Developing a radiation error scoring system to monitor quality control events in a radiation oncology department.

Purpose: The aims of this study were to evaluate the factors contributing to errors in the treatment of cancer patients undergoing radiation therapy and to develop a grading system that allows for the comparison of errors.
Methods: Deviations in the prescribed treatment of patients undergoing radiation therapy were collected during 2003 in the Department of Radiation Oncology at Fox Chase Cancer Center. The deviations were classified according to responsibility as follows: therapist, physician, dosimetrist, physicist, machine, or all. The deviations in treatment were graded on an increasing scale ranging from 1 to 4, according to severity. Error analysis was made corresponding to treatment machine, therapist, dosimetrist, type of treatment, palliative or definitive treatment, type of cancer, time of occurrence, and machine census at the time of occurrence.
Results: A total of 33,757 patient treatments were delivered on 4 linear accelerators with 3,646 dose calculations performed by the physics staff during 2003. All treatments, both intensity-modulated radiation therapy and conventional therapy, were considered in this analysis. A total of 25 quality control (QC) events occurred during the study period. The crude error rate for therapists was 0.041%, and the crude error rate for dosimetrists and physicists was 0.22%. There were 17 level I errors (2 machine, 1 block mismounting, 4 dosimetrist, 1 all, and 9 therapist), 5 level II errors (3 dosimetrist, 1 therapist and dosimetrist, and 1 therapist), and 3 level III errors (3 therapist). Fifteen of the 25 QC events and 8 of 13 therapist events occurred after 12 pm. A correlation did not exist between the time of occurrence and machine census at the time of the QC events. However, more level I events occurred after 12 pm than before 12 pm.
Conclusions: A review of QC events occurring during the course of radiation treatments allowed a change in the process of care to prevent the occurrence of some QC events. The QC event rate experienced by the department compares favorably with published results from similar academic centers. The periodic review of QC events allows for the opportunity to identify processes that can be adapted to reduce the occurrence of QC events in the future.