Quality assurance in radiotherapy: evaluation of errors and incidents recorded over a 10 year period

Abstract

Background and purpose: To establish an incident reporting system to (1) record and classify incidents, (2) assess the impact of incidents on patients in terms of dose errors, and (3) evaluate the effectiveness of the quality assurance checking program implemented at the Radiation Treatment Program at the Northeastern Ontario Regional Cancer Centre (NEORCC).

Materials and methods: An 'incident' is defined as an event or a series of events that has led to, or would have led to if undiscovered, dose errors to a patient undergoing radiation therapy treatment. The incidents reported between November 1992 and December 2002 were analyzed according to their source of error, stage of discovery and dose errors.

Results: Between November 1992 and December 2002, 13385 patients have undergone radiation treatment at the NEORCC. Over this period of time, 624 'incidents' were reported. Source of error: the majority of the incidents (42.1%) were related to errors in 'documentation' and most of these could be attributed to 'error in data transfer' or 'inadequate communication'. 'Patient set-up error' accounted for 40.4% of the incidents and about half of these errors were related to shielding. Errors in 'treatment planning' accounted for 13.0% of the incidents. Stage of discovery: independent checks by another dosimetrist/physicist and checking during patient first set-up and port film were effective in detecting documentation errors and errors in treatment planning. The use of portal imaging (Siemens Beamview) has enabled us to detect and correct for more than 85% of reported shielding errors in patient set-up. Dose errors: 40% of the incidents were discovered before the first treatment with no dose error to patients. Overall 97.9% of the incidents had dose error of <5%.

Conclusions: Human errors occur during the various stages of the complex process of radiation therapy. If uncorrected, these could lead to substantial dose errors to patients. The implementation of a quality assurance checking program can substantially reduce these human errors but never totally eliminate them.