Evaluating the role of expertise-based trial designs in trials of surgical interventions

Randomised Controlled Trials (RCTs) have played a very important role in advancing health care by allowing accurate assessment of a new treatment by comparison with a standard treatment. The field of surgery has been slower to use RCTs than medicine in general. Different options for how to run such studies exist. Conventionally participating surgeons have been expected to perform both treatments under assessment. An alternative approach is to require a participating surgeon to perform the intervention they are most comfortable with. This has been called an expertise-based approach. This is attractive from a surgeon‘s perspective though such an approach does raise some scientific (such a study may need to be larger) and practical challenges (organising corresponding expertise surgeons to be available). The aim of this research is to evaluate the relative merit of the different approaches and specifically, in which circumstances is an expertise-based approach preferable from both a practical and scientific perspective.

While many surgeons accept the need in principle for RCTs to evaluate surgical interventions, they struggle to reconcile their personal involvement with their surgical experience. Under a standard 2 arm surgical trial design, participating surgeons are expected to delivery both interventions. Where interventions differ substantially, surgeons may routinely only conduct one or the other of the interventions. It is likely even where surgeons perform both they have a preference and/or have more expertise in one than the other. As a consequence some surgeons may be reluctant to participate in a standard RCT. The expertise-based trial design, where participating surgeons provide only the intervention in which they have expertise, has been proposed to overcome this problem. Other purported benefits include increased surgical compliance with randomisation, addressing the learning curve effect along with patient preference for such a design. While such a design is not new, the profile of the design has been raised and the use of it is becoming more common. However, this approach has been criticised on a number of grounds including its impact upon sample size, practicality, generalisability and ethics.

The aim of this research is to evaluate the relative merit of expertise-based designs over the standard trial design in surgical evaluations. A three year programme of methodological research is proposed which would seek to answer the following research questions:
1. Under what circumstances is the expertise-based approach an acceptable and feasible design?
2. What impact do the various trial design options have upon the statistical efficiency?

The specific elements of research are:
1. Systematic (methodology) review of the use of expertise-based trials in the surgical and non-surgical medical literature focussing on when the design has been used and implemented.
2. Assessment of the acceptability and feasibility of expertise-based designs through semi-structured interviews with surgeons and trialists. This will consider areas in which the design could be used and is practical.
3. An extensive simulation study evaluating the various options for expertise-based trial design and corresponding statistical analysis and their impact upon statistical efficiency. This will inform design of surgical trials and assessment of the relative trade-off between approaches.
4. Statistical analyses of two expertise-based trials assessing the impact of different analysis strategies.
5. Guidance documentation on the use of expertise-based designs in surgery summarising the published evidence to date and the knowledge gained from elements 1-4 above and recommendations for their use.