On Becoming a Domestic and Laboratory Goddess

 Dr. Isis has returned to MRU and is cuddled up, drinking a cup of Orange Chocolate Green Tea, and reading some science. 

Figure 1: Dr. Isis adores her Eldreth Pottery mugs for drinking tea and reading science. She also owns a pie plate and soup crock and has found no other ceramic cookware that heats as evenly or retains heat as well. Check out his collection and order here. 

Now, Dr. Isis quite frequently blogs about being a girl, so a letter in this week's Science entitled Flaunting the Feminine Side of Research Studies certainly caught the eye of the domestic and laboratory goddess.  In her letter, Phyllis Greenberger of the Society for Women's Health Research notes that men and women are fundamentally, physiologically different. Dr. Isis could not agree more and thinks it important to note that the uniqueness of our feminine physiology is not in opposition to struggles for gender equality.  Ms. Greenberger notes that women are currently underrepresented in clinical study cohorts and that data are rarely analyzed for sex-based effects. Indeed, she notes the percent of women participating in studies investigating cardiovascular disease is 27%.  All of this in spite of the NIH Revitalization Act of 1993  and the ORWH's 2008 Research Priorities for Women's Health Statement.

Figure 2: The current state of clinical research.

Yeah.

This is a particular issue that Dr. Isis, as a vascular physiologist and a woman, is painfully aware of and, yet, the difficulties associated with including women in clinical research can be more pragmatic than simple gender discrimination.  For example, in the animal model of the disease Dr. Isis currently studies, she limits her animal population to young, male critters.  It is her hope to establish her findings in male creatures before considering the effect of femaleness on the mechanism of this disease.  Does Dr. Isis limit her research model because she secretly hates women (especially those hotter than her) and does not want to cure them?  No, but she appreciates that the effects of estrogen on vascular function are hugely complex and temporally based.  This stuff is so complex that people spend their careers studying it and there are all sorts of fancy books devoted to the topic.  In the human studies Dr. Isis has been/is involved with,  we have to be careful to study women in the context of their menstrual cycle.  In a previous study, we studied them only during menstruation when all that nuisance estrogen wasn't floating around.  In a current study I am collaterally involved with, the principal investigators are studying their population during menstruation and again during ovulation.  This means that each female participant must be studied twice while each male participant need only be studied once.  Thus, in this study, women are twice as expensive as men.  I could not agree more with Ms. Greenberger that the inclusion of women in clinical research is vital to the understanding of our unique physiology and I applaud her for reminding us of this public health disparity.  However, except in phase III or IV clinical trials where sample sizes may be sufficiently huge to allow for the inclusion of secondary endpoints and the splitting of populations for secondary analyses without devastating one's statistical power, splitting the populations of smaller, mechanistic studies can result in meaningless findings.  Let's ignore the fact that, as reported in the New England Journal of Medicine, may studies don't have sufficient statistical power to detect important differences in the original population to begin with.

I suppose there are three solutions to the issue of study design.  First, one can take the approach Dr. Isis has taken: I study my model in male animals first and determine if something is important. Then, I design a second study with sex effects as the primary endpoint and add intact and ovariectomized female animals to determine the importance of being XX as opposed to XY.  A major potential problem with this approach is that, just because something ends up not being important in male animals doesn't mean it's not important in female animals. I potentially may miss something important.  I could reverse the order of the sexes studied, but the same disadvantages are still applicable.  The second option is to study both sexes simultaneously in a study with a sample size large enough to allow for division of the population.  The disadvantage to this approach is, although you end up with more data to answer the fundamental question, you also potentially end up using many more subjects before you find out that your stuff doesn't work.  Finally (and probably Dr. Isis's least favorite option) we can throw sex effects to the wind, pool the data, and study single species cohorts composed of 50% male and 50% female creatures/humans.

But, I think there are more immediate fundamental solutions to the problem.   First, researchers have to be willing to tackle the investigation of sex effects in disease, even though it means larger samples sizes and (frankly) more work.  Second, as we interpret published findings, especially as they apply to the treatment of human patients, we have to remember to ask who comprised the cohort and ask if it is appropriate to apply the findings to female patients.  Finally, we have to continue to support groups like the Society for Women's Health Research that  remind us of  a major gap in medical knowledge, appreciate the effect this gap has on public health, and aims to close it.