Patients' preferences for enrolment into critical-care trials

Interdepartmental Division of Critical Care Medicine, University of Toronto, Toronto, ON, Canada.
European Journal of Intensive Care Medicine (Impact Factor: 5.17). 07/2009; 35(10):1703-12. DOI: 10.1007/s00134-009-1552-y
Source: PubMed

ABSTRACT Most critically ill patients are incapable of providing informed consent for research.
We sought to determine patients' preferences for different consent frameworks for enrolling incapable patients into critical-care trials.
Prospective observational and structured interview study.
Five university-affiliated hospitals in Ontario.
Two-hundred and forty consecutive capable and consenting survivors of critical illness.
Participants considered four frameworks for enrolling incapable patients into clinical trials using a baseline scenario and three permutations for: risk (very low vs. high), treatment type (new vs. currently available), and availability of substitute decision-maker (yes vs. no).
For each scenario, patients chose their preferred framework and rated the acceptability of each framework using a seven-point Likert scale. Most (180/240; 76%) patients selected "consent by substitute prior to enrolment" as their preferred framework; this also received the highest baseline acceptability ratings ("acceptable" or "highly acceptable" 207/240; 87%). Modifying risk or treatment type did not substantially change these ratings. A minority of patients rated delayed consent as unacceptable or highly unacceptable in both the baseline scenario (48/240, 20% delayed to substitute; 57/240, 24% delayed to patient) and when a substitute was unavailable (34/240; 15%).
Most survivors of critical illness found the usual practice of obtaining informed consent from a substitute decision-maker prior to enrolment in a clinical trial to be acceptable. Nearly half of patients considered foregoing informed consent to be unacceptable, whereas a minority considered enrolment followed by delayed consent to be unacceptable even when a substitute was unavailable. These approaches should, therefore, only be considered when deviating from the usual practice of obtaining consent from a substitute decision-maker is truly justified, such as where treatments being tested need to be delivered as soon as possible in order to be effective.