Obtaining surrogate consent for a minimal-risk research study in the intensive care unit setting

Clin Trials. 2013 Feb;10(1):93-6. doi: 10.1177/1740774512464727. Epub 2012 Nov 20.


Background: Obtaining surrogate consent for clinical research studies conducted in the intensive care unit (ICU) setting is logistically challenging.

Purpose: To determine whether differences in proportions consenting to trial enrollment existed among patients eligible to consent directly versus those requiring surrogate decision makers in a minimal-risk study to evaluate the accuracy of continuous glucose monitoring in the ICU setting.

Methods: Low initial enrollment rates prompted a detailed tracking of the screening and consent process. We analyzed the subset of eligible patients identified during a single year to document whether they were approached about trial enrollment, whether they consented or declined, the reasons for declining, and the method of consent (self or surrogate). The proportion of participants who consented and the reasons for declining were compared for self-consenting and surrogate-consenting participants.

Results: Of the 3041 patients screened, one-third (n = 982) were eligible; 119 of the 982 were approached regarding enrollment. Absence of a surrogate accounted for the majority of eligible patients (726; 84%) not approached. The most common reasons for refusal in the self versus surrogate groups included feeling overwhelmed (13% vs 24%), fear of discomfort (22% vs 12%), and fear of risk (7% vs 4%). Of the 57 eligible patients capable of consenting directly, 11 (19%) enrolled versus 12 (19%) of the 62 who required surrogate consent. When recruitment hours were expanded to include evening time, more eligible patients or their surrogates could be approached than during the day-shift hours alone. Consent was obtained for a larger proportion of potential participants with a history of diabetes (40%) than for those without a history of diabetes (14%).

Limitations: The findings are from a subset of the entire study sample; data were available only for participants who could be approached, who may have differed from those who could not be approached.

Conclusions: Surrogate and self-consent rates were similar. Surrogate unavailability was a major barrier to enrollment; overlap of staffing with usual visiting hours should be considered when planning trials in the ICU.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Biomedical Research / methods*
  • Diabetes Mellitus / drug therapy*
  • Female
  • Glucose / administration & dosage*
  • Humans
  • Intensive Care Units / statistics & numerical data*
  • Legal Guardians*
  • Male
  • Massachusetts
  • Middle Aged
  • Third-Party Consent / statistics & numerical data*
  • Young Adult


  • Glucose