Brief Consent Methods Enable Rapid Enrollment in Acute Stroke Trial: Results From the TICH-2 Randomized Controlled Trial

Abstract

Background and Purpose: Seeking consent rapidly in acute stroke trials is crucial as interventions are time sensitive. We explored the association between consent pathways and time to enrollment in the TICH-2 (Tranexamic Acid in Intracerebral Haemorrhage-2) randomized controlled trial. Methods: Consent was provided by patients or by a relative or an independent doctor in incapacitated patients, using a 1-stage (full written consent) or 2-stage (initial brief consent followed by full written consent post-randomization) approach. The computed tomography-to-randomization time according to consent pathways was compared using the Kruskal-Wallis test. Multivariable logistic regression was performed to identify variables associated with onset-to-randomization time of ≤3 hours. Results: Of 2325 patients, 817 (35%) gave self-consent using 1-stage (557; 68%) or 2-stage consent (260; 32%). For 1507 (65%), consent was provided by a relative (1 stage, 996 [66%]; 2 stage, 323 [21%]) or a doctor (all 2-stage, 188 [12%]). One patient did not record prerandomization consent, with written consent obtained subsequently. The median (interquartile range) computed tomography-to-randomization time was 55 (38–93) minutes for doctor consent, 55 (37–95) minutes for 2-stage patient, 69 (43–110) minutes for 2-stage relative, 75 (48–124) minutes for 1-stage patient, and 90 (56–155) minutes for 1-stage relative consents ( P <0.001). Two-stage consent was associated with onset-to-randomization time of ≤3 hours compared with 1-stage consent (adjusted odds ratio, 1.9 [95% CI, 1.5–2.4]). Doctor consent increased the odds (adjusted odds ratio, 2.3 [1.5–3.5]) while relative consent reduced the odds of randomization ≤3 hours (adjusted odds ratio, 0.10 [0.03–0.34]) compared with patient consent. Only 2 of 771 patients (0.3%) in the 2-stage pathways withdrew consent when full consent was sought later. Two-stage consent process did not result in higher withdrawal rates or loss to follow-up. Conclusions: The use of initial brief consent was associated with shorter times to enrollment, while maintaining good participant retention. Seeking written consent from relatives was associated with significant delays. Registration: URL: https://www.isrctn.com ; Unique identifier: ISRCTN93732214.