Interview with Dr. Cara Tannenbaum

Show Notes for Podcast Eleven of seX & whY

Host: Jeannette Wolfe

Interview with Dr. Cara Tannenbaum, Professor in the Faculties of Medicine and Pharmacy at the Université de Montréal in Canada, and Scientific Director of the Institute of Gender and Health of the Canadian Institutes of Health Research

Definitions

Biological Sex- chromosomes, hormones, reproductive anatomy, usually binary

Gender- social and cultural construct- falls on a spectrum

• For a really nice summary of current use of definitions please see this excellent review.
• Excellent websites with tons of resources
  • Institute of Gender and Health- Canadian Institutes of Health Research
  • Video on how to conduct better science that considers the potential influence of sex and gender

Historically factors that limited the inclusion of women in clinical trials.

• Belief that outside of reproductive zones, males and females were the same
• Dogma that the female estrous cycle screwed up data and that male animals produced “cleaner” results
  • Two interesting facts: 1) Many female rodents’ entire estrous cycle is only 4 days!; and 2) We now know that male animals also have significant hormonal fluxes and that overall they are actually just as variable as females- see review
• Concern after the worldwide thalidomide nightmare* and the public backlash from the discovery of several unethical government sponsored clinical trials, that fetuses (along with prisoners and children) needed extra protection from the potential of unnecessary harm by participation in a research trial. This led to regulatory protection via the Common Rule. As any women of child-bearing age could theoretically become pregnant, they (and ultimately by cultural proxy all women) were essentially excluded from most human trials and early clinical phase drug trials from 1970’s to the mid 1990’s.
  • To read and an inspiring story as to why most of American was saved from the limb-shortening horrors of thalidomide, read here. (Essentially, FDA scientist Dr. Oldham Kelsey refused to sign off on its application, even amidst considerable pressure from the drug company, due to concern of inadequate evidence.)

Interesting sex and gender differences in car crashes

• Crash dummy 101
  • Historically crash dummy is Hybrid III which is 5’9’’ 170 pounds representing an average male
  • Hybrid III female model- 5’ 110 pounds
  • Other models- used by NHTSA
  • Why injury patterns may be different between men and women
    • Differences in baseline anthropometric measures (like height)
    • Biomechanical differences (women more prone to whiplash due to differences in neck muscular)
    • Mechanical design (Smaller adults sit closer to steering wheel and increase risk of lower extremity injury, and are more vulnerable to side impact since more of their head is in front of window)
  • NASS CDS data
    • Weight annual sample of US 5000 police reported tow away crashes
    • Collects data on
      • Occupant demographics (Age, sex, weight, BMI; Restraint use; Injuries obtained (via medical records and interviews) standardized into an abbreviated injury scale (AIS). It examines fatality and whole body and regional injuries, on a 1-6 scale of severity
      • Vehicle properties (Type, model year)
      • Crash conditions (Estimated speed, mechanism of impact)

What we know from NHTSA data and Insurance Institute for Highway Safety

• Overall, males represent about 70% of overall fatalities for crashes
  • Greatest gender differences is in 20-29 age group
  • Men more likely to have alcohol involved in accident
• On average men drive about 5000-6000 miles/yr more than women
  • Women more likely to work closer to home
    • Crashes more likely to be low speed and to occur in more congested areas
  • If a man and a woman are both in car
    • Males more likely to be driver
  • Summary of Bose study Vulnerability of female drivers involved in motor vehicle crashes: An analysis of US population at risk.
    • Question they asked- for a comparable crash do male and female drivers sustain similar rates of injuries.
      • Examined injury outcomes in men and women using 1998-2008 NASS CDS data set
      • For a comparable crash, women had 47% percent greater chance of being severely injured than men (had a higher risk of chest and spine injuries)
      • Of note the researchers controlled for weight and BMI

Other evidence that the clinical relevance of studying different sized and biomechanical models in crashes is important is shown by data obtained in 2011 after the NHTSA changed their safety star ratings to include testing of a female sized dummy in the front passenger seat. Many cars found their ratings go down, for example the 2011 Sienna minivan saw its ratings for passenger frontal crashes go from 5 star to 2 after it was shown that at 35mph that 20-40% of female dummies were killed or seriously injured compared to the industry average of 15%.

Underscoring the “literal” blind spots that can occur if you don’t consider factors associated with diversity in study design, a recent study from Georgia Tech suggested that some of the visual recognitions systems used that are critical for self-driving car safety may not adequately recognize dark skinned faces showing a 5% increased chance of error in recognition compared to that of fair skinned faces. Of note, there is a significant lack of gender and racial diversity in the self-driving car technology teams and in artificial intelligence/tech research overall.

Who makes up the team influences what gets studied, click here for a recent Lancet article and here for a Nature Human Behavior one both  showing that sex-related outcomes are far more likely to be reported in medical research consisting of diverse teams.

Take home points

• Including the variables of biological sex and gender in research results in better science and has led to the discovery of huge knowledge gaps that need to be closed if we want to optimize the care of all of our patients
• Our historical medical research model has been predominately based on the study of male animals. There are multiple reasons for this including a true belief that: outside our reproductive zones that men are women are exactly the same; using males animal produces cleaner data; and including women of child bearing age in clinical research trials exposes women to unnecessary risks without significant benefit. We now know that all these reasons are fundamentally flawed. Every cell has a sex and the differences between men and women outside their reproductive zones are often quite clinically important. Studying males and females side by side helps us to optimize the care of both sexes. In women it allows us to double check that therapies that were originally developed in men actually work in women and have the same benefit/side effects profiles. And for men, in instances when it is discovered that women have more favorably outcomes, it allows us to go back to the lab, figure out why there is a difference and then to use that knowledge to develop new therapies to help men.
• To move the scientific community and its deeply ingrained culture to a new model that incorporates the variables of sex and gender will require a comprehensive multi-targeted approach. Key considerations include- engagement, education, skill building around research methodology and analysis, mentoring and funding incentivization. Of note Institutional review boards, journal editors, grant reviewers and conferences directors have great power to jump start this transition by including an expectation of sex and/or gender inclusion in submission requirements.
• As we live in an ever increasingly complex world, now more than ever, it is essential that we pay attention to who is actually doing the research and developing new technologies. A diverse world requires diverse teams.

Next month we will look at the science pipeline from bench to bedside to identify opportunities to do better science.