Evaluation of frequency and type of errors detected by a computerized record and verify system during radiation treatment

Background: Computerized record and verify systems (RVS) have bean introduced to improve the precision of radiation treatment delivery. These systems prevent the delivery of ionizing radiations when the settings of the treatment machine do not match the intended parameters within some maximal authorized deviation. Purpose: To assess the potential alteration of the frequency of errors associated with the use of RVS during radiation treatment delivery. Materials and methods: The software of the RVS was altered in order to record the settings actually used for radiation treatment delivery whereas the verification function was suppressed. At the end of the study period, the settings used during daily administration of radiation treatment were compared to the parameters recorded in the RVS using the computer. They were also compared with the planned ones written in the patient treatment chart. Results: Out of the 147 476 parameters examined during the study period; 678 (0.46%) were set erroneously. At least one error occurred in 628 (3.22%) of the 19 512 treated fields, an erroneous parameter was introduced in the RVS memory in 22 (1.17%) of the 1885 fields. Conclusions: RVS has the potential to improve precision of radiation treatment delivery by detecting a significant number of setting errors. However, excessive confidence in RVS could lead to repeated errors as there is a potential for the entry of erroneous parameters into the RVS memory. (C) 1999 Elsevier Science Ireland Ltd. All rights reserved.