In my previous blog post I told about our letter to editor concerning a study criticizing sham-controlled studies in orthopaedics.
Another letter was also submitted and accepted at the same time by Harris, Poolman and Buchbinder. They had also very important aspects regarding the criticism towards sham-controlled RCTs:
We agree that there are other methods (e.g., the method and timing of randomization, blinding the statisticians, ensuring high rates of follow-up) that would add to the reduction of error, but to “invalidate” such high-level (by the authors own ratings) research implies that we should instead rely on studies that do not include blinding, are not randomized, or do not contain a comparator. To do so is to accept inferior methods of measurement, simply because the best methods are not perfect.
Original authors responded to both our letter and that by Harris et al. in a single response. Firstly, the response of the original authors was almost 3000 words long. The journal instructions say that Letter to Editor and reply should be limited to 500 words. Original authors responded to two letters but still, it was not a quite fair game. We could have also used additional 1000 words to discuss these issues.
Original authors replied:
The randomized controlled trial (RCT) has always been the ideal investigation to identify whether an intervention works: Is an experimental surgical technique (e.g., knee arthroscopy) “better” when compared with a control group (i.e., superior, non-inferior, or equivalent) in ideal (efficacy) or real-world generalizable (effectiveness) settings? In theory, randomization may prove causation. However, in practice, randomization may fail.10
I think it is ludicrous that when overall methodology in orthopedics is below average and has been that for a long time, these aspects in randomization are discussed for the first time when truly high-quality studies are finally published in our field. Theoretical aspects are discussed only when age old paradigms are challenged. Whether randomization sometimes fail or does not fail is not really what we need to be discussing at this moment. We need to discuss those things which belong to the “former paradigm” in the clinical practice. Much of orthopaedics is still practiced according to this “former paradigm“.