Informed Consent that Confounds Data Analysis

Your research project will study the impact of a new way to care for patients with Alzheimer’s disease.  Guidelines for research with human subjects require obtaining informed consent from each subject.  For most Alzheimer’s patients, a family member or other authorized person must provide this consent.  You suspect that Alzheimer’s patients who have family or a health care proxy will respond differently from those who lack such support.  Do you:

1. Request an exemption from the requirement for informed consent.
2. Increase the sample size sufficiently that data on subjects who provided informed consent can be analyzed separately from any available data on subjects who did not provide informed consent.
3. Remove individual identifiers from data on subjects who did not provide informed consent and analyze this data anonymously.
4. Exclude subjects who did not provide informed consent from the study and ignore the confounding effect of having informed consent.

J1G says:  Work the dilemma into the research design and analyses. Consider covariance, crossover, counterbalancing, etc.

R3H says:  This is not an uncommon problem.  There is a mechanism in the institutional review process for dealing with exactly this problem.  An approved IRB can grant approval for a research project under particular circumstances (minimal risk, no individually identifyable data) without the requirement for prior informed consent if the IRB concurs that prior consent would taint the data.  In some cases, consent may be required after the data has been acquired but before it can be used as part of the study database.  All of this is included in the purview of an appropriate IRB.