Room: AAPM ePoster Library
Purpose: The AAPM TG-275 report recommends the use of tools such as checklists and automation to increase the effectiveness of physics chart review, and thereby reduce errors. The purpose of this work is to evaluate the utility of a commercially-available automation tool and freely-available checklists for reducing errors in radiation therapy.
Methods: An automated plan checking tool (ClearCheck, Radformation, New York, NY) and in-house developed intelligent dynamic checklists (IDCs) implemented within the ARIA oncology information system (Varian Medical Systems, Palo Alto, CA) are evaluated. IDCs for physics initial plan review and physician peer review were considered. Using the highest risk failure modes identified in the AAPM TG-275 report, Table 1.A, the number of failure modes addressed by ClearCheck and by checklists were determined. Overlapping and complementing check coverage was also evaluated.
Results: The TG-275 report identified 46 failure modes with highest risk to patient safety. ClearCheck is currently able to automatically check 13% of these failure modes. The IDC for physics initial plan review prompts the user to check 78% of these failure modes, including the 13% covered by ClearCheck. When both tools are coupled with an IDC for physician peer review, failure mode coverage increased to 97.8%.
Conclusion: ClearCheck is capable of automating the check of many items, notably dosimetric constraints and target and organs-at-risk planning margins, which is a qualitative advantage over checklists. However, its ability to check critical failure modes as identified in TG-275 is currently limited and should be used as part of a comprehensive physics plan review and physician peer review strategy, including low-cost, easily implemented and widely-available checklists. Automated plan check tools may be most effective when used to offload parametric and binary condition checks, increasing time available and decreasing mental workload so that physicists may focus review efforts on complex plan quality issues.