Quality control and error detection in the radiotherapy treatment process

Abstract

Purpose: In 1995, a post of quality control (QC) officer was established in the Radiotherapy Department of the University Hospital Gent (R-UZG). We report here the evolution of error detection in the domain of information transfer during the different steps of the treatment chain, in our department, since the creation of this job.

Material and Methods: From January 1995 to December 1997 (1995 n = 831; 1996: n = 1095; 1997: n = 1091), data on errors were recorded. At the start, an inventory was made of the existing situation and a Quality Assurance (QA) program was outlined for process control. According to the site of origin of errors in the treatment preparation chain, errors were separated into five levels: medical treatment prescription, simulation, treatment planning, treatment data transfer and daily set-up.

Results: The total number of errors found was 459 in 1995; 809 in 1996 and 1046 in 1997. During 1995 and 1996 the medical prescription protocols were adapted to the increasing need of the radiation technologists for more information. This explains an increased number of errors (from 80/459 to 276/809) in 1996. After a period of adaptation, the number of errors decreased in 1997 (257/1046). The second level, where many errors were found, is at the transfer phase (1995: 181/459 1996: 210/809; 1997: 336/1046). Most of these errors were made during the transfer of data from the prescription chart to the computer. These errors were due to lack of attention, human mistakes and calculation errors. The number of errors during simulation increased due to rotation of personnel in 1996. The increase persisted in 1997 for the same reason. Transfer errors due to the automatic transfer of leaf settings decreased (1995:18/29; 1996:15/17; 1997:7/31) Thanks to the start of QC management, errors were detected and corrected in the entire treatment process at R-UZG. Once changes were accepted, new challenges were initiated. After each evaluation, initiatives were taken to try to decrease specific errors. Changing attitudes was a difficult and slow process, but progress was made. The most important change in attitude certainly was the acceptation of the concept of QC.