Main Session
Sep
28
PQA 03 - Digital Health Innovation and Informatics, Patient Safety & Quality, and Radiation and Cancer Biology
2476 - Identifying Risk Factors in the Re-Irradiation Using Risk Analysis - A Multi-Site Single-Institution Experience
Presenter(s)
Yulun He, PhD - University of Washington, Seattle, WA
Y. He1, J. Kang1, J. Meyer2, S. R. Bowen1, J. Takayesu2, D. Melancon2, S. Apisarnthanarax2, A. A. Yorke3, and E. C. Ford3; 1Department of Radiation Oncology, University of Washington/Fred Hutchinson Cancer Center, Seattle, WA, 2Department of Radiation Oncology, University of Washington, Seattle, WA, 3Department of Radiation Oncology, University of Washington and Fred Hutchinson Cancer Center, Seattle, WA
Purpose/Objective(s):
As cancer survivorship increases, re-irradiation (reRT) has become more prevalent yet remains high-risk due to complex workflows and radiobiological uncertainties. Notably, reRT-related errors are among the most serious in large national databases (AAPM TG-275 and RO-ILS). Accordingly, this study systematically evaluates the reRT workflow to identify and mitigate risks across a multi-modality, multi-site institution.Materials/Methods:
A multi-professional team performed a Failure Modes and Effects Analysis per the AAPM TG-100 methodology. The team chronologically mapped the pre-treatment reRT workflow from intake through final physician peer-review, identified core tasks and failure modes (FMs) along with their causes and effects, and calculated Risk Priority Numbers (RPNs).Results:
Analysis of 14 tasks and 29 FMs across intake, physician, dosimetry, and physics roles (2, 5, 3, 4 tasks; 5, 8, 4, 9 FMs, respectively) yielded a median RPN=50 (range: 7–280) (Table). Ten high-risk FMs (RPN>100) arose primarily from unawareness of prior treatments, incorrect non-overlap assumptions, incorrect plan identification during peer review, limited radiobiological understandings, and inattention, leading to omission of the reRT pathway or generation of suboptimal treatment plans. Two FMs reached RPN >200; both were of physician-dependent and were largely driven by to poor detectability. Based on these findings, targeted interventions were implemented, including integration of prior radiation history into simulation notes and standardized templates for composite dose review during chart rounds.Conclusion:
High-risk FMs in reRT were identified in all stages of the workflow and were primarily associated with incorrect initiation of the workflow and gaps in final peer-review. Formalized risk analysis, such as described here, can identify these vulnerabilities and guide targeted interventions to safely manage the growing volume and complexity of reRT.| Rank | Role / Task | Failure Mode | Cause(s) | Effect(s) | Severity | Occurrence | Detectability | RPN |
| 1 | Physician / creates simulation order | Physics consult not requested | Non-overlap assumption; inattention; unaware of priors | Omitting reRT pathway | 7 | 5 | 8 | 280 |
| 2 | Physician / peer review | Composite not reviewed | Unclear biological vs. physical dose; unaware of priors; composite not generated | Omitting reRT pathway | 8 | 3 | 9 | 216 |
| 3 | Physician / peer review | Wrong plan reviewed | Confusing plan naming; multiple priors; unproved plan displayed | Omitting reRT pathway | 8 | 2 | 9 | 144 |
| 4 | Physician / Independently confirms prior RTs | Confirmation omitted | Inattention | Omitting reRT pathway | 7 | 3 | 6 | 126 |
| 4 | Physician / creates planning instructions | clinical goals omitting priors | Inattention | Omitting reRT pathway; suboptimal plan | 6 | 3 | 7 | 126 |
| 6 | Physician / creates planning instructions | Intentionally omitting biological dose constraints | Limited radiobiological understandings | Suboptimal plan | 8 | 5 | 3 | 120 |