The Problem of Randomization within a Standard of Care Range: A Case Study

James Cummings

Background: Recent controversy in the press and medical literature surrounding a clinical trial in preterm infants raised several questions about the informed consent process in comparative effectiveness trials. An important consideration that was missing from these discussions was the implications of randomizing subjects within a standard of care, when that standard of care is not defined discretely, but by a range of physiologic measurement.
Summary: This paper discusses the risk/benefit implications when subjects are randomized within a standard of care that is defined by a range of physiologic measurement, rather than discrete therapeutic interventions. The recent controversy surrounding the informed consent process for a large, multicenter, clinical trial in preterm infants is used as the backdrop for this discussion, and a hypothetical study design built around a common clinical problem is used to further demonstrate the significant alteration in risk/benefit that might occur when subjects are randomized to narrower ranges of response within a larger standard of care range. While it may be possible to mitigate the negative effects of this type of randomization by alterations in study design, specifically closer monitoring and intervention, this potentially increased risk must clearly be addressed in the informed consent process.