A clinical trial is a preplanned experiment on humans that aims to assess the effects or benefits of at least 1 treatment. A randomized clinical trial (RCT) uses randomization and is considered the gold standard for assessing intervention effects. Randomization reduces selection bias by randomly assigning interventions and creating similar treatment groups that are in all respects similar except for the intervention and in which any observed differences between treatment groups are expected to be due to chance.

Randomization in clinical trial methodology has a specific meaning and is implemented by using explicit procedures consisting of the following 3 steps: generation of random allocation, allocation concealment, and implementation of randomization. It should be clearly understood that frequently used terms in the orthodontic literature to describe randomization such as consequent allocation , randomly assigned , and allocation based on some deterministic measure such as date of birth, day of the week, or file number are not considered true randomization approaches in RCT methodology and have been associated with biased results. Random number generation is usually produced by using random tables or software algorithms capable of producing random sequences. Several methods such as simple and permuted block randomization, stratified randomization, and minimization have been described as useful in orthodontic research. Simple randomization resembles a toss of a coin, and, although it is truly random, it is problematic especially in small trials because the probability of unequal treatment arms is high. Permuted blocks ensure balanced treatment groups of so many patients and solve the problem of simple randomization, whereas stratified randomization and minimization ensure balanced treatment groups on important prognostic factors of the outcome.

Allocation concealment ensures the unpredictability of treatment allocation by patients and investigators, and is always feasible, unlike blinding, which is often difficult to implement in orthodontic trials. Methods of allocation concealment can include the use of sequentially numbered sealed opaque envelopes, computerized allocation, or allocation assigned from an external center. The best method to achieve allocation concealment is to use an independent and centralized assignment protocol that does not involve any persons associated with the trial. With this method, the randomization lists are generated and held securely in remote locations, thus reducing the chance of peeking at treatment assignments and eliminating the chance of predicting the next treatment assignment.

Implementation of randomization pertains to who implemented the procedures of randomization, and when, where, and how these procedures (generation of randomization lists, allocation concealment, and treatment assignment) were implemented. In small studies, it is likely that the same investigators will perform all tasks. However, it would be best if different persons performed the various randomization procedures, since this would decrease the chance of biasing the process.

It has been reported that lack of appropriate randomization and allocation concealment might introduce bias and thus render the study results invalid. Among the various study designs, the RCT is considered the one with the potential to provide the highest quality evidence. However, there is evidence in the biomedical literature, including dentistry, that randomization and RCT quality, in general, are suboptimal, and often clinical trials described as RCTs are not really RCTs.

The term randomized clinical or controlled trial has become more common in the titles of articles published in orthodontic journals, but no study has assessed whether studies published as RCTs are truly RCTs. Therefore, the purposes of this study were to investigate the extent to which studies published in orthodontic journals with RCT in the title are mislabeled and to identify probable associations between journal type and other publication characteristics on correct classification.