When clinical data is spread in different computer systems of different orthodontic clinics and practices and in different formats, it cannot be collected and used for research easily. Even if a researcher where to get permission and collect a large amount of data (hundreds of thousands of patients) from many different practitioners, he/she would have to harmonize it before being able to perform any kind of analysis on it. The process of going through each data of each practitioner and categorizing it in a common standard way would be so tedious and expensive, that it would render the research not worth it. This is why current clinical orthodontic research is being done only less than 100 patients.
Research would probably lead to different more accurate results, if the sample size was three orders of magnitude greater. And the data is certainly not missing. We have the data, it’s just not formatted in a research friendly way.