Randomized controlled trials in adult traumatic brain injury: a review of compliance to CONSORT statement.

OBJECTIVE

To describe the extent to which adherence to Consolidated Standards of Reporting Trials (CONSORT) statement in randomized controlled trials (RCTs) in adult traumatic brain injury (TBI) has improved over time.

DATA SOURCES

MEDLINE, PsycINFO, and CINAHL databases were searched from inception to September 2013.

STUDY SELECTION

Primary report of RCTs in adult TBI. The quality of reporting on CONSORT checklist items was examined and compared over time. Study selection was conducted by 2 researchers independently. Any disagreements were solved by discussion.

DATA EXTRACTION

Two reviewers independently conducted data extraction based on a set of structured data extraction forms. Data regarding the publication years, size, locations, participation centers, intervention types, intervention groups, and CONSORT checklist items were extracted from the including trials.

DATA SYNTHESIS

Of 105 trials reviewed, 38.1%, 5.7%, and 32.4% investigated drugs, surgical procedures, and rehabilitations as the intervention of interest, respectively. Among reports published between the 2 periods 2002 and 2010 (n=51) and 2011 and September 2013 (n=16), the median sample sizes were 99 and 118; 39.2% and 37.5% of all reports detailed implementation of the randomization process; 60.8% and 43.8% provided information on the method of allocation concealment; 56.9% and 31.3% stated how blinding was achieved; 15.7% and 43.8% reported information regarding trial registration; and only 2.0% and 6.3% stated where the full trial protocol could be accessed, all respectively.

CONCLUSIONS

Reporting of several important methodological aspects of RCTs conducted in adult TBI populations improved over the years; however, the quality of reporting remains below an acceptable level. The small sample sizes suggest that many RCTs are likely underpowered. Further improvement is recommended in designing and reporting RCTs.