The striking male dominance of surgical specialities is difficult to miss, and this appears to have at least partially motivated this study’s research question. While I acknowledge the study’s methodologic and statistical rigor, I respectfully submit that there are some serious philosophical concerns with its general line of inquiry that deserve consideration.
First, the variable in question is an effectively non-modifiable one. Rather than examining specific attributes or behaviours (e.g., communication skills, adherence to guidelines) which any physician -- regardless of sex -- could manifest, the authors studied sex and subsequently speculated on what behaviours may have driven the small difference observed. If one’s goal is quality improvement, then a focus on adjustable attributes or behaviours that all physicians could theoretically be taught to develop would be more fruitful.
Second, as the authors state, “These results do not support the preferential selection of a surgeon of either sex in clinical practice”. Indeed, if the within-group differences in outcome exceed the between-group differences (and this is almost certainly the case), one truly could not conclude that any individual surgeon of a given sex is better or worse than his or her counterpart based on sex alone. Unfortunately, the subtlety of such an argument is compromised in the era of “sound bites” and tweets. The headline of a prominent article about this work on time.com, for example, is “Researchers Find Women Make Better Surgeons Than Men”. 1 This effect is exacerbated by the phenomenon of “stereotype threat,” in that the group expected to have worse outcomes comes to underperform over time, essentially participating in a self-fulfilling prophecy.2 This would clearly be a undesirable outcome.
Third, given the provocative nature of the study’s findings, it is nearly impossible not to wonder about a political subtext. I question, for example, whether this paper would have been published (or even submitted for publication) had the opposite result been found. Similarly, would the authors have felt as comfortable encouraging female surgeons to learn from the behaviour patterns of their male counterparts? I suspect not.
I challenge our profession to avoid the “slippery slope” of physician identity-based research questions. Should we compare outcomes of Asian providers to those of Caucasian or Hispanic ones, for example? Physicians, like all populations, can be fragmented into an infinite number of subcategories; deciding which, if any, of these are worthy of study is not a trivial task, particularly when results are either not actionable or almost certain to be misinterpreted and misapplied.
Diana Toubassi, MD, CCFP
1. http://time.com/4975232/women-surgeon-surgery/ Accessed October 12, 2017.
2. Spencer SJ, Logel C, Davies PG. Stereotype threat. Annu Rev Psychol 2016;67:415-37. doi: 10.1146/annurev-psych-073115-103235. Epub 2015 Sep 10.
Competing interests: No competing interests