Orthopaedic surgeons prefer to participate in expertise-based randomized trials

Abstract

Empiric data and theoretical arguments suggest an alternative randomized clinical trial (RCT) design, called expertise-based RCT, has enhanced validity, applicability, and ethical integrity compared with conventional RCT. Little is known, however, about whether physicians will participate in an expertise-based RCT. In a cross-sectional survey of Canadian orthopaedic surgeons, we evaluated preference for and willingness to participate in an expertise-based versus a conventional RCT if given the opportunity to participate in a trial investigating the effectiveness of high tibial osteotomy versus unicompartmental knee arthroplasty. Using an electronic survey ((c)2005 SurveyMonkey.com), we invited all 767 members of the Canadian Orthopaedic Association (2005) to participate; 276 surgeons completed the questionnaire (37.5% response rate). One hundred two surgeons (53.4%) were willing to participate in an expertise-based RCT compared with 35 surgeons (18.3%) willing to participate in a conventional RCT. Ninety-seven surgeons (52.4%) strongly or moderately preferred the expertise-based design compared with 25 (13.5%) who preferred the conventional design. For the clinical example we presented, the majority of Canadian orthopaedic surgeons were willing to participate in and preferred the expertise-based design. The expertise-based randomized clinical trial design may overcome some of the barriers to conducting clinical trials in orthopaedic surgery and improve the validity of their conclusions.

Fig. 1A–B

Comparisons of (A) conventional randomized clinical trial (RCT) and (B) expertise-based RCT designs are shown. In the conventional randomized clinical trial (A), patients are randomized to one of two surgeries (Surgery A or Surgery B), and surgeons administer Surgery A to some participants and Surgery B to others regardless of the surgeon’s level of expertise and/or preference. In an alternative RCT design (B), patients are randomized to surgeons with expertise in Surgery A who are committed to performing only Surgery A or to surgeons with expertise in Surgery B who are committed to performing only Surgery B. This alternative design is referred to as an expertise-based RCT.

Fig. 2A–B

We constructed forest plots to illustrate the proportion of respondents in successive survey waves who (A) preferred and (B) were willing to participate in an expertise-based (EB) randomized clinical trial (RCT). The pooled estimate is calculated using a random effects meta-analysis of successive survey waves. Error bars represent 95% confidence intervals (CIs).

Fig. 3

Surgeons prefer to participate in an evidence-based versus a conventional RCT for comparison of outcomes after HTO versus UKA in patients with medial compartment osteoarthritis.

Fig. 4

Surgeons’ level of expertise in a particular intervention influences their opinion regarding the relative superiority of that intervention. Expertise is defined as having performed 10 or more of a procedure during the past year.