Critical incident reporting over time: a retrospective, descriptive analysis of 5493 cases

As recommended by the WHO and many national healthcare authorities, health care institutions of most industrialised countries have employed a critical incident reporting system (CIRS). However, little is known about differences in critical incidents across clinical specialties, the use of CIRSs amongst different professional groups, the types, severity and risk of reoccurrence of critical incidents, their contributing factors and the preventive actions taken in response.

In this retrospective, descriptive study we critically reviewed all reports filed in the CIRS of our institution between 2013 and 2019 and analysed characteristics over time.

Of the 5493 analysed incidents, the main types were related to medications (32.8%), clinical procedures (32.6%) or behaviour of employees (23.3%). Only 21.6% of reports were made by physicians, 51.3% were rated at least "high risk". Major contributing factors were personal factors (44.0%), lack of training and knowledge (43.7%) and communication errors (36.1%). Most actions taken to prevent similar events aimed at improving communications (23.6%); in 46.3% no actions whatsoever were taken. Longitudinal analysis revealed that reporting increased in internal medicine and obstetrics but collapsed in laboratory medicine / microbiology. Steady increases in medication-type incidents as well as lack of training and knowledge as contributing factors were observed.

This study revealed that the efferent loop (feedback, preventive actions taken) was the weakest part of the CIRS in our institution, indicating that no learning may have resulted from a large number of reports. In particular, the actions taken appeared to not adequately address the major contributing factors. This highlights that special attention must be paid to the efferent loop of a CIRS to fulfil the purpose of such a reporting system and ultimately to improve patient safety.