Can orthodontics be evidence-based?
I receive many comments on my blog and individual posts. These vary from praise to extreme personal criticism. Recently, Jean-Marc E. Choufani, a first-year resident from a programme in New England, made some tremendous perceptive comments following my recent post on MSE. I thought it would be good to publish these in full along with my response.
As much as I admire your systematic and thorough approach to evaluating the articles you review, I sometimes wonder whether certain conclusions regarding quality of evidence and risk of bias are genuinely worth the assigned weights in day-day practice.
After reviewing the book “evidence-based orthodontics” by Huang et al. – I found it quite entertaining how very few procedures and clinical decisions in orthodontics are truly evidence-based.
Does this absence of evidence mean it doesn’t work? Should we focus more on evaluating the cost-benefit ratio for each patient when proposing treatment plans? Of course, it should be considered incompetence to suggest something sure to cause harm to the patient. However, how do we assess this in orthodontics?
When looking at some of the biological effects of orthodontic treatment, many patients are arguably ‘healthier’ before orthodontic treatment, especially when side-effects of treatment and mechanotherapy are not anticipated.
Although the implications of these comments may seem to defeat the purpose of carrying out studies as the scientific method dictates, I’d like to mention an article published by Straus and Sackett in 1999 titled “Applying evidence to the individual patient” that provided me with some clarity when thinking about this post.
Non-lethal nature of orthodontics
Due to the non-lethal nature of orthodontic treatment, the ‘medical’ approach to evidence-based practice may not be as crucial as in other medical specialties. In theory, offering an approach that might be more invasive and expensive, for example, the techniques advocated by Wilmes or De Clerk) maybe logical. Particularly if the treatment has the potential to provide a result that may surpass conventional treatment. We may also eliminate the need for appliances such as extra-oral headgears and facemasks.
However, this doesn’t mean one should advocate trying everything and anything available in the market. Some may argue the need for statistical and clinical significance evaluated in a 6-10mm profession may be unwarranted when considering all the factors involved in certain types of orthodontic treatment. Might it not be more appropriate to develop a system of guidelines that can help gear practitioners make better decisions, not based on experience, but on viable and effective options?
Besides statistical significance, we should consider why specific methods are gaining more traction other than being novel. Although I’m taking the stance of a ‘devil’s advocate,’ I wonder about certain conclusions made about clinical studies in a field that tackles so much biological variability where ‘all roads lead to Rome’, so to speak. I understand that there is a dire need for a scientific approach to orthodontic treatment. Indeed, we may realign the focus to address the most purposeful and effective techniques for tackling the problem. However, this may not be enough. Certainly, patient psychology and quality of life should always determine why one treatment could be selected over another. Perhaps, we need to develop these outcomes further?
I would like to share a disclaimer that I am currently an orthodontist in training. My comments reflect the limited information I’ve gained over the past few months in my residency. I’ve shared my reflections to stir up discussion. I’m aware that this may lead to criticism, and it is welcomed! However, can we truly move forward and evolve unless we share our philosophies and put our ideas to the test? I’ll leave that for you to decide.
What did I think?
I thought these were all great points, and I would like to respond as best I can. So a good place to start would be my first post on my blog. I outlined the aims of this blog. It is worth revisiting. This post was published on September 1st, 2013, which seems long ago. I stated
“Over the past ten years there has been a marked increase in the number of high quality prospective trials that have been published. They have provided clear evidence on the effectiveness of our treatment and delivery of care. However, at the same time there has been greater changes to practice that occurred because of the effect of statements of so-called “gurus” and advertising. Orthodontists have accepted many of these claims and we have to consider if we are losing our way as an evidenced based specialty”?
This, in effect, is the challenge that Jean-Marc raised. His first point was about whether we have an evidence base because most of our trials are deemed to be at risk of bias and that there is generally an absence of evidence. My response is simple, I agree. Many of our trials are small, and they are at high risk of bias according to most assessment tools used. There are two reasons for this. Firstly, we are still gaining experience in trials, as this is still a relatively new technique for orthodontists. Furthermore, the risk of bias tools are very unforgiving when applied to orthodontics. For example, we can’t blind the operators and patients to the interventions.
We need to consider how to deal with this criticism. I feel that we need to interpret all these classifications in light of the nature of orthodontics and come to our own conclusions. A great example of this has been the recent discussion on intercepting impacted canines. Contributors made comments pointing out that removing a primary canine is a very low-risk procedure. As a result, should continue this practice, even if the evidence for its success is not strong. In effect, we are trying to do good with minimal intervention. The most important thing that we can do is interpret the literature using our critical appraisal skills.
This, of course, is very different from the promotion of methods that are more costly or likely to cause harm to our patients. This is the realm of the few extreme KOLs and snake oil salespeople who blight our speciality. I do not feel that they can hide behind the non-lethal nature of orthodontics.
Popular methods and all roads lead to Rome.
Like many thinking orthodontists, I am heavily influenced by Lysle Johnston. One of his most compelling quotes is that “everything works”. This is entirely correct. I agree with this concept when applied to morphological change from treatment. It is rare to see an intervention that does not do some “good”. However, we do not know the effects of our therapy on patient values. Again, I agree with Jean-Marc about the need to measure quality of life to make conclusions about arguably the most important effects of treatment.
Finally, for the benefit and safety of our patients,we do need to put our ideas to the test. We can only progress by carrying out the optimum research design, increasing education in critical appraisal, applying this to our publications, and finally stopping making unsubstantiated claims on the effect of treatment. It is only then that we can call orthodontic treatment an evidence-based speciality. We are on the way to achieving this, and great comments like this help us think about the way forwards.
Emeritus Professor of Orthodontics, University of Manchester, UK.