Size and quality of randomised controlled trials in head injury: review of published studies

摘要: Abstract Objective: To assess whether trials in head injury are large enough to avoid moderate random errors and designed biases. Design: All randomised controlled on the treatment rehabilitation of patients with published before December 1998 were surveyed. Trials identified from electronic databases, by hand searching journals conference proceedings, contacting researchers. Data extracted number participants, quality concealment allocation, use blinding, loss follow up, types interventions, outcome measures. Results: 279 reports identified, containing information 208 separate trials. The average participants per trial was 82, no evidence increasing size over time. total 203 which reported 16 613. No detect reliably a 5% absolute reduction risk death or disability, only 4% an 10%. Concealment allocation adequate 22 inadequate unclear 25 47 (23%) it reported. Of 126 assessing 111 followed up 19%. measuring 26 (21%) that assessors blinded. Conclusions: Randomised too small poorly refute but clinically important benefits hazards treatment. Limited funding for research unfamiliarity issues consent may have been obstacles.