Quality of surgery in oncology trials

Abstract

Randomised controlled trials (RCTs) with surgical interventions often lack a framework to ensure surgical quality. Although recent oncology trials, such as ADDICT (D1+ vs. D2 gastrectomy), have sought to monitor surgery there has been no demonstrably reliable tool to assess surgical quality. We aimed to investigate SQA in oesophagogastric oncology trials and to develop a robust framework of consensus strategies to overcome challenges to design and implementation of SQA. A multi-method approach including both qualitative and quantitative methodologies were applied in order to address the research objectives. On systematic review of previously reported challenges to SQA in trials the most commonly encountered included: constraints of using case volume for credentialing surgeons; inter-centre variation in the definition and execution of interventions, and; insufficient monitoring of surgical quality. A meta-analysis of SQA and protocol utilisation within oesophagogastric RCTs revealed public availability of protocols and Eastern country of origin were associated with improved survival. Semi-structured interviews were subsequently conducted with expert stakeholders examining challenges to SQA in trials. Prominent mitigating strategies included operative monitoring using photographs and/or videos with a structured objective assessment tool. Expert consensus was reached for 59 strategies to overcome challenges to SQA in oncology trials. 14 (74%) of the 19 included expert stakeholder proposed strategies from chapter 4 gained consensus amongst ADDICT trial stakeholders within 2 Delphi rounds, indicating their relevance within oesophagogastric oncology RCTs. A patient focus group and survey, established to gain insight into service user perception of quality of surgery, reinforced the importance of considering operative volume and monitoring surgery using a structured methodology. Robust monitoring methods are required to assess surgical quality and oesophagectomy assessment tools were demonstrated to be reliable using generalisability theory. Condensing the expert Delphi consensus allowed formulation of a 33-item framework of strategies to overcome challenges to implementation of SQA in oncology trials (FOSQAT). Given the relevance of the expert Delphi strategies within ADDICT, in future we recommend trial committees and surgeons will not be required to conduct a Delphi process, but rather will be able to select relevant strategies for implementation from the FOSQAT consensus. Clinical validation of this framework assessing impact of implemented strategies on short and long-term outcomes should be the focus of future research in this area.