Radiotherapy Application of Causal Analysis Based On Systems Theory (CAST)
L Wong1*, N Leveson1, T Pawlicki2, (1) Massachusetts Institute Of Technology, Cambridge, MA (2) UC San Diego, La Jolla, CA
Presentations
SU-D-TRACK 6-5 (Sunday, 7/25/2021) 2:00 PM - 3:00 PM [Eastern Time (GMT-4)]
Purpose: Radiation therapy (RT) accidents are commonly investigated with Root Cause Analysis (RCA) based on chain-of-events models. This work aims to analyze an RT incident using a technique based on a systems- and control-theory model and to describe early impressions of this approach.
Methods: Causal Analysis based on Systems Theory (CAST) was applied to an RT incident recreated as a publicly-available video (i.treatsafely.org, #118350879) involving a pediatric patient that received a suboptimal treatment. We performed the five parts of the CAST methodology: 1) gather basic information about the accident, 2) create a safety control structure, 3) analyze the relevant components, 4) identify systemic factors, and 5) create a safety improvement program. These steps were performed by an experienced clinical physicist who has extensive experience with RCA and an expert in systems safety engineering and CAST. A qualitative comparison between CAST and RCA was completed.
Results: A diverse set of contextual factors contributed to the incident. Including the mental model flaws of the frontline staff, these contextual factors revealed additional contributions by technology providers, management, as well as regulatory and advisory bodies. Four systemic factors (e.g., communication and coordination, economics, etc.) were also involved in the incident. Multiple intervention ideas were generated covering technology, practice change, and management to prevent incident reoccurrence. Overall, CAST enabled a robust analysis of the accident. For the non-safety expert, the CAST analysis forced incident analysis from multiple levels which is different than the conventional RCA approach that primarily targets frontline personnel. However, the CAST methodology is more complicated to use than RCA for the non-safety expert.
Conclusion: While CAST appears to be a more robust methodology than RCA, achieving its potential requires prior knowledge about the procedure. Also, some terminology modifications can be beneficial to make it more easily comprehensible when introducing it to clinicians.
Handouts
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email