PERSPEC TIVE
# B LACK LIVES MATTER
tives to demand change in law, policy, and practice. Rightfully or not, medical professionals often have a societal status that gives our voices greater credibility. Af ter the grand-jury decision last November not to indict the police officer who shot a black teenager in Ferguson, Missouri, I wrote to my staff noting that in this time An audio interview of public Outcry, w ith Dr. Bassett it is important is available a t NEJM .org to assert our un wavering commitment to reduc ing health disparities. We can all do at least that. As a mother o f black children, I feel a personal urgency for soci ety to acknowledge racism’s im pact on the everyday lives of mil
lions of people in the United States and elsewhere and to act to end discrimination. As a doctor and New York City’s health com missioner, I believe that health professionals have much to con tribute to that debate and pro cess. Let’s not sit on the sidelines.
2. H arp er S, MacLehose RF, Kaufm an JS. Trends in the black-white life expectancy gap am ong US states, 1990-2009. H ealth A ff
Disclosure forms provided by the author are available with the full text of this arti cle at NEJM.org.
(h ttp ://w h o .in t/a b o u t/d e fin itio n /e n /p rin t .h tm l). 4. W ashington HA. M edical apartheid: the
From th e O ffic e o f th e C om m issioner, New York C ity D e p a rtm e n t o f H ealth and M e ntal
dark histo ry o f m edical e xp erim e ntatio n on black A m ericans fro m colon ial tim e s to the present. New York: Doubleday, 2006.
H ygiene, N ew York. This a rtic le was pu blished on February 18, 2015, at NEJM.org. 1. Krieger N. D is c rim in a tio n and health in eq uities. In: Berkm an LF, Kawachi I, Glym o u r M , eds. Social ep id e m io lo g y. 2nd ed. New York: O xfo rd U n ive rsity Press, 2014:63125.
(M illw o o d ) 2014;33:1375-82. 3. Preamble to th e c o n s titu tio n o f th e W orld H ealth O rg anizatio n as adopted by the Inter national H ealth Conference, N ew York, 1922 June, 1946: signed on 22 July 1946 by the representatives o f 61 States (o fficial records o f the W orld H ealth O rg anizatio n, no. 2, p. 100) and entered in to force on 7 A p ril 1948
5. D ive rsity in th e physician w o rkforce: fa c ts & fig u re s . W a shing ton, DC: A ssocia tio n o f A m erica n M edical Colleges, 2014 (h ttp ://a a m c d iv e rs ity fa c ts a n d fig u re s .o rg / s e c tio n -ii-c u rre n t-s ta tu s -o f-u s -p h y s ic ia n -w o rk fo rc e /).
DOI: 10.1056/NEJMpl500529 C opyright © 2015 M assachusetts M e d ica l Society.
Bias, Black Lives, and Academic Medicine D avid A. A n se ll, M .D ., M .P .H ., and E dw in
t noon Pacific Standard Time on December 10, 2014, thou sands o f students from 70 medi cal schools throughout the United States held silent “White Coats for Black Lives” die-ins. These demonstrations, the largest coor dinated protests at U.S. medical schools since the Vietnam War era, were initiated by medical students in California and spread across the country in response to the follow ing call to action posted online at thefreethoughtproject.com: “We feel it is essential to be gin a conversation about our role in addressing the explicit and im plicit discrimination and racism in our communities and reflect on the systemic biases embedded in our medical education curricula, clinical learning environments, and administrative decision-making. We believe these discussions are needed at academic medical cen ters nationwide.” Though the stim
A
K.
M c D o n a ld , M .D .
ulus for the die-ins was the nation wide protests in response to the killing of unarmed black men by police officers, the students de manded an examination of racial bias within our country’s aca demic medical centers. What are the systemic biases within academic medical centers, and what do they have to do with black lives? Two observations about health care disparities may be relevant. First, there is evidence that doc tors hold stereotypes based on pa tients’ race that can influence their clinical decisions.*1 Implicit bias refers to unconscious racial ste reotypes that grow from our per sonal and cultural experiences. These implicit beliefs may also stem from a lack of day-to-day in terracial and intercultural interac tions. Although explicit race bias is rare among physicians, an un conscious preference for whites N ENGLJ
M ED 3 7 2 ;i2
N E JM .O R G
MARCH
1 9 ,2 0 1 5
as compared with blacks is com monly revealed on tests o f im plicit bias.1 Second, despite physicians’ and medical centers’ best intentions of being equitable, black-white dis parities persist in patient out comes, medical education, and faculty recruitment. In the 2002 report Unequal Treatment, the In stitute of Medicine (IOM) reviewed hundreds o f studies o f age, sex, and racial differences in medical diagnoses, treatments, and health care outcomes.2 The IOM’s con clusion was that for almost every disease studied, black Americans received less effective care than white Americans. These dispari ties persisted despite matching for socioeconomic and insurance status. Minority patients received fewer recommended treatments for diseases ranging from AIDS to cancer to heart disease. And ra cial gaps in health care outcomes 1087
PERSPECTIVE
B IA S , B L A C K LIVES , A N D A C A D E M IC M E D IC IN E
W h ite m e n
—■ —
A s ia n m e n
—■ —
B la ck m e n
—■ —
H is p a n ic m e n
W h ite w o m e n
—■ —
A s ia n w o m e n
—■ —
B la ck w o m e n
—g j —
H is p a n ic w o m e n
B —■ —
N a tiv e A m e ric a n m e n N a tiv e A m e ric a n w o m e n
White M en
9972
1 1 ,2 8 1
1 0 ,1 0 5
8617
8005
8084
7727
6821
6763
W om en
2589
3 ,7 7 5
4 ,3 8 0
4344
4925
5117
5712
5895
5533
Asian M en
400
652
924
1526
2188
3191
3452
2777
2427
W om en
120
270
516
951
1470
2313
2996
2838
2434
M en
411
456
468
471
516
614
632
569
517
W om en
286
362
405
493
675
824
893
954
880
Black
Hispanic or Latino M en
369
629
706
696
741
976
1024
912
766
W om en
1 25
276
345
384
525
721
869
856
696
30
47
51
31
63
83
77
90
81
3
19
21
30
48
73
57
85
72
American Indian or Alaska Native M en W om en
Number o f U.S. Physicians by Graduation Year, Race, Ethnic Group, and Sex, 1980-2012. D ata are fro m th e A s s o c ia tio n o f A m e ric a n M e d ic a l C o lleges. An in te ra c tiv e g ra p h ic is a va ila b le a t N E JM .org.
have persisted. For example, gaps in blood pressure, cholesterol, and glycated hemoglobin control be tween black and white members of Medicare health maintenance organizations were found through out the period 2006 to 2011.3 The IOM found “strong but circumstantial evidence for the role of bias, stereotyping, and prejudice” in perpetuating racial health disparities.2 The finding that physicians have implicit ra cial bias does not prove that it affects patient-doctor relationships or changes treatment decisions. But some research suggests that there’s a direct relationship among 1088
physicians’ implicit bias, mistrust on the part of black patients, and clinical outcomes.1 Although the causes o f health care disparities are certainly multifactorial, im plicit bias plays some role. Implicit bias may also influ ence administrative decisions at academic medical centers — de cisions ranging from what ser vices are provided, to whether to accept insurance plans that serve the most disadvantaged members of minority groups, to which neighborhoods to choose when establishing new physicians’ of fices. The likelihood of such in fluence does not mean that bias N ENGLJ MED 372J12
NEJM.ORG
MARCH 19, 2015
is the only explanation for un equal treatment or administrative decisions that favor one group over another. The point is simply that there is potential for making racially biased decisions, and it generally goes unexamined. Implicit racial bias might con tribute to the failure to achieve greater inclusion o f black stu dents in medical education. Though there has been progress in the recruitment of some un derrepresented minority groups to medical schools, the percent age o f black men among all medical school graduates has de clined over the past 20 years (see
PERSPECTIVE
BIAS, BLACK LIVES, A N D A C A D E M I C M E D I C I N E
graph). The country’s tradition just 11% of the 3952 black faculty ally black medical colleges — members. The paucity of black Howard, Meharry, and Morehouse faculty members contributes to a — continue to graduate a dispro climate in which black medical portionate number o f black med students may lack accessible ical students. In 2012, there were black role models. The IOM has just 517 black men among the defined the climate for diversity more than 20,000 graduating stu as “the perceptions, attitudes, dents at U.S. medical schools (see and expectations that define the graph). Black medical students institution, particularly as seen are more than twice as likely as from the perspectives of individ white students to express a desire uals o f different racial or ethnic to care for underserved communi backgrounds.” Though there may ties of color. Our inability to re be various drivers of poor re cruit black men into medicine is cruitment, retention, and promo alarming, given the urgency of tion of black faculty members, racial health care disparities in the role o f institutional bias and the United States. the climate for black faculty at Recruitment and retention of academic medical centers deserve black faculty members have also scrutiny. By any measure, aca long challenged academic medi demic medicine’s persistent diffi cine. Only 2.9% of all faculty culty in developing black faculty members at U.S. medical schools members is a serious concern. are black.*1234 A 2010 study showed For the sake of not only black that among faculty members lives but all lives, we should heed who had been hired in 2000, our students’ call to examine the blacks were less likely to have implicit biases in our academic been retained than any other medical centers. We can begin by demographic group. Black facul assessing how bias contributes ty members are less likely than to the persistence o f black-white their white counterparts to be disparities in health care, medi promoted, to hold senior faculty cal school recruitment, and fac or administrative positions, and ulty retention in our own institu to receive research awards from tions. We can audit the care we the National Institutes of Health.5 deliver to ensure that the right Thirty-one percent of the 84,195 treatments are provided and the white faculty members at U.S. best outcomes are achieved re medical schools were full profes gardless of patients’ race, class, sors in 2011, as compared with or sex. We can assess the climate
within our centers and strive to ensure that our recruitment pro cesses, classrooms, clinics, admin istrations, and boardrooms are inclusive to all. But most impor tant, we should talk about bias, with our students, our faculties, our staff, our administrations, and our patients. Maybe then we’ll have a chance to finally eliminate the racial health care disparities that persist in the United States. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. From th e D e p a rtm e n t o f Interna l M e dicine, Rush U n iv e rs ity M e dical Center, Chicago. T h is a rtic le was p u blished on February 18, 2015, atN E JM .o rg . 1. Chapm an EN, Kaatz A, Carnes M . Physi cians and im p lic it bias: ho w d o c to rs may u n w ittin g ly perpetuate health care d is p a ri ties. J Gen Intern M ed 2013;28:1504-10. 2. Smedley BD, Stith AY, N elson AR, eds. Unequal tre a tm e n t: c o n fro n tin g racial and eth nic d ispa rities in healthcare. W ashington, DC: N ational A cadem y Press, 2002. 3. Ayanian JZ, Landon BE, N ew house JP, Zaslavsky A M . Racial and eth n ic disparities a m ong enrollees in M edicare Advantage plans. N Engl J M ed 2014;371:2288-97. 4. Castillo-Page L. D iversity in m edical ed u cation: facts and figu res 2012. W ashington, DC: Am erican A ssociation o f M edical Colleg es, 2012 (http s ://m e m b e rs .a a m c .o rg /e w e b / u p lo a d /D iv e rs ity % 2 0 in % 2 0 M e d ic a l % 2 0 E d u ca tion_ F acts% 20 and % 2 0F igu res % 202012.pdf). 5. Guevara JP, Adanga E, Avakam e E, Carth o n MB. M in o rity facu lty d e velopm ent p ro gram s and un derrepresented m in o rity fac ulty representation at US m edical schools. JA M A 2013;310:2297-304. DOI: 10.1056/NEJMpl500832 C opyright © 2015 M assachusetts M e d ica l Society.
Having and Fighting Ebola — Public Health Lessons from a Clinician Turned Patient Craig Spencer, M.D., M.P.H.
hile treating patients with Ebola in Guinea, I kept a journal to record my perceived level o f risk o f being infected with the deadly virus. A friend who’d volunteered previously had
W
told me that such a journal com forted him when he looked back and saw no serious breach of protocol or significant exposure. On a spreadsheet delineating three levels of risk — minimal, N EN G LJ
M E D 372J12
N E J M .O R G
M ARCH
19, 2015
moderate, and high — I’d been able to check off minimal risk every day after caring for patients. Yet on October 23, 2014, I en tered Bellevue Hospital as New York City’s first Ebola patient. 1089
Copyright © Massachusetts Medical Society 2015.
# B LACK LIVES MATTER
tives to demand change in law, policy, and practice. Rightfully or not, medical professionals often have a societal status that gives our voices greater credibility. Af ter the grand-jury decision last November not to indict the police officer who shot a black teenager in Ferguson, Missouri, I wrote to my staff noting that in this time An audio interview of public Outcry, w ith Dr. Bassett it is important is available a t NEJM .org to assert our un wavering commitment to reduc ing health disparities. We can all do at least that. As a mother o f black children, I feel a personal urgency for soci ety to acknowledge racism’s im pact on the everyday lives of mil
lions of people in the United States and elsewhere and to act to end discrimination. As a doctor and New York City’s health com missioner, I believe that health professionals have much to con tribute to that debate and pro cess. Let’s not sit on the sidelines.
2. H arp er S, MacLehose RF, Kaufm an JS. Trends in the black-white life expectancy gap am ong US states, 1990-2009. H ealth A ff
Disclosure forms provided by the author are available with the full text of this arti cle at NEJM.org.
(h ttp ://w h o .in t/a b o u t/d e fin itio n /e n /p rin t .h tm l). 4. W ashington HA. M edical apartheid: the
From th e O ffic e o f th e C om m issioner, New York C ity D e p a rtm e n t o f H ealth and M e ntal
dark histo ry o f m edical e xp erim e ntatio n on black A m ericans fro m colon ial tim e s to the present. New York: Doubleday, 2006.
H ygiene, N ew York. This a rtic le was pu blished on February 18, 2015, at NEJM.org. 1. Krieger N. D is c rim in a tio n and health in eq uities. In: Berkm an LF, Kawachi I, Glym o u r M , eds. Social ep id e m io lo g y. 2nd ed. New York: O xfo rd U n ive rsity Press, 2014:63125.
(M illw o o d ) 2014;33:1375-82. 3. Preamble to th e c o n s titu tio n o f th e W orld H ealth O rg anizatio n as adopted by the Inter national H ealth Conference, N ew York, 1922 June, 1946: signed on 22 July 1946 by the representatives o f 61 States (o fficial records o f the W orld H ealth O rg anizatio n, no. 2, p. 100) and entered in to force on 7 A p ril 1948
5. D ive rsity in th e physician w o rkforce: fa c ts & fig u re s . W a shing ton, DC: A ssocia tio n o f A m erica n M edical Colleges, 2014 (h ttp ://a a m c d iv e rs ity fa c ts a n d fig u re s .o rg / s e c tio n -ii-c u rre n t-s ta tu s -o f-u s -p h y s ic ia n -w o rk fo rc e /).
DOI: 10.1056/NEJMpl500529 C opyright © 2015 M assachusetts M e d ica l Society.
Bias, Black Lives, and Academic Medicine D avid A. A n se ll, M .D ., M .P .H ., and E dw in
t noon Pacific Standard Time on December 10, 2014, thou sands o f students from 70 medi cal schools throughout the United States held silent “White Coats for Black Lives” die-ins. These demonstrations, the largest coor dinated protests at U.S. medical schools since the Vietnam War era, were initiated by medical students in California and spread across the country in response to the follow ing call to action posted online at thefreethoughtproject.com: “We feel it is essential to be gin a conversation about our role in addressing the explicit and im plicit discrimination and racism in our communities and reflect on the systemic biases embedded in our medical education curricula, clinical learning environments, and administrative decision-making. We believe these discussions are needed at academic medical cen ters nationwide.” Though the stim
A
K.
M c D o n a ld , M .D .
ulus for the die-ins was the nation wide protests in response to the killing of unarmed black men by police officers, the students de manded an examination of racial bias within our country’s aca demic medical centers. What are the systemic biases within academic medical centers, and what do they have to do with black lives? Two observations about health care disparities may be relevant. First, there is evidence that doc tors hold stereotypes based on pa tients’ race that can influence their clinical decisions.*1 Implicit bias refers to unconscious racial ste reotypes that grow from our per sonal and cultural experiences. These implicit beliefs may also stem from a lack of day-to-day in terracial and intercultural interac tions. Although explicit race bias is rare among physicians, an un conscious preference for whites N ENGLJ
M ED 3 7 2 ;i2
N E JM .O R G
MARCH
1 9 ,2 0 1 5
as compared with blacks is com monly revealed on tests o f im plicit bias.1 Second, despite physicians’ and medical centers’ best intentions of being equitable, black-white dis parities persist in patient out comes, medical education, and faculty recruitment. In the 2002 report Unequal Treatment, the In stitute of Medicine (IOM) reviewed hundreds o f studies o f age, sex, and racial differences in medical diagnoses, treatments, and health care outcomes.2 The IOM’s con clusion was that for almost every disease studied, black Americans received less effective care than white Americans. These dispari ties persisted despite matching for socioeconomic and insurance status. Minority patients received fewer recommended treatments for diseases ranging from AIDS to cancer to heart disease. And ra cial gaps in health care outcomes 1087
PERSPECTIVE
B IA S , B L A C K LIVES , A N D A C A D E M IC M E D IC IN E
W h ite m e n
—■ —
A s ia n m e n
—■ —
B la ck m e n
—■ —
H is p a n ic m e n
W h ite w o m e n
—■ —
A s ia n w o m e n
—■ —
B la ck w o m e n
—g j —
H is p a n ic w o m e n
B —■ —
N a tiv e A m e ric a n m e n N a tiv e A m e ric a n w o m e n
White M en
9972
1 1 ,2 8 1
1 0 ,1 0 5
8617
8005
8084
7727
6821
6763
W om en
2589
3 ,7 7 5
4 ,3 8 0
4344
4925
5117
5712
5895
5533
Asian M en
400
652
924
1526
2188
3191
3452
2777
2427
W om en
120
270
516
951
1470
2313
2996
2838
2434
M en
411
456
468
471
516
614
632
569
517
W om en
286
362
405
493
675
824
893
954
880
Black
Hispanic or Latino M en
369
629
706
696
741
976
1024
912
766
W om en
1 25
276
345
384
525
721
869
856
696
30
47
51
31
63
83
77
90
81
3
19
21
30
48
73
57
85
72
American Indian or Alaska Native M en W om en
Number o f U.S. Physicians by Graduation Year, Race, Ethnic Group, and Sex, 1980-2012. D ata are fro m th e A s s o c ia tio n o f A m e ric a n M e d ic a l C o lleges. An in te ra c tiv e g ra p h ic is a va ila b le a t N E JM .org.
have persisted. For example, gaps in blood pressure, cholesterol, and glycated hemoglobin control be tween black and white members of Medicare health maintenance organizations were found through out the period 2006 to 2011.3 The IOM found “strong but circumstantial evidence for the role of bias, stereotyping, and prejudice” in perpetuating racial health disparities.2 The finding that physicians have implicit ra cial bias does not prove that it affects patient-doctor relationships or changes treatment decisions. But some research suggests that there’s a direct relationship among 1088
physicians’ implicit bias, mistrust on the part of black patients, and clinical outcomes.1 Although the causes o f health care disparities are certainly multifactorial, im plicit bias plays some role. Implicit bias may also influ ence administrative decisions at academic medical centers — de cisions ranging from what ser vices are provided, to whether to accept insurance plans that serve the most disadvantaged members of minority groups, to which neighborhoods to choose when establishing new physicians’ of fices. The likelihood of such in fluence does not mean that bias N ENGLJ MED 372J12
NEJM.ORG
MARCH 19, 2015
is the only explanation for un equal treatment or administrative decisions that favor one group over another. The point is simply that there is potential for making racially biased decisions, and it generally goes unexamined. Implicit racial bias might con tribute to the failure to achieve greater inclusion o f black stu dents in medical education. Though there has been progress in the recruitment of some un derrepresented minority groups to medical schools, the percent age o f black men among all medical school graduates has de clined over the past 20 years (see
PERSPECTIVE
BIAS, BLACK LIVES, A N D A C A D E M I C M E D I C I N E
graph). The country’s tradition just 11% of the 3952 black faculty ally black medical colleges — members. The paucity of black Howard, Meharry, and Morehouse faculty members contributes to a — continue to graduate a dispro climate in which black medical portionate number o f black med students may lack accessible ical students. In 2012, there were black role models. The IOM has just 517 black men among the defined the climate for diversity more than 20,000 graduating stu as “the perceptions, attitudes, dents at U.S. medical schools (see and expectations that define the graph). Black medical students institution, particularly as seen are more than twice as likely as from the perspectives of individ white students to express a desire uals o f different racial or ethnic to care for underserved communi backgrounds.” Though there may ties of color. Our inability to re be various drivers of poor re cruit black men into medicine is cruitment, retention, and promo alarming, given the urgency of tion of black faculty members, racial health care disparities in the role o f institutional bias and the United States. the climate for black faculty at Recruitment and retention of academic medical centers deserve black faculty members have also scrutiny. By any measure, aca long challenged academic medi demic medicine’s persistent diffi cine. Only 2.9% of all faculty culty in developing black faculty members at U.S. medical schools members is a serious concern. are black.*1234 A 2010 study showed For the sake of not only black that among faculty members lives but all lives, we should heed who had been hired in 2000, our students’ call to examine the blacks were less likely to have implicit biases in our academic been retained than any other medical centers. We can begin by demographic group. Black facul assessing how bias contributes ty members are less likely than to the persistence o f black-white their white counterparts to be disparities in health care, medi promoted, to hold senior faculty cal school recruitment, and fac or administrative positions, and ulty retention in our own institu to receive research awards from tions. We can audit the care we the National Institutes of Health.5 deliver to ensure that the right Thirty-one percent of the 84,195 treatments are provided and the white faculty members at U.S. best outcomes are achieved re medical schools were full profes gardless of patients’ race, class, sors in 2011, as compared with or sex. We can assess the climate
within our centers and strive to ensure that our recruitment pro cesses, classrooms, clinics, admin istrations, and boardrooms are inclusive to all. But most impor tant, we should talk about bias, with our students, our faculties, our staff, our administrations, and our patients. Maybe then we’ll have a chance to finally eliminate the racial health care disparities that persist in the United States. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. From th e D e p a rtm e n t o f Interna l M e dicine, Rush U n iv e rs ity M e dical Center, Chicago. T h is a rtic le was p u blished on February 18, 2015, atN E JM .o rg . 1. Chapm an EN, Kaatz A, Carnes M . Physi cians and im p lic it bias: ho w d o c to rs may u n w ittin g ly perpetuate health care d is p a ri ties. J Gen Intern M ed 2013;28:1504-10. 2. Smedley BD, Stith AY, N elson AR, eds. Unequal tre a tm e n t: c o n fro n tin g racial and eth nic d ispa rities in healthcare. W ashington, DC: N ational A cadem y Press, 2002. 3. Ayanian JZ, Landon BE, N ew house JP, Zaslavsky A M . Racial and eth n ic disparities a m ong enrollees in M edicare Advantage plans. N Engl J M ed 2014;371:2288-97. 4. Castillo-Page L. D iversity in m edical ed u cation: facts and figu res 2012. W ashington, DC: Am erican A ssociation o f M edical Colleg es, 2012 (http s ://m e m b e rs .a a m c .o rg /e w e b / u p lo a d /D iv e rs ity % 2 0 in % 2 0 M e d ic a l % 2 0 E d u ca tion_ F acts% 20 and % 2 0F igu res % 202012.pdf). 5. Guevara JP, Adanga E, Avakam e E, Carth o n MB. M in o rity facu lty d e velopm ent p ro gram s and un derrepresented m in o rity fac ulty representation at US m edical schools. JA M A 2013;310:2297-304. DOI: 10.1056/NEJMpl500832 C opyright © 2015 M assachusetts M e d ica l Society.
Having and Fighting Ebola — Public Health Lessons from a Clinician Turned Patient Craig Spencer, M.D., M.P.H.
hile treating patients with Ebola in Guinea, I kept a journal to record my perceived level o f risk o f being infected with the deadly virus. A friend who’d volunteered previously had
W
told me that such a journal com forted him when he looked back and saw no serious breach of protocol or significant exposure. On a spreadsheet delineating three levels of risk — minimal, N EN G LJ
M E D 372J12
N E J M .O R G
M ARCH
19, 2015
moderate, and high — I’d been able to check off minimal risk every day after caring for patients. Yet on October 23, 2014, I en tered Bellevue Hospital as New York City’s first Ebola patient. 1089
Copyright © Massachusetts Medical Society 2015.
