ESTRO 38 Abstract book

S329 ESTRO 38

was expected that the embedment of QI processes within the practice would enable both the delivery of accurate and meaningful radiation dose as well as the reduction in the number of imaging related radiotherapy errors (RTE). Material and Methods The Institute for Healthcare Improvement’s Model for Improvement framework utilising plan, do, study, act (PDSA) cycles was employed to help structure and drive quality within this project. Four project measures were identified to monitor progress against the QI aim: subjective image quality, image dose, RTE and staff feedback. The QI programme was initiated in May 2017 and is still ongoing. During this period an RTE audit, frontline staff survey, subjective image quality audit, and a ‘live’ staff and RTE feedback was initiated to inform our intervention strategy, which consisted of 6 PDSA cycles (Figure 1). Following each intervention, relevant project measures were monitored to establish efficacy of the intervention.

Conclusion The use of a QI model to drive continuous improvement within this setting has been highly successful and has resulted in a clear reduction in imaging related RTE. Utilising the PDSA cycles has highlighted that whilst improvements have been made, a key number of interventions remain and require continual development and monitoring. OC-0620 Stop: No! Take Action: Yes! A new approach to act on anatomical changes seen on CBCT M. Buijs 1 , F. Pos 1 , M. Frantzen-Steneker 1 , F. Koetsveld 1 , P. Remeijer 1 1 Netherlands Cancer Institute, Radiotherapy, Amsterdam, The Netherlands Purpose or Objective Currently for the most common treatment sites Traffic Light Protocols (TLP) have been developed to recognize and react to anatomical changes seen on CBCT scans. For the RO’s as well as the RTT’s this method proved to be quite labor-intensive, as it involves alerting the RO, handover of the findings, and finally decision making by the RO. Therefore a new approach was developed to act on anatomical changes: the Take Action Protocol (TAP). In this approach the RTT’s do not only have a role in detecting anatomical changes, but are also deciding on the appropriate action and follow up in a standardized way, resulting in a significant shift in responsibility. The aim of this study was to evaluate the implementation of this “Take Action Protocol” and assess its impact on workload, burden of the RTT’s responsibility, and the accuracy of the reviewing and decision making. Material and Methods During a pilot period from Sept 2017 to Febr 2018 the TAP, designed in collaboration with RO’s and Medical Physicists, was applied for bladder and prostate patients. The protocol consists of a primary decision to act on the anatomical change seen on the online CBCT. When the anatomical change appears to be systematic, a flow chart guides the decision regarding the follow up of the treatment, see figure 1. To evaluate the use of the TAP, CBCT images were retrospectively reviewed by an IGRT specialist to assess the accuracy of the reviewing and decision making. Also the impact on the workload was evaluated by scoring the amount of contact moments with an RO. Two surveys were

Results During the QI period the ‘live’ feedback from staff and RTE identified collimation and image quality as key areas for improvement, which was consistent with the image quality audit. For the staff survey 31 out of 52 therapeutic radiographers responded, with 80% of respondents being involved in some aspect of 2DkV imaging ≥ 5 times per month. There was a good level of confidence in all aspects of 2DkV imaging, however poor digitally reconstructed radiograph (DRR) quality was identified by 35% of staff as a major issue in the image guidance process. Other factors affecting quality related to the imaging workflow, inadequate collimation, and lack of confidence with utilising higher dose protocols. These findings were congruent with the RTE audit results (Figure 2), which also identified inadequate record keeping as a significant contributory factor to RTE. Figure 1 presents a run- sequence plot of RTE per quarter, alongside the timeframe of each PSDA cycle. Across the course of the QI initiative a steady and sustained reduction in RTE was observed, with current quarterly error rates running at one fifth of their pre QI values.