Waiver of consent in studies of acute brain injury.

A multicenter trial of hypothermia in patients with acute brain injury, designed to accrue 140 patients per year and randomizing in less than 6 h from injury, enrolled 392 patients. The design was to achieve 33 degrees C within 8 h after injury. For the first 9 months of the trial, the only consent mechanism permitted by federal regulations was prospective, informed consent. In the subsequent 33 months, after a change in federal regulations, waiver of consent could be used when family could not be located. Waiver of consent was used in 62% of patients enrolled. In the first 9 months of the trial, accrual was 65 patients. In the subsequent 3 years, an average yearly accrual was 127 patients. In the first 9 months, time from injury to randomization was 4.5 +/- 1.2 h; time to achievement of target temperature was 11.7 +/- 2.6 h. In years when waiver of consent was permitted, randomization time was 4.1 +/- 1.1 h, and time to target temperature was 7.9 +/- 2.7 h. For all years of the study, waiver of consent was used for 53% of minorities, 47% of unskilled workers, 33% of nonminorities, and 29% of skilled or professional workers. Minorities were underrepresented by 30% in the first 9 months of the study. We conclude that it is impracticable and unjust to perform studies of acute brain injury without use of waiver of consent when the treatment window is less than 6 h.