I received an inquiry from our Institutional Review Board about a study they were doing a continuing review for. This was a non-randomized comparison of two surgical techniques.
The first question you might ask is: “Why didn’t they randomize the treatments?” In general, it is difficult to randomize in a surgery trial. In a 2002 BMJ article [Medline], McCulloch offers a wide range of reasons.
- Unlike drug trials, there are no large commercial sponsors of surgical research.
- Emergency surgery often makes consent and randomization impossible.
- Many surgical techniques are rare and cannot accumulate sufficient sample sizes.
- The learning curve for new techniques make it impractical two different surgical methods in the same practice.
- Outcomes may be more strongly related to the skill of the particular surgeon rather than the technique used.
- Variations in an operation may occur during the operation itself and these variations can often influence the outcome.
- While it is possible to monitor the purity of a drug, it is difficult to assess the technical quality of an operation.
- Patients may be unwilling to allow randomization to choose the type of surgery they will receive.
This last point was also emphasized in a 1994 World J Surg article [Medline], where Plaisier noted that most patients did not want to participate in a randomized study of extracorporeal shock wave lithotripsy with open cholecystectomy, and that a newer technique, laparoscopic cholecystectomy, was so popular that it would probably be impossible to ever run a randomized trial for this surgery.<U+FFFD>
In a 1984 NEJM article [Medline], Taylor noticed poor enrollment in a randomized comparison of mastectomy approaches and asked physicians why. Some of the reasons offered included:
(1) concern that the doctor-patient relationship would be affected by a randomized clinical trial (73 per cent);
(2) difficulty with informed consent (38 per cent);
(3) dislike of open discussions involving uncertainty (22 per cent);
(4) perceived conflict between the roles of scientist and clinician (18 per cent);
(5) practical difficulties in following procedures (9 per cent); and
(6) feelings of personal responsibility if the treatments were found to be unequal (8 per cent).
In a 1995 BMJ editorial [Medline], Russell argues that most of the constraints on randomization in surgical trials can be overcome with sufficient effort.
The use of placebos in randomized trials of surgery is extremely controversial. I talk about this at length on my page about placebos, and will not repeat the discussion here.
Lots of other areas of medicine have trouble with recruiting patients for randomized trials (would you volunteer for a randomized trial of birth control methods, especially if you knew that one of the arms of the study was a placebo arm?). Still, the problems are thoughtfully documented for surgery trials, and this offers us a good example on some of the constraints on randomization.
You can find an earlier version of this page on my original website.