Objectives: To assess the risk of bias in multicenter randomized controlled trials (RCTs) investigating laparoscopic colorectal cancer surgery and review the use of quality assurance mechanisms to reduce performance bias.
Background: RCTs represent the criterion standard comparison for health care interventions. For trials investigating interventional techniques, performance bias can arise through variation in delivery of the intervention.
Methods: A comprehensive systematic review was undertaken using MEDLINE and EMBASE databases to identify all large RCTs investigating laparoscopic colorectal cancer surgery. Risk of performance bias was evaluated through assessment of publications and protocols to identify methods used for quality assurance of surgical technique. In addition, the Cochrane Collaboration's “risk of bias” tool was used to evaluate other potential sources of bias.
Results: The literature search identified 48 publications, reporting upon 8 individual RCTs. All studies used mechanisms for quality assurance of laparoscopic colorectal surgery. Methods employed included credentialing of surgeons or units through assessment of experience and expertise, standardization of surgical technique, and monitoring. None report the use of structure objective assessment tools for accrediting expertise. All 8 were assessed as low risk of bias using the Cochrane tool. A framework is proposed for use as a model for quality assurance in future surgical trials.
Conclusions: Consideration of risk of performance bias is important when appraising trials investigating an interventional technique. Laparoscopic colorectal surgery RCTs have all employed quality assurance mechanisms to reduce risk of performance bias. Further research is indicated to investigate adopting objective assessment tools for quality assurance within multicenter RCTs.