The New Science of Unconscious Bias:
My name is Jin My name is Jean My name is Gene My name is Jeanne. Do you know me? The New Science of Unconscious Bias: Workforce and Patient Care Implications Presented by: David B. Hunt Jerry Benston (612) Under Accreditation Council for Continuing Medical Education guidelines disclosure must be made regarding relevant financial relationships with commercial interests within the last 12 months. David Hunt Commercial Interest Nature of Relevant Financial Relationship What was received? For what role? Critical Measures Payment Equity Owner Under Accreditation Council for Continuing Medical Education guidelines disclosure must be made regarding relevant financial relationships with commercial interests within the last 12 months. Jerry Benston Commercial Interest Nature of Relevant Financial Relationship What was received? For what role? Critical Measures Consulting fees Consultant Disclosure Statement David Hunt discloses that he is the sole equity owner of the company Critical Measures and that Critical Measures is engaged in the business of diversity-related, healthcare consulting. Critical Measures does not offer any products or services that relate to the clinical care of patients and will not make any clinical recommendations during the course of this presentation. Critical Measures does, however, offer various, diversity-relatedorganizational assessment tools and CME-accredited e-learning programs for doctors and nurses. Jerry Benston is an independent consultant with a long-term, consulting and training affiliation with Critical Measures. Learning Objectives Upon completion of this session, participants will improve their competence and performance by being able to: Explain how human biology and sociology contribute to unconscious or implicit bias. Identify their own unconscious biases by taking the Implicit Association Test. Recognize the implications of implicit bias research for the OU workforce and racial and ethnic disparities in patient care. Use emerging best practices to overcome or minimize the effects of unconscious bias both personally and organizationally. Develop an organizational action agenda to address the effects of implicit bias in the workforce and patient care. Learning Objectives (contd)
Upon completion of this session, participants will improve their competence and performance by being able to: Identify ten core cross-cultural differences that can contribute to communication problems and cross-cultural conflict. Utilize the Intercultural Conflict Style Model as a framework for managing and resolving cross-cultural conflict. Identify the leadership competencies that managers and providers will need to manage a culturally diverse workforce and care for culturally and linguistically diverse patients. Agenda The Science of Unconscious Bias
Measuring Implicit Bias The Implicit Association Test Workplace and Patient Care Implications Application: Racial and Ethnic Disparities, Language Access, Caring for Immigrants and Refugees Dealing With Cross-Cultural Differences BARNGA Dealing With Cross-Cultural Conflict ICS Model Becoming Conscious of Unconscious Biases Next Steps For Individuals and Organizations Understanding the New Science of Bias
* 07/16/96 Understanding theNew Science of Bias * ## Know But Dont Know I Know Think I Know
We may want to discuss a way to incorporate this with the conversation about the important considerations listed below. Awareness: New Research re: Bias
* 07/16/96 Awareness: New Research re: Bias In the past, bias was regarded as aberrant, conscious and intentional. Today, we understand that bias is normative, unconscious and largely unintentional. Social Cognition Theory establishes that mental categories and personal experiences become hard-wired into cognitive functioning. 4. As a result, human biases can be seen as evolutionarily adaptive behaviors. * ## Unconscious Bias: How Does It Work?
The problem? Too much information to process. Scientists estimate that we are exposed to as many as 11 million pieces of information at any one time, but our brains can only functionally deal with about 40. The solution? Mental short-cuts. The brain seeks to conserve energy. Decision-making, ambiguity, novelty and problem solving all take heavy cognitive reserve. Weve evolved to have mental short cuts that save time and usually yield reliable results. Our Brains at Work Seeks the simplest path to conclusions
The brain asa prediction-machine Seeks the simplest path to conclusions Wired for threat identification Perceptions of Groupness Distort Perception and Behavior
Experiments by Tajfel and others showed that, as soon as people are divided into groups even on trivial or random bases strong biases resulted. Subjects perceived members of their group as more similar to them and members of other groups as more different. Subjects saw in-group members highly differentiated individuals and out-group members as largely homogenous. Subjects were better able to recall undesirable behavior of outgroup members than similar behavior of ingroup members. Ingroup members failures were attributed to situational factors while outgroup failures were attributed to innate characteristics. Subjects permitted to allocate monetary rewards maximized rewards to their own group and minimized rewards to outgroups. Video -The Eye of the Storm
* 07/16/96 Video -The Eye of the Storm Reactions to film/video? Impact of the collars on students socially & academically. Who wears collars at OU Medicine? Take-aways for you: -Professionally -Personally * ## The Implicit Association Test
* 07/16/96 The Implicit Association Test * ## YouTube Video: Do the Test
* 07/16/96 YouTube Video: Do the Test * ## SLB CFLTK SPRND HLMG John Ridley Stroop, 1935
Copyright Cook Ross Inc. - Unconscious Bias Learning Lab RED GREEN YELLOW BLUE BROWN
John Ridley Stroop, 1935 Copyright Cook Ross Inc. - Unconscious Bias Learning Lab What Activates Our Biases?
* 07/16/96 What Activates Our Biases? Our biases are most likely to be activated by four key conditions. They are: stress time constraints multi-tasking need for closure * ## https://implicit.harvard.edu YouTube Video: How Biased Are You?
* 07/16/96 YouTube Video: How Biased Are You? * ## * 07/16/96 Key IAT Findings - Age Age: Around ninety percent of Americans mentally associate negative concepts with the social group "elderly"; only about ten percent show the opposite effect associating elderly with positive concepts. Older people do not, show an automatic preference for their own group. Remarkably, the preference for young is just as strong in those in the over-60 age group as it is among 20-year-olds. * ## Key IAT Findings - Gender
* 07/16/96 Key IAT Findings - Gender Gender: Seventy-five percent of men and women do not associate female with career as easily as they associate female with family. (Women show an implicit attitudinal preference for females over males, but they nonetheless show an implicit stereotype linking females closer to family than career.) * ## * 07/16/96 Key IAT Findings - Race Race: Whiteparticipants consistently show a preference for White over Black on the IAT a substantial majority of White IAT respondents (75% to 80%) show an automatic preference for White over Black. Data collected from this website consistently reveal approximately even numbers of Black respondents showing a pro-White bias as show a pro-Black bias. Other key race findings: younger people are just as likely to display an implicit race bias as older adults, women are as likely to display an implicit race bias as men and educational attainment appears to make no difference with respect to implicit race bias. * ## Summary: Key IAT Trends
* 07/16/96 Summary: Key IAT Trends Implicit biases are pervasive. People are often unaware of their implicit biases. Implicit biases predict behavior. People differ in levels of implicit bias. Educational attainment makes no difference with respect to implicit biases. * ## Workplace Implications
* 07/16/96 Workplace Implications * ## Biases Impact Decision-Making
* 07/16/96 Biases Impact Decision-Making Unconscious bias can infect management decisions throughout the employment life cycle: a.Interviewing. Recruitment, hiring & retention. b.Expectations of and interactions with employees. (Micro-inequities) c.Employee evaluations. (Set Up to Fail Syndrome Harvard Business Review) d.Decisions about promotions, training and other job benefits. e.Termination and discharge decisions. * ## The Big Five Orchestras
Chicago and Boston None of the Big Five employed more than 12% women until the 1980s Blind auditions Improved the chances that a woman would ultimately be hired Female musicians in the Big Five increased five-fold from 1970 to 2000 Orchestrating Impartiality: the Impact of Blind Auditions on Female Musicians, 94 Am. Econ. Rev. 715 (2000). Susan Boyle Britains Got Talent
Great talent often doesnt look and act like you Can you spot great talent no matter how it is packaged? Selection/Signing of Professional Athletes Are Emily & Greg More Employable than Lakisha & Jamal?
Study of actual racial hiring bias in Chicago and Boston Resumes sent to actual want ads 4 resumesper position 2 high quality and 2 low quality African American sounding names assigned to one high quality and one low quality Primary measurement was the callback rate Results: people with "white-sounding" names are 50 percent more likely to get a response to their resume than are those with "black-sounding" names. Marianne Bertrand and Sendhil Mullainathan, Are Emily and Greg More Employable Than Lakisha and Jamal?Field Experiment on Labor Market Discrimination, 94 Am. Econ. Rev. 991 (2004). Racial Disparities in Medical Education
In 2012, there were just 517 black men among the more than 20,000 graduating students among the more than 20,000 graduating students at U.S. medical students. Only 2.9% of all faculty members at U.S. medical schools are black. Black faculty members are less likely than their white counterparts to be promoted, to hold senior faculty or administrative positions and to receive research awards from the National Institutes of Health. 31% of the 84,195 white faculty members at U.S. medical schools were full professors in 2011, compared with just 11% of the 3,952 black faculty members. Source: Bias, Black Lives, And Academic Medicine, New England Journal of Medicine, March 19, 2015 Diversity and Productivity
Effective diversity programs are associated with higher productivity (+18%). (National Urban League, 2004) Gallup found that 27.7 million U.S. workers, or 18%, are actively disengaged. Another 52% of workers were not engaged, while only 30% of workers were actively engaged. Result: 70% of workers are not fully engaged. Actively disengaged" employees -- those fundamentally disconnected from their jobs -- cost the U.S. economy between $450 billion and $550 billion a year. (Gallup 2012) What causes workers to disengage at work? One notable cause is DRIs Diversity Related Incidents of Disrespect. Workplace Incivility DRIs
Studies have found that over 71 percent of the workforce has experienced some form of workplace incivility in the last five years. Incivility is evidenced by disrespectful behavior. Source: Don Zander, Brookings Institution, 2002 Of the reported incidents of workplace-related DRIs: 32% were related to gender; 28% were related to race; 20% were related to age; 14% were related to sexual orientation and 6% were related to religion. Workplace Incivility DRIs
Fiscal Impact of Workplace Incivility: Of those who experienced work-place related DRIs: 28% lost work time avoiding the instigator of the incivility; 53% lost time worrying about the incident/future interactions; 37% believe their commitment at work declined; 22% have decreased their effort at work; 10% decreased the amount of time that they spent at work; 12% actually changed jobs to avoid the instigator. Source: The Sparticus Group: 2003. Race, Ethnicity and Perceptions of Workplace Relationships in Healthcare Management
White Asian Black Hispanic Race relations within my company Women 79% 60% 41% 55% are good. Men 90% 70% 53% 73% Managers of Color usually have to Women 6% 29% 75% 47% be more qualified to get ahead here. Men 3% 33% 66% 35% White managers share vital growth Women 57% 29% 10% 18% and career-related information with Men 55% 37% 12% 30% managers of color. The evaluation of both whites and Women 69% 51% 18% 33% employees of color are equally Men 75% 50% 22% 43% thorough and carefully evaluated . Has a strong feeling of belonging Women 82% 70% 58% 71% to the organization. Men 85% 72% 72% 79% Source: A Race/Ethnic Comparison of Career Attainment in Healthcare Management: American College of Healthcare Executives; Institute for Diversity in Healthcare Management, 2002 Its Not Just Physicians -- Racial Discrimination Among NBA Referees
Price, Joseph and Wolfers, Justin, "Racial Discrimination Among NBA Referees," NBER Working Paper Series, Vol. w13206 (2007). Available at Does Unconscious Racial Bias Affect Trial Judges?
This article reports the results of the first study of implicit racial bias among judges Jeffrey J. Rachlinski, Sheri Lynn Johnson, Andrew J. Wistrich & Chris Guthrie, Does Unconscious Racial Bias Affect Trial Judges?, 84 Notre Dame L. Rev (2009) Patient Care Considerations
* 07/16/96 Patient Care Considerations * ## The Effect of Race and Sex on Physicians' Recommendations for Cardiac Catheterization
720 physicians viewed recorded interviews Reviewed data about ahypothetical patient The physicians then made recommendations about thatpatient's care Source:Schulman et.al. NEJM 1999;340:618. New Study Finds Unconscious Bias In M.D. Decision-Making
Emergency room doctors in the study were told two men, one white and one African-American, were each 50 years old and complained of chest pain. The patients were not actually real people, but rather computer-generated images seen by the doctors only on a monitor. After the doctors in the study evaluated the two simulated patients, they were then given an implicit association test examining unconscious racial biases. The result was most of the doctors were more likely to prescribe a potentially life-saving, clot-busting treatment for the white patients than for the African-American patient. The study, by the Disparities Solutions Center, affiliated with Harvard University and Massachusetts General Hospital, is the first to deal with unconscious racial bias and how it can lead to inferior care for African-American patients. It was published in the online edition of the Journal of General Internal Medicine in June, 2007. M.D.s and Implicit Bias/IAT
Sabin examined implicit race bias in physicians as a whole using data from Harvards Project Implicit website. Overall, the 2,535 website participants who reported having an MD degree showed significant pro-White bias. J. Health Care Poor Underserved. 2009:20(3): The degree of implicit race bias varies by physician race and gender. In Sabins data, the presence of pro-White bias was significant among physicians of all racial groups except African Americans, who were neutral, while women showed less implicit race bias than men. Lessbut not zeropro-White bias has also been found among non-White vs. White resident physicians and medical students. But Does MD Bias Impact Patient Care?
Demonstrating that physicians have measureable implicit bias does not prove that this bias affects patient-doctor interactions or alters the treatment patients receive. However, research supports a link between disparate treatment decisions and implicit provider bias. This research exists in two forms: studies comparing treatment recommendations for patients who are identical except for social category information and studies directly measuring implicit bias and then determining the correlation between measured bias and physicians treatment decisions. Studies Show MD Bias Impacts Patient Care
Two studies found that Black patients seen in emergency departments receive less analgesia than White patients. Hispanic patients in one study were seven times less likely to receive opioids in the ED than non-Hispanic patients, even after adjusting for other factors. These findings were duplicated in Black patients. In a follow-up study, researchers assessed physicians ability to quantify pain in Hispanic patients compared to non-Hispanic patients. They found that physicians could accurately judge patients pain severity regardless of ethnicity yet still provided less analgesia to Hispanic patients with severe injuries. Todd KH, Lee T, Hoffman JR. The effect of ethnicity on physician estimates of pain severity in patients with isolated extremity trauma. JAMA 1994:271(12)925-8. When Health Care Isnt Caring
Lambda Legal surveyed 4,916 GLBT people and people living with HIV nationwide in the spring of Results showed that these populations were frequently: Denied care; Treated in a discriminatory manner while obtaining care; Subjected to harsh or abusive language by health professionals; Treated by health professionals who refused to touch them or used excessive precautions when doing so; Blamed for their conditions by health professionals When Health Care Isnt Caring
Source: When Health Care Isnt Caring: Lambda Legals Survey on Discrimination Against LGBT People and People Living With HIV, (New York: Lambda Legal, 2010). Available at: When Health Care Isnt Caring
Source: When Health Care Isnt Caring: Lambda Legals Survey on Discrimination Against LGBT People and People Living With HIV, (New York: Lambda Legal, 2010). Available at: How Might Implicit Biases Impact Clinical Objectivity?
Taking a sexual history (presumptions of heterosexuality, monogamy) Incidence of HIV Travel history Country of Origin Health literacy use of the teachback method Making the unusual into the common and vice versa Race Gender U.S. Patient Satisfaction Data Race
Research has found that Hispanic, Asian, and African Americans, compared to whites, report lower quality in their overall interaction with their physicians, less time spent with their physicians, poorer patient-physician communication, diminished trust in their physicians, and less respect from their physicians. A Harvard School of Public Health/Robert Wood Johnson Foundation survey of 4,334 randomly selected U.S. adults compared perceptions of the quality of physician care among fourteen racial and ethnic groups with those of whites. On each measure examined, at least five and as many as eleven subgroups perceived their care to be significantly worse than care for whites. In many instances, subgroups were at least fifteen percentage points more negative than whites. Many of the differences remained after socioeconomic characteristics and language skills were controlled for. Health Affairs, May 08. Picker Inpatient Satisfaction with Doctors By Race, CLIENT A
Question/StatementWhiteOf Color Signif? Didnt always have confidence/trust14.5%26.1% Yes in my doctors. Doctors talked as if I wasnt there %23.2% Yes Courtesy of doctors fair or poor2.5%5.5% Yes Doctors/nurses gave conflicting info.21.5%26.5% Yes *Scores over 20% are considered problems by Picker. Picker Inpatient Satisfaction with Nurses By Race, CLIENT A
Question/StatementWhiteOf Color Signif? Didnt always have confidence/trust24.8%34.7% Yes in my nurses. Nurses talked as if I wasnt there %22.9% Yes Courtesy of nurses fair or poor3.5%5.6% Yes Nurses answers to questions %29.6% Yes werent always understood. *Scores over 20% are considered problems by Picker. Picker Treated with Courtesy, By Race, CLIENT A
Question/StatementWhiteOf Color Signif? Courtesy of admissions staff rated2.0%5.9% Yes fair or poor. Courtesy of people who took blood2.8%8.8% Yes samples rated fair or poor Courtesy of people who brought food5.0%8.8% Yes rated fair or poor. Courtesy of people bringing to and from1.2%6.2% Yes room rated fair or poor. Courtesy of people taking x-rays rated1.4%7.6% Yes Courtesy of people who cleaned room3.3%8.6% Yes rated fair or poor Picker Other Key Indicators of Care By Race, CLIENT A
Question/StatementWhiteOf Color Signif? Not always treated with respect and13.1%21.6% Yes dignity. Didnt always get help in time going 20.4%30.8% Yes to the bathroom. After using call button, had to wait > %4.3% Yes minutes for help. Staff definitely did not do everything19.7% 26.3% Yes they could to control pain. Didnt have enough say about pain26.1% 38.4% Yes control during delivery. Probably would or would not % 28.8% Yes recommend to family/friends. Application Disparities Worlds Apart Video Vignettes
* 07/16/96 Application Disparities Worlds Apart Video Vignettes * ## Worlds Apart: Video Vignettes
Mohammad Kochi, an Afghani man with stomach cancer, refuses chemotherapy in part because of poor communication between his doctors and his daughters, who act as his translators. Robert Phillips, a 29-year-old black man, has been waiting three years for a kidney transplant. He's frustrated with the medical bureaucracy and feels that black patients may not be readily referred for a new kidney because physicians think "they're just going to ruin it anyway." Justine Chitsena, a 4-year-old girl from Laos, needs surgery for a congenital heart defect. Her grandmother adheres to traditional Laotian and Buddhist beliefs and worries that the scar will affect Justine's spirit in her subsequent lives, while her mother worries that her family will blame her if something goes wrong. Video: Mr. Kochis Case Issues Presented: Family decision-making and
* 07/16/96 Video: Mr. Kochis Case Issues Presented: Family decision-making and withholding of information. Effects of immigration and acculturation on family dynamics. Language barriers and communication Religious beliefs, spirituality and negotiation. The culture of American medicine. * ## Mr. Kochis Case Take-Away Points
Assess how patients prefer to make medical decisions individually, as a family, or through a specific authority figure and try to accommodate. Even within the same culture, there can be very different values and perspectives on health and illness. Language barriers can lead to miscommunication and poor health outcomes. LEP patients have a legal right to an interpreter. Use of family members risks poor communication. A non-confrontational patient may be agreeing with the physician just to show respect Beliefs in fatalism can lead a patient to refuse treatment, but they are rarely absolute. Probe for a deeper understanding of patients beliefs and negotiate acceptable treatment options. Video: Robert Phillips Case
* 07/16/96 Video: Robert Phillips Case Issues Presented: The distinction between disease and illness Discrimination and racial/ethnic disparities in health care. Stereotyping and clinical decision-making Mistrust and communication style Conflict style differences * ## Video: Justine Chitsenas Case
* 07/16/96 Video: Justine Chitsenas Case Issues Presented: Understanding the familys health and illness beliefs. Family decision-making and authority figures. Traditional/alternative medical practices Cross-cultural medical ethics. Barriers to effective communication. * ## Battling Bias What Works?
* 07/16/96 Battling Bias What Works? * ## Useful Metaphors for Unconscious Bias
Unconscious bias is like a chronic illness it needs constant monitoring and attention. We are all carriers Unconscious bias is like a pilot flying above the clouds. With no reference point on land, pilots must learn to fly using instrument panels Key is Pattern Recognition Battling Bias As Individuals
* 07/16/96 Battling Bias As Individuals Use tools to explore your own unconscious biases (IAT, ICS) Slow down, shift from think fast brain systems (amygdala) to think slow brain systems (pre-frontal cortex). (Daniel Kahneman) In particular, there are several strategies that appear to make a difference: A.Information re: the psychological basis of bias B.Motivation - internal (vs. external) motivation to change C.Individuation learning to see diverse others as individuals rather than as members of groups. D.Direct contact with members of other groups. E.Working together on teams, as equals, in pursuit of common goals. F.Context/environment display positive images of leaders from diverse groups 4.Obtain 360 degree feedback from diverse employees/colleagues. Reverse mentoring processes can also help. * ## * 07/16/96 Perspective-Taking Another strategy to mitigate the impact of implicit bias is perspective-taking. Perspective taking is a conscious attempt to envision another persons viewpoint. Drwecki et. al. applied perspective-taking in a clinical setting. Nurses were shown pictures of either Black or White patients with genuine expressions of pain and asked how much pain medication they recommended. Nurses told to use their best judgment recommended significantly more pain medication for White than Black patients, whereas Nurses instructed to imagine how the patient felt recommended equal analgesic treatment regardless of race.Drwecki BB, Moore, CF Ward SE, Prkachin KM. Reducing racial disparities in pain treatment: the role of empathy and perspective taking. Pain. 2011:152(5): * ## Reducing Bias In Physicians
* 07/16/96 Reducing Bias In Physicians Racial and ethnic disparities in the quality of medical outcomes are widely documented. A.Institute of Medicine report Unequal Treatment (2002) B.Research indicates that little progress has been made since 2002. Michelle van Ryn of the University of Minnesota and colleagues recently published some evidence-based recommendations for combating bias among health care providers. See: Burgess, D., van Ryn, M., et. al. Reducing Racial Bias Among Health Care Providers: Lessons from Social-Cognitive Psychology, 22 Journal of General Internal Medicine (2007). Their conceptual model (depicted on the next slide), recognizes the importance of motivation, information and skills as key ingredients for successful interventions. The next slide discusses the approach advocated by van Ryn. * ## Can Implicit Bias Be Controlled?
* 07/16/96 Can Implicit Bias Be Controlled? * ## Battling Bias Within Hospitals
* 07/16/96 Battling Bias Within Hospitals Collect patient race, ethnicity and language (REL) data. Tie patient REL data to patient outcomes. (