gms | German Medical Science

Objective: Critical incident monitoring is an important tool for quality improvement and maintenance of high safety standards. It was developed for air-travelling safety and is now widely accepted as a useful quality improvement technique for reducing medical-care related morbidity and mortality. It has been applied by most medical subspecialties in Germany and internationally. Despite this widespread acceptance, literature does not report of neurosurgical applications. Nevertheless like in other subspecialties critical incidents may have severe consequences for neurosurgical patients. Therefore we decided to establish and evaluate the usefulness of a critical incident reporting system (CIRS) for quality control in a neurosurgical department.

Methods: Data collection started in mid-September. All staff members of the neurosurgical department (physicians, nurses, OR-staff, physiotherapists and pharmacists) were asked to report critical incidents. The anonymous reporting form contains a space for describing the incident, several multiple-choice questions about specific risk factors, place and assumed cause of the incident, severity of the consequences and a space to suggest counter-measures. All reporting forms are collected centrally and anonymously. The incidents are transcribed in a specific on-line documentation system (ADKA-DokuPik). The data is analysed by the department’s quality-representative and an external specialist. Every two month the staff is informed about severe security breaches and the suggested counter-measures.

Results: Initially reporting rate is about 25 incidents per month (presently 51 in total). Human error was identified in 35 incidents (85%). The most frequent factors underlying errors were communication problems (39%), wrong assessment of the situation (22%), and excess of work load (12%). Most of the reported critical incidents occurred on the ward (56%), followed by the operating room (30%). 81% of the critical incidents so far reported were considered preventable.

Conclusions: Implementation of the reporting system was uncomplicated. Directly after launching of the project a high number of formerly unnoticed incidents were reported. The reporting rate is very good compared to literature. As most incidents were considered preventable, identification of these incidents should have the potential to lower the incidence rate by improving communication, increasing specific neurosurgical knowledge and lowering the work load.