Doctors Can Fight Unconscious Racial Bias

(Bloomberg Opinion) -- I am a first generation immigrant from West Africa. Growing up there, and in Europe and North America, I observed the many ways in which race influences identity, opportunity and social interactions. I am also a primary-care and public-health physician. Ever since I was a medical resident, I’ve noticed time and time again how my African-American patients seem more comfortable sharing sensitive information with me than with my white colleagues.

I’ve also noticed that these patients do not always get the same level of attention or care as white patients do. This apparent disconnect between my medical colleagues and many of our ethnic minority patients is what led me to devote my career to studying how race influences health care.

Racial disparities in health care are well documented. Hundreds of studies have shown that racial and ethnic minorities experience worse care and poorer health outcomes than whites. Researchers have found this to be true among pregnant women, children, youth, middle-aged and older adults, and people with heart disease, diabetes, kidney disease and cancer. They’ve found it in nursing homes and outpatient clinics, emergency rooms and hospitals, whether patients are getting flu shots or mammograms, bypass surgery or organ transplants.

My own early research in 2003 and 2004 demonstrated more specifically that doctors communicate differently with their African-American and white patients, dominating conversations much more with the former than with the latter. I also found that when African-American patients see doctors of their own race, the communication is more relaxed, engaging and enjoyable — and patients report better experiences.

Around the same time, I learned about a test called the Race Attitude Implicit Association Test, developed by Harvard researchers, which measures the extent to which people associate the faces of black and white people with good and bad words. People who associate good words with white faces and bad words with black ones more quickly than they associate good words with black faces and bad words with white ones are said to have an implicit preference for whites over blacks. The hundreds of thousands of people who have done the test online have revealed that most Americans — and most people from other countries, as well — implicitly favor whites over blacks.

To figure out whether implicit bias was contributing to racial disparities in health care, my colleagues and I decided to give the Race IAT to doctors and measure their communication with African-American and white patients. Our study showed that the medical world is much like the rest of society: About two-thirds of primary care doctors implicitly preferred whites over blacks, and two-thirds perceived whites as more cooperative, blacks as more mistrusting or reluctant to cooperate.

We also found that the greater doctors’ unconscious bias was, the more they dominated conversations with black patients, and the less they discussed social and personal issues with them. At the same time, black patients perceived these physicians as less respectful and less trustworthy. These doctors were notably not uncommunicative with their white patients. In fact, the white patients perceived them as more respectful than other doctors.

Other studies have substantiated our findings and suggested that these biases could influence clinical decisions and health-care outcomes. A study of medical residents found that the greater a doctor’s pro-white implicit bias, the less likely he or she was to treat black patients suffering from insufficient blood flow to the heart with medications to prevent heart damage — the same medications the doctor would readily provide for white patients in the same condition. Unconscious bias, in other words, causes doctors to make mistakes.

One good way to prevent mistakes is to use checklists. For instance, a five-step checklist used in intensive-care units in Michigan to prevent catheter-related infections saved some 1,500 lives in just three months. Another checklist designed to reduce gender disparity in blood-clot prevention lowered the incidence of preventable blood clots in one hospital to zero.

This methodology, so successful in certain corners of medicine, should be applied to helping doctors recognize and overcome unconscious bias. Though we don’t yet understand what strategies would work best to accomplish this change, a simple checklist I have devised could incorporate what we do know about effective communication between doctors and patients of different racial and ethnic groups. It also incorporates lessons from a tool called the Ladder of Inference, created by the organizational psychologists Chris Argyris and Peter Senge, which can help interrupt the unconscious processes that people often follow as they progress from observing things to making decisions. My checklist contains six items — easy to remember because the first letters spell the word “relate”:

  • Respect the humanity of the people in front of you, regardless of whether you like them or agree with what they are saying.
  • Empathize — imagine yourself in the patient’s shoes.
  • Listen more, and talk less.
  • Ask yourself what assumptions you may be making and whether they are based on facts about this particular person.
  • Talk with patients about their personal lives, not just their medical problems.
  • Engage patients in problem-solving and decision-making by asking what they think about their condition and the care plan.

Although doctors may already use many of these strategies, applying them deliberately could counteract bias and improve patients’ experiences, and ultimately their health. The researcher in me knows this checklist still needs to be tested in a carefully designed study. The healer in me greatly hopes that it can bring doctors closer to the goal of equitable health care for all.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Lisa Cooper is a Bloomberg Distinguished Professor at the School of Medicine, School of Nursing and the Bloomberg School of Public Health at Johns Hopkins University, and also the James F. Fries Professor of Medicine.

©2018 Bloomberg L.P.