Inclusion: The politics of difference in medical research (review)
- Inclusion: The politics of difference in medical research
In 1993, American law scholar Vanessa Merton published a wonderful article entitled "The exclusion of pregnant, pregnable, and once-pregnable people (a.k.a. women) from biomedical research." This article drew attention to the existence of U.S. Food and Drug Administration (FDA) rules published in 1977 excluding women of "childbearing potential" from most clinical drug trials, whether or not they were pregnant, or likely to become so. Although the rule was intended to apply mainly to Phase I and Phase II trials, it was quickly taken to apply to Phase III trials by most of the pharmaceutical industry. The rationale behind the exclusion was to prevent a fetus suffering drug-related injury as a result of a woman's participation in a trial. This exclusion could be seen as an attempt to protect the unborn from harm caused in clinical experimentation, based on a history of products initially licensed and then withdrawn from sale after their effects on fetal development became known, most notably the drug Thalidomide. Yet the logic of this approach was flawed. Of course, exclusion of women from such trials did not mean that women could not be prescribed such drugs after the drugs were licensed. The prescription would be at their (and perhaps their doctor's) risk, rather [End Page 174] than at the risk of the manufacturer or regulator. Prescribing would take place in the absence of reliable safety and efficacy data for women—including pregnant women. Hence, the effects of the regulation were to make clinical research with women less attractive, to deny women the power to make responsible choices about their own participation in research, to transfer risk from manufacturer to patient (and perhaps clinician), and to prioritize (at least in the research context) the interests of the non-existent or at any rate, not-yet-personal, over the interests of actual, autonomous, female persons (a.k.a. women). Moreover, the regulation reinforced the notion that women's bodies, biomedically speaking, were simply minor variants of men's, so that treatments could be safely and effectively developed on uncomplicated male bodies instead of the more complicated (if only marginally, relevantly so) female bodies.
By the time Merton's article was published, the FDA changed its guidance to allow the inclusion of women at all stages of clinical drug trials, provided they used effective contraception. Similarly, Congress passed the National Institutes of Health (NIH) Revitalization Act, imposing responsibilities on the NIH to take steps to ensure that women and "members of minority groups" be included in research projects funded by the NIH, and that subgroup analyses be performed to estimate the effectiveness and safety of the treatments tested in such trials within such subgroups. In other words, inclusion had to be in sufficient numbers to permit such subgroup analyses to be statistically meaningful.
Steven Epstein's marvelous new book describes and analyzes how this transition took place using the methods of the sociology of science and political sociology. He describes the shift that took place from the late 1980s to the late 1990s as a change of paradigm—from considering the basic unit of analysis in clinical research as the standard human (considered to be the white, male, middle-class American) to a diverse set of "standard humans" under what he calls the "inclusion-and-difference" paradigm. The point of the inclusion-and-difference paradigm is that clinical trials built around the standard human tended to be highly unrepresentative in terms of the demographic characteristics of their participants and thus, could be considered unfair in two ways. First, recruitment into trials would exclude people (especially, as we have seen, women) for the convenience of analysis and/or recruitment. To the extent that access to experimental treatment within clinical trials is important as a matter of treatment access, this exclusion could be considered distributively unjust. Second, it was assumed that notwithstanding the low participation rates of some groups—either because of formal exclusion or other informal kinds of underrecruitment—the [End Page 175...
