Application of an updated severity‐based FMEA approach to Radiation Therapy QA: A multi‐institutional exercise
Alejandra Rangel, Ryan Studinski, Ady Abdellatif, Leigh Conroy, Jenna King, Humza Nusrat, Michael OliverAbstract
Background
External beam radiation therapy (RT) involves many complex steps to safely and reliably treat patients. The radiation oncology community has adopted process maps and failure modes and effects analysis (FMEA) to analyze risk using the TG‑100 methodology, but the traditional TG‑100–endorsed risk priority number (RPN) has recognized limitations. An updated risk‑profile introduced by the automotive industry, action priority (AP), offers a severity‑weighted alternative that may better support risk‑based decision‑making in RT.
Purpose
To compare the traditional RPN‐based FMEA approach with the updated AP‐based approach for identifying high‐risk potential failure modes (PFMs) in a representative external beam RT workflow across multiple institutions, and to assess whether the AP framework provides additional value to the radiation oncology community.
Methods
Nine physicists from different cancer centers in Ontario, Canada collaboratively developed and refined a list of 63 PFMs associated with a generic RT workflow requiring data transfer. Each PFM was scored for occurrence (O), severity (S), and detectability (D) using TG‑100 recommendations. Three scoring scenarios were evaluated: average, worst‑case, and a severity‑focused score. High‑risk PFMs were identified using the traditional (top RPN quartile and/or S ≥ 8) and the updated (High or Medium AP categories) FMEA criteria.
Results
Seven high‑risk PFMs were selected using the updated FMEA, and an additional 23 were selected using the traditional FMEA. All PFMs identified by the updated FMEA approach were also selected by the traditional FMEA approach; however, the traditional FMEA approach identified many additional PFMs due to its quartile‑based and severity‐threshold criteria. The updated AP approach appeared less affected by the scoring variability which may reduce the need for consensus discussion, it was generally easier to apply, showed more consistent behaviour across FMEA iterations and provided a clearer action guidance.
Conclusions
The use of the updated AP approach in this study suggests a clearer, severity‐driven, and more streamlined strategy to prioritize high‐risk PFMs, potentially reducing the burden of multi‑institutional FMEA while maintaining consistent identification of critical risks.