It's the little THANGS: Recording frequently unreported errors for dosimetry quality improvement

The reporting of errors resulting in dose deviations are well-studied. Less studied is the amount of inconsequential errors that will not harm the patient but could lead to inefficiency. This paper reports an institutional effort to quantify and reduce these less significant errors. Dosimetry items discovered during physicist plan/record and verify (R&V) check prior to treatment were recorded in a shared document and called Therapy Anomaly Gathering System (THANGS) and individual items were called a "thang." Items were categorized to 1 of 4 types: Treatment Plan, Plan Document, R&V, and Secondary MU. The aggregate numbers were presented to the dosimetry staff at regular staff meetings. It was emphasized to the staff that this was a Quality Improvement (QI) study and would not be used punitively. Thangs were tracked over a 4-year period. In Q1 of year 1 of the study, the average number of errors identified was 179/month. This was reduced to 114/month by Q4 of year 1 and 68/month by the end of year 4, a 62% reduction. The number of errors/plan in Q1 Year 1 was 1.25, and that was reduced by Q4 Year 4 to 0.4, a 68% reduction. The percentage of errors by type did not vary much over the 4 years. By far, R&V errors were the most common, and QI efforts were primarily aimed at them. We have developed a simple method to identify areas in dosimetric work that are vulnerable to minor errors and, through consistent reminders, drastically reduce them. This leads to a seamless throughput for a given plan ultimately resulting in improved physics, therapist, and most importantly patient satisfaction.