Against One-Size-Fits-All Research Ethics

Many feel the Common Rule treats an unwieldy range of activities identically under the monolithic label "human subjects research." Past objections centering on the conflation of biomedical and behavioral research have gained new currency with the increase in biobanking and Internet-based research. A more nuanced approach to research is overdue. Regulation will no doubt remain a major component of any new approach. But in some research contexts, investigators and subjects should be permitted to reach voluntary, informed agreements about certain aspects of their relationship.

Consider the National Institutes of Health's new "Guidelines for Human Stem Cell Research."¹ The guidelines owe their existence to the NIH's recognition that research diversity requires regulatory diversity. Under the Common Rule, the sources of existing, coded biological specimens are not considered research subjects when certain conditions are met. Thus, neither their consent to, nor even their knowledge of, research is required.² When the source's only interest is in informational privacy, this rule is defensible. But when the source—or her relatives, or other genetic groups—objects to the kind of research being conducted, deidentification is irrelevant. While leaving the deidentification-is-enough rule otherwise intact, the NIH singled out human embryonic stem cell (hESC) research on deidentified embryos left over from in vitro fertilization, sensibly requiring providers' voluntary, informed consent, as defined by the guidelines' criteria.

But if the guidelines as a whole show the promise of disaggregating research, one of the first decisions made in their name shows the peril of failing to do so. While newly derived lines must meet the guidelines' rigorous informed consent criteria, the provenance of existing lines is reviewed by a specially formed working group of the Advisory Committee to the Director, "taking into account the principles articulated" in both the guidelines and the Common Rule.³ The working group found that lines derived by the Reproductive Genetics Institute fell short of these principles because donors were required to waive any legal claims arising from the research.⁴ The guidelines' consent criteria, designed specifically for this context, say nothing about exculpatory language; however, the Common Rule—applicable here even to nonfederally funded entities like RGI as a source of ethical guidance—prohibits it. Although the NIH said it would approve the lines if donors consented without the waiver, deidentification makes contacting them again difficult.

Why is the exculpatory clause incompatible with the principles of ethical research? According to NIH Director Francis Collins, the "use of exculpatory language . . . was inconsistent with the basic ethical principle of voluntary consent."⁵ But if voluntariness is a problem at all, it is so only with respect to the waiver term. The most plausible argument to be made from involuntariness is this: the waiver term was unfavorable, but donating their embryos was so important, and the opportunity to do so without agreeing to a waiver so lacking, that donors' acceptance of the waiver was coerced. If true, the solution is to sever the offending clause, not to invalidate the parties' intent. Having agreed to donate despite the offending term, they needn't be contacted to ask if they're willing to donate without it. Instead, in exchange for approving the lines, the NIH should require RGI to waive its right to rely on the waiver in any lawsuit brought by donors. (Conversely, if the waiver somehow did infect the agreement to donate, it's unclear how contacting donors again for their consent would constitute a cure. A party who didn't authorize a contract made in its name can later ratify it, making her consent retroactive to the date of the contract. But here, consent today isn't equivalent to initial consent: embryo destruction is a fait accompli, and donors may "choose" only between permitting the lines to be used for potentially beneficial research or insisting that they be wasted. Contact under these circumstances seems more like begging for forgiveness than asking for meaningful permission.)

Now consider a more provocative possibility: that in this particular case, the waiver itself is ethically acceptable. Like most contracts we sign every day, RGI's was a take-it-or-leave...