Abstract 3397: Do Patients Enrolled by Family Members in an Acute Stroke Trial Continue to Participate When They Regain Competency?

Abstract

Background: Stroke renders many patients aphasic or cognitively impaired, requiring their initial research consent for acute stroke trials be obtained from a proxy - a family member or other legally authorized representative (LAR). Federal law requires that proxy-enrolled patients who regain competency be approached for explicit self-consent to continue or discontinue trial participation. The frequency of competency recovery and the delayed decisions made have not been well studied. Methods: We analyzed patients enrolled in the NIH Field Administration of Stroke Therapy-Magnesium (FAST-MAG) testing prehospital initiation of magnesium sulfate within 2h of stroke onset. In FAST-MAG, a physician obtains consent by cellphone from: 1) the patient, 2) an on-scene LAR, if patient noncompetent. Per California law, LARs are: spouse, adult child, sibling, parent, registered domestic partner, and individual identified in an advanced health care directive /guardian. Results: Among the first 1008 explicitly consented patients, 397 (39.4%) were enrolled by proxy consent. Providers of proxy consent were: spouse (49.8%), daughter (24.9%), son (18.7%), sister (3.0%), brother (2.0%), mother (0.8%), and registered/domestic partner (0.3%). Among proxy consent patients, mean age was 74.0 (SD 12), 46.1% were female, and initial NIHSS was median 16 (IQR 2-14). Over the three month course of trial follow-up, 99/397 (24.9%) of these patients regained competency. The rate of consent to continue participation after competency recovery was 99/99 (100%, 95CI 96.4%-100%). Among the 298 patients who did not regain competency, 295 (99.0%) continued to final study visit under LAR consent and 3 had family consent withdrawn (1.0%). The all-source (self or LAR) rate of continued consent was 394/397 (99.2%). Among patients providing explicit self-reconsent, the median interval to reconsent was 1 day (IQR 0-2, range 0-118) and final 3 month MMSE scores were median 27.5 (IQR 25-29). Conclusions: In this prehospital stroke trial, 4 of every 10 patients were noncompetent and enrolled by proxy. Among these patients, 1 of every 4 later regained competency. All consented to continue. The high rate of delayed self-consent suggests LARs understand patient wishes well and proxy consent is an acceptable method for enrolling cognitively impaired patients in acute stroke trials.