What outcomes should orthodontic researchers be measuring?
This post is about my last paper in the AJO-DDO. This was concerned with finding out what we should measure in orthodontic research. As a result, we hope to improve the relevance of research to both patients and orthodontists.
Regular readers of my blog will know that I often draw attention to problems with what we measure in orthodontic research. This is because we tend to include measures that are only relevant to orthodontics. For example, large complex cephalometric measurements that have little meaning (cephalometric festivals). In addition, to very precise occlusal indices and now ever more complex CBCT analyses. These levels of measurement result in heterogeneity in systematic reviews and simply increase clinical uncertainty. Furthermore, most of our outcomes are not relevant to our patients.
The final consequence of this is that as cannot agree on what we measure we may be missing the important effects of treatment. As a result, a large amount of research may simply be wasted.
What is the solution to this?
One solution is to develop Core Outcome Sets (COS). A COS is an agreed set of outcomes that should be reported in clinical trials for a specific area of healthcare. Significantly, these are developed by both the providers of care, our patients and the public. We developed a COS for orthodontic trials.
A team from London, Manchester and Sheffield (North of England) did this study to develop a COS for Orthodontics. The AJO-DDO published the paper.
The AJO-DDO has enabled open access for this paper. Padhraig and I were part of the large team that did this work. The lead investigator was Aliki Tsichlaki, and she joins us in writing this post.
What did we want to do?
We aimed to
“Identify a set of clinician and patient-informed core outcomes for use in clinical trials of orthodontic treatment”.
What did we do?
Aliki led a lengthy and complicated study that was divided into several main stages. I am not going into these now as do not have enough space. But, these were:
- A scoping review to find outcomes that have previously been used in orthodontic research.
- A series of qualitative interviews to find out the motives for having treatment and the expectations of orthodontic patients. These were having treatment in five UK settings on the National Health Service and privately.
- A computerised Delphi survey that collected information on the importance of the outcomes we identified from the scoping review and interviews. Orthodontists and patients all over the world ranked the importance of the outcomes. We approached people using social media (this blog) and membership lists of societies.
- A final face to face consensus meeting where a group of patients and orthodontists discussed the ranked outcomes.
- The final set of outcomes was developed by the Study Advisory Group.
What did we find?
Firstly, the team identified outcomes from 164 trials in the review. They held 7 focus groups and 16 qualitative interviews for a total of 35 participants. The Delphi exercise ranked 34 outcomes. The participants in this part were 28 general dentists and 196 orthodontists from 64 countries. Finally, 14 people attended the final consensus meeting to review the importance of the outcomes developed from the earlier stages. The panel gave a high ranking to the impact of orthodontics on feelings/emotions.
At the end of this process, we identified a final core outcome set. These were:
- Impact of self-perceived aesthetics
- Alignment or occlusion
- Skeletal relationship
- Adverse effects on teeth or supporting structures
- Patient-related adherence
We felt we developed a good COS that incorporated the point of view of clinicians and patients.
We made several conclusions. I thought that the most crucial recommendation was;
“Clinical trials that evaluate the whole duration of orthodontic treatment should include each outcome from the final set, apart from stability”.
What did I think?
I thought that this was a great paper! But I am conflicted because I was one of the authors! We started this work nearly 7 years ago with our initial work looking at orthodontic outcomes and COS. As a result, I was really pleased that I had stayed on this team to help complete the task.
As with all research, other steps are needed. For example, we need to decide how to measure the outcomes. For example, how do we measure skeletal relationship? My feeling is that we should just select one simply measure. More importantly, how do we measure the impact of treatment on our patients’ perceptions of their aesthetics. I cannot help feeling that this is the most crucial measure.
I hope that this work continues and we can start to make real progress in measuring what matters to our patients. We may even get rid of the cephalometric festival!
Emeritus Professor of Orthodontics, University of Manchester, UK.