc. Researchers saw a considerable reduction in the amount of errors during the planning phase, but eight months later the errors spiked back up.

“It turned out that complacency can be a problem,” wrote Hossain et al. “Attention can fade, and the process of checking off items can become routine, leading to checking off boxes without making sure that the items have actually been checked.”

At the end of the nearly year-long survey the authors measured a 40 percent decrease in incidents, copmparing favorably to the national average (shown above in red).  

“It is possible for a community hospital to establish and rapidly implement an ILS. An ILS can be implemented rapidly by focusing on the origin of events dictating modification in the workflow,” wrote Hossain et al. “Patient safety can improve continuously, and its progress can be monitored through various metrics including adverse ratios.”