Skip to content

Biases

Biases of Clinicians and Institutions

Share This Section

Throughout the consultation process, health leaders highlighted how institutional values, culture, and behavioral norms perpetuate structural racism and interact with the individual biases of clinicians to determine quality of care for patients of color. 

Yale University scholar Dr. Miraj U. Desai shared his findings that “clinics have biases and norms about what an ‘ideal’ patient is, largely based on how well they help the organization to run efficiently. In mental health, the ‘ideal’ patients are verbal, admit a problem, and accept services. If a patient does not exhibit these behaviors considered as normal, modeled in part on Eurocentricity, the system may winnow them out. This creates barriers for Black, Indigenous and people of color, and others who do not fit the predominant racial and cultural norms.”1

Organizational biases intersect with the racial biases of care providers, who have a great deal of autonomy and power to make life-changing and life-ending choices for BIPOC patients.

One person reflected on “seeing 30 of my colleagues perish on the front lines from COVID—and the majority of them were Black and Brown. As hospitals were pushing for autonomy so that they could make determinations around who and when a patient gets a ventilator, I knew that both implicit bias and explicit bias was going to show up. I knew who was gonna end up getting the ventilator and who was not. I realized that just as studies are now indicating that Black babies live longer when they have Black doctors, Black people would probably live longer with COVID if they had Black doctors too.”

Racial bias also has become embedded in medical training and diagnostic tools, including guidelines that predict who is most likely to die in the hospital from acute heart failure, whose kidney transplant is higher risk, and whose lungs have more capacity. 

“Originally the spirometer, which is designed to measure lung capacity, was [used to] justify that the hard work that slaves were doing in the field was a good thing because they have lower lung capacity and this [hard work] helps increase their capacity,” one leader said. “And even today, the standard measurements are different for whites and Blacks.”

Further, the health system has a data and research bias that effectively masks structural racism. That bias shows up in what research gets done, what research informs institutional and public policy, and the continued dominance of white men in medical study samples. 

Many leaders also noted how bias and discrimination around race and ethnicity in medical research, diagnostics, treatment, and public health also reflected systemic bias against women, the LGBT community, and people with disabilities. And for people with multiple marginalized identities, the harm of discrimination and bias is compounded.

Most health systems don’t track or effectively analyze data associating race, ethnicity, language, and other marginalized identities of patients with their outcomes, so they don’t see the real impacts of their policies and practices in BIPOC people. 

Black woman consulting with a male doctor

As one person shared, “To collect data on race, ethnicity, age, language proficiency, you need a fairly sophisticated person to do that, but right now you have the lowest paid intake people collecting that data. So, in general, most organizations are not working with perfect data. We need regulators to require the collection of data and to reward organizations who are reducing gaps and penalize organizations who are not. Until we have that kind of system in place, it’s not going to happen.” 

Many said that good data provides an opportunity to define and work with clear quality metrics to drive improvements and close gaps in race-associated outcomes.

“The architecture of data systems reflects inherent values and biases and discrimination even in terms of who is invisible in our data systems,” a health leader said. “We need to reimagine a different table set for who is designing our data systems [where patient/communities most impacted are included], what data would be in there, but also, who decides how those data are used, and how we could use data to drive incentives and alignment to get us to the goal of equity in our health outcomes.”

Added another leader: “There is a need for an organization to turn a lens on itself, for them to explore individual bias and prejudices, especially leaders. There’s a need for greater self-reflection and self-honesty.” 

These are the views of health leaders who participated in an 8-month project to analyze structural racism across the U.S. health system and provide recommendations for the collective leadership required to dismantle it.

  • 1

    “Implicit Organizational Bias: Mental Health Treatment Culture and Norms as Barriers to Engaging With Diversity,” Desai, Paranamana, Restrepo-Toro, O’Connell, Davidson (2021).