Reported patient safety incidents in radiology - understanding the relation between risk areas and their underlying...

Learning from errors and underlying causes creates important knowledge, enabling re-evaluation of routines and preventing recurrence of errors. Because of the nature of activities in radiology, measurement of specific risk areas is needed. The aim of this study was to analyse reported events in radiology and identify the underlying causes in relation to the risk.

Patient Safety Reporting System (PSRS) data from three radiological clinics were retrieved, covering 923 risk events. The reports were studied deductively based on a risk assessment framework (i.e. risk areas) using qualitative content analysis.

The events were distributed between six risk areas covering risks that "the patient could be exposed to unnecessary radiation" (7 %); "the patient could receive an inaccurate diagnosis" (26 %); "the patient could incur drug-induced damage" (2 %); "the patient could suffer direct physical injury" (9 %); "the patient's examination and treatment could be delayed or not carried out" (47 %); and that "the patient's general health condition could deteriorate" (9 %). Twenty-two subcategories were identified and linked to these risk areas, representing underlying causes.

This study validates a previously designed risk assessment framework for radiology. Collaboration complexities between internal and external actors are evident, particularly when utilizing external practitioners and teleradiology services. Because of capacity issues, radiological expertise is often not given due regard, and risks escalate when information systems fail to meet radiology documentation requirements. To improve patient safety, radiological competence needs to be maintained, established standards and quality controls must be followed, and documentation and reporting need to be accurate.

An awareness of long-term effects of errors in radiology needs to be created, and collaboration challenges for improved patient safety and well-timed diagnoses need to be solved.