Clinical studies in orthodontics--an overview of NIDR-sponsored clinical orthodontic studies in the US.

A number of clinical trials sponsored by the National Institutes of Health (NIH) use rigorous methods of data acquisition and analysis previously developed in fundamental biology and the physical sciences. The naive expectation that these trials would lead relatively rapidly to definitive answers concerning the therapeutic strategies and techniques under study is dispelled. This presentation focuses on delineating differences between the study of central tendencies and individual variation, more specifically on the strategy to study this variation: measure additional sources of variance within each patient at more timepoints and perhaps with greater precision. As rigorous orthodontic research is still in its infancy, the problem of defining the proper mix between prospective and retrospective trials is discussed. In view of the high costs of prospective clinical trials, many of the questions germane to orthodontics can be answered by well-conducted retrospective trials, assuming that properly randomized sampling procedures are employed. Definitive clinical trials are likely to require better theoretical constructs, better instrumentation, and better measures than now available. Reasons for concern are the restricted resources available and the fact that current mensurational approaches may not detect many of the individual differences. The task of constructing sharable databases and record bases stored in digital form and available either remotely from servers, or locally from CD-ROMs or optical disks, is crucial to the optimization of future investigations.