I saw a study the other day outline in Contemporary Obgyn magazine. The study was originally published in JAMA Network Open. This study attempted to evaluate the possibility of gender and racial bias in patient selection of their physicians. In other words, does it matter if the physician is male or female? Black or white? Here’s how the study went. They took about 3600 online US respondents from a very geographically diverse population. Two different crowdsourcing platforms MTurk and Lucid were used to recruit participants. Each participant was shown a clinical scenario of a gastrointestinal issue being evaluated by an emergency room physician. The participant was supposed to play the role of the patient. They were shown pictures of the different ER physicians with different diagnoses: one with a diagnosis of gastroenteritis and a conservative treatment plan and one with a diagnosis of appendicitis and a more aggressive plan. The researchers did what is called a random assignment of the participants to images of physicians with different genders and races. 823 were assigned to black female physicians, 791 to black males, 821 to white females, and 835 to white males. They used a total of 10 different faces for each of these groups from the Chicago Face Database. This means that they were trying really hard to determine whether a participant’s reaction was truly due to a particular skin color or gender, rather than one specific face. None of the participants knew about the purpose of the study. In research terms, that means that they were “blinded.” Each participant then had to rate each simulated physician on confidence in him or her, satisfaction with care, likelihood of recommending the physician, trust in the diagnosis, and the likelihood that they would request additional tests.
I found the initial report of the results as reassuring. They did not find any significant difference in participant satisfaction and confidence based on race or gender. Ok that sounds good at first, but it is not time to celebrate just yet. They went on to say that about 40% of the participants endorsed some type of white superiority. It does not explain how they knew that. So then, they essentially threw the accuracy of the results out the window and said that it was possible that the participants actually knew the purpose of the study and hid their prejudices in their answers to look more socially acceptable. Mind you, that this is by no means the first study evaluating gender and racial bias in patient physician selection. Also keep in mind that a lot of these studies are failing to show any systemic bias in regards to gender or race, but we keep looking anyway….and then not believing the results. I don’t get it. Would we prefer that there is bias? Are we unhappy that we can’t prove it? Are we looking for yet another race or gender inequality issue to be upset about? There are plenty of real issues out there with regard to race and gender that desperately need to be addressed. Maybe we should leave this one alone if we cannot truly prove that it is there in the first place. Have a fantastic day everyone!
Dr. Katz
