The Food and Drug Administration Modernization Act of 1997 made it mandatory for all phase II through IV trials regulated by this Act to be registered. After this, the National Institutes of Health created ClinicalTrials.gov , which is a registry of publicly and privately supported clinical studies of human participants. The objective of this study was to examine the characteristics of registered studies in orthodontics.
The ClinicalTrials.gov Web site was used to query all registered orthodontic studies. The search term used was “orthodontics.” No limitations were placed for the time period. All registered studies regardless of their recruitment status, study results, and study type were selected for analysis.
A total of 64 orthodontic studies were registered as of January 1, 2014. Of these, 52 were interventional, and 12 were observational. Close to 60% of the interventional studies and 66.7% of the observational studies had sample sizes of 50 or fewer subjects. About 21.2% of the interventional studies and 16.7% of the observational studies had sample sizes greater than 100. Only 1 study was funded by the National Institutes of Health, and the rest were funded by “other” or “industry” sources. Close to 87.7% of the interventional studies were randomized. Interventional model assignments included factorial assignment (3.9%), parallel assignments (74.5%), crossover assignment (7.8%), and single-group assignment (13.7%). Most studies were treatment oriented (80.4%). The types of masking used by the interventional studies included open label (28.9%), single blind (44.2%), and double blind (26.9%). Outcome assessors were blinded in only 6 studies.
Orthodontic studies registered in ClinicalTrials.gov are dominated by small single-center studies. There are wide variations with regard to treatment allocation approaches and randomization methods in the studies. These results also indicate the need for multicenter clinical studies in orthodontics.
This is an overview of the orthodontic clinical studies in the ClinicalTrials.gov registry.
Most registered studies are single-center studies with sample sizes of 50 or fewer.
There are wide variations about treatment allocation approaches and randomization methods.
Publication bias should be assessed and kept in perspective when interpreting their results.
Our results indicate the need for multicenter clinical studies in orthodontics.
Well-designed and conducted controlled clinical studies are the gold standard for clinical decision making and occupy the highest echelon in the pyramid of evidence-based medicine. However, clinical studies are heterogeneous and differ considerably in terms of conduct of the study, funding sources, randomization protocol, end outcomes, and blinding strategies. Consequently, the results across clinical studies are not consistent, are specific only to the cohorts examined, and are not generalizable. Typically, journals tend to publish results from studies that demonstrate statistically significant findings. Publication bias associated with publication of positive results and nonpublication of negative findings in peer-reviewed journals has precluded us from understanding the true effects of an intervention. Several peer-reviewed journals now mandate that clinical trials be registered before the start of the study. The Food and Drug Administration Modernization Act of 1997 made it mandatory for all phase II through IV trials regulated by the Food and Drug Administration Act to be registered. After this, the National Institutes of Health created ClinicalTrials.gov , which is a registry of publicly and privately supported clinical studies of human participants. Clinical studies conducted all across the world are registered in ClinialTrials.gov . This database provides information on the characteristics of registered clinical studies as well as results from the interventions and the observations. Although several studies have examined the characteristics of registered studies for a wide variety of medical conditions, to our knowledge, no study to date has documented the characteristics of clinical studies in orthodontics that are registered in ClinicalTrials.gov . Such reports will enable us to understand the areas of clinical interest that are the focus of orthodontic clinical studies, study designs, funding sources, and outcome measurements. The objective of our study was to examine the characteristics of registered orthodontic clinical studies. Both interventional and observational clinical trials were examined.
Material and methods
ClinicalTrials.gov was created as a result of the Food and Drug Administration Modernization Act of 1997 and was made available to the public in February 2000. ClinicalTrials.gov “registry” or “results database,” launched in September 2008, is a World Wide Web-based resource that gives patients, providers, researchers, and the general public easy access to information on publicly and privately supported clinical studies on a wide range of diseases and conditions. Each ClinicalTrials.gov record presents summary information about a study protocol and includes information about the disease or condition, intervention (eg, medical product, behavior, or procedure performed), description and design of the study, eligibility criteria, geographic locations where the study is being conducted, and contact information for the study locations. In addition, descriptions of study participants (ie, number starting and completing the study and their demographic data), outcomes of the study, and summary of adverse events experienced by the participants are also available for some records. Submission of adverse event information was optional when the results database was released, but this became required in September 2009.
The ClinicalTrials.gov Web site was used to query all registered orthodontic clinical studies. The search term used was “orthodontics.” No limitations were placed for the time period. All registered studies regardless of their recruitment status (open to recruiting, closed to recruiting, or actively recruiting), study results (with or without results), and study type (interventional, observational, and expanded access) were selected for analysis. The characteristics of the clinical studies examined included types of interventions, outcome measures, types of blinding, study sponsors, geographic origins of study, sexes of participants, ages of participants, and phase of the study.
As per the University Hospitals Case Medical Center institutional review board and in agreement with Federal Regulations 45 CFR 46.101 (b), which states that “research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects,” such studies are permitted to be classified as research that is “exempt” from full or expedited review by the institutional review board. This study was also exempt from institutional review at the College of Dentistry of the University of Iowa. This study was a cross-sectional analysis of orthodontic clinical studies that are available publicly from ClinicalTrials.gov .
The first author (V.A.) and the second author (S.R.) compiled the data using predefined terminology and search parameters. Information on clinical studies was downloaded in a comma-delimited format from the registry and converted into an SPSS (version 21.0; SPSS, IBM Corp, Armonk, NY) data file. Simple descriptive statistics were used to summarize the characteristics of the clinical studies. The same software was used to generate the descriptive statistics.
Since the inception of the trial registry to the start of this study (January 1, 2014), a total of 158,356 clinical studies were registered. These included studies conducted in all 50 states of the United States and 184 other countries. Of these, 128,139 (80%) were interventional (drug or biologic, 84,344; behavioral and other, 32,757; surgical procedure, 14,231; and device, 11,793), and 29,470 (18%) were observational. Data are updated every day; as of January 1, 2014, a total of 10,869 registered studies had posted results. With the search term “orthodontics,” the initial search yielded 72 studies. Each study was reviewed by the first 2 authors (V.A. and S.R.). Eight of the 72 studies were not included because they examined nonorthodontic outcomes, excluded subjects who were undergoing orthodontic treatment, or were studies conducted by faculty affiliated with orthodontic departments, but the actual study was not an orthodontic-related clinical outcome study.
Our final sample included 64 clinical studies. There were 52 interventional studies and 12 observational studies. The most frequently examined clinical topics by the different studies included accelerated tooth movement, white spot lesions, and plaque control or gingivitis. Eleven of the 64 studies were conducted in the United States, 9 in the United Kingdom, and the rest in 20 different countries ( Table I ). Seventeen studies were completed, 19 studies were still recruiting study subjects, 10 studies were no longer recruiting study subjects, 1 study was terminated, and 5 studies were enrolling participants by invitation.
|Characteristic||Interventional study (n = 52)||Observational study (n = 12)|
|Country of trial location|
|Belgium||1 (1.9%)||2 (16.7%)|
|Canada||4 (7.7%)||3 (25%)|
|Denmark||1 (1.9%)||1 (8.3%)|
|Hong Kong||2 (3.8%)||–|
|Israel||2 (3.8%)||2 (16.7%)|
|Saudi Arabia||1 (1.9%)||–|
|Taiwan||2 (3.8%)||2 (16.7%)|
|United Kingdom||9 (17.3%)||–|
|United States||10 (19.2%)||1 (8.3%)|
|Active, not recruiting||9 (17.3%)||1 (8.3%)|
|Completed||16 (30.8%)||1 (8.3%)|
|Enrollment by invitation||4 (7.7%)||1 (8.3%)|
|Not yet recruiting||6 (11.5%)||6 (50%)|
|Recruiting||16 (30.8%)||3 (25%)|
With the exception of 1 study that included only female subjects, all other studies included both male and female subjects ( Table II ). Nineteen studies included only children. Adult subjects were included in 45 studies. Most studies had sample sizes of 50 or fewer ( Table III ). Close to 60% of the interventional studies and 66.7% of the observation studies had sample sizes of 50 or fewer. About 21.2% of the interventional studies and 16.7% of the observational studies had sample sizes greater than 100. Information on the phase of the clinical trial was available for 25 interventional studies. Of these, 28% were phase IV, 12% were phase III, 8% were phase II or III, 16% were phase II, 8% were phase I or II, and 28% were phase I. Details on funding sources are provided in Table IV . Only 1 interventional study was funded by the National Institutes of Health. Most studies were funded by “other” or “industry” sources.
|Characteristic||Interventional study (n = 52)||Observational study (n = 12)|
|Both male and female||51 (98.1%)||12 (100%)|
|Only female||1 (1.9%)||–|
|Adult/senior||2 (3.8%)||1 (8.3%)|
|Child/adult||14 (26.9%)||8 (66.7%)|
|Child/adult/senior||7 (13.5%)||3 (25%)|
|Sample size||Interventional study (n = 52)||Observational study (n = 12)|
|1-25||13 (25%)||5 (41.7%)|
|26-50||18 (34.6%)||3 (25%)|
|51-75||3 (5.8%)||1 (8.3%)|
|76-100||7 (13.5%)||1 (8.3%)|
|>100||11 (21.2%)||2 (16.7%)|
|Funding source||Interventional study (n = 52)||Observational study (n = 12)|
|Industry and others||4 (7.7%)||–|
|Others/National Institutes of Health||1 (1.9%)||–|
Information on allocation protocol was available for 49 interventional studies. Of these, 87.7% (n = 43) were randomized, and 12.3% were nonrandomized (n = 6). The different interventional model assignments included factorial assignment (3.9% of interventional studies), parallel assignments (74.5%), crossover assignment (7.8%), and single-group assignment (13.7%) ( Table V ). Three observational studies were cohort studies, and 8 were case-control studies. One observational study did not provide information on the model. Data regarding the primary purpose of the interventional studies were available for 51 studies ( Table VI ). Most studies were treatment oriented (80.4%). The types of masking used by the interventional studies included open label (28.9%), single blind (44.2%), and double blind (26.9%). Outcome assessors were blinded in 6 studies.