60PART 1 The Profession of Medicine
race/ethnicity, education, and socioeconomic status have persisted. For
example, at every level of education and income, African Americans have
lower life expectancy at age 25 than whites and Hispanics/Latinos. Blacks
with a college degree or more education have lower life expectancy than
whites and Hispanics who graduated from high school. Blacks have had
lower life expectancy compared to whites for as long as data have been
collected. From 1975 to 2003, the largest difference in life expectancy
between blacks and whites was substantial (6.3 years for males and 4.5
years for females) (Fig. 10-1). The gap in life expectancy between the black
and white populations decreased by 2.3 years between 1999 and 2013 from
5.9 to 3.6 years (4.4 years for males and 3.0 years for women) (Fig. 10-2).
The life expectancy gap is augmented by worse health and higher
disease burden. Cardiovascular-related diseases remain the leading
cause of black-white differences in life expectancy. If all cardiovascular
causes and diabetes are considered together, they account for 35% and
52% of the gap for males and females, respectively. Finally, place matters
for health. Analysis of data from 2010 to 2015 demonstrate large geographic life expectancy gap variation at the census tract level (Fig. 10-3).
Socioeconomic and race/ethnicity factors, behavioral and metabolic risk
factors (prevalence of obesity, leisure-time physical inactivity, cigarette
smoking, hypertension, diabetes), and health care factors (percentage
of the population younger than 65 years who are insured, primary care
access and quality, number of physicians per capita) explained 60%,
74%, and 27% of county-level variation in life expectancy, respectively.
Combined, these factors explained 74% of this variation. Most of the
association between socioeconomic and race/ethnicity factors and life
expectancy was mediated through behavioral and metabolic risk factors.
In addition to racial and ethnic disparities in health, there are racial
and ethnic disparities in the quality of care for persons with access to
White females
Life expectancy (years)
60
70
1985 1995 2005
80
65
75
85
Black females
1985 1995 2005
Year
White males
1985 1995 2005
Black males
1985 1995 2005
FIGURE 10-1 Life expectancy at birth among black and white males and females in the United States, 1975–2003. (Adapted from S Harper, J Lynch, S Burris, GD Smith:
Trends in the black-white life expectancy gap in the United States, 1983-2003. JAMA 297:1224, 2007.)
White male
Black male
1999
0
65
70
75
80
85
2001
Age (years)
2003 2005 2007 2009 2011 2013
Black
Black female
White female
White
FIGURE 10-2 Life expectancy, by race and sex: United States, 1999–2013. (From KD
Kochanek et al: NCHS Data Brief 218:1, 2015.)
the health care system. Seminal studies over several decades have consistently documented disparities in health care. For instance, studies
have documented disparities in the treatment of pneumonia and congestive heart failure, with blacks receiving less optimal care than whites
when hospitalized for these conditions. Moreover, blacks with endstage renal disease are referred less often to the transplant list than are
their white counterparts (Fig. 10-4). Disparities have been found, for
example, in the use of cardiac diagnostic and therapeutic procedures
(with blacks being referred less often than whites for cardiac catheterization and bypass grafting), prescription of analgesia for pain control
(with blacks and Hispanics/Latinos receiving less pain medication than
whites for long-bone fractures and cancer), and surgical treatment of
lung cancer (with blacks receiving less curative surgery than whites
for non-small-cell lung cancer). Again, many of these disparities have
occurred even when variations in factors such as insurance status,
income, age, comorbid conditions, and symptom expression are taken
into account. Finally, disparities in the quality of care provided at the
sites where minorities tend to receive care have been shown to be an
important additional contributor to overall disparities.
The 2019 National Healthcare Quality and Disparities Report, released
by the Agency for Healthcare Research and Quality, tracks about 250 health
care process, outcome, and access measures, across many diseases and settings. This annual report is particularly important because most studies
of disparities have not been longitudinally repeated with the same methodology to document trends and changes in disparities over time. This
report found that some disparities were getting smaller from 2000 through
2016–2018, but disparities persisted and some even worsened, especially
for poor and uninsured populations. For about 40% of quality measures,
blacks (82 of 202 measures) and American Indians and Alaska Natives
(47 of 116 measures) received worse care than whites. For more than onethird of quality measures, Hispanics (61 of 177 measures) received worse
care than whites. Asians and Native Hawaiians/Pacific Islanders received
worse care than whites for about 30% of quality measures, but Asians also
received better care for about 30% of quality measures (Fig. 10-5). Of note,
for those quality measures that demonstrated disparities at baseline, >90%
of these measures showed no improvement since 2000 (Fig. 10-6).
■ ROOT CAUSES OF DISPARITIES
Race, Racism, and Health Race and racism are core elements of any
explanatory model on racial and ethnic disparities in health and health
care. Our nation’s history of slavery, segregation, separate but “equal”
health care, and medical experimentation, among a myriad of other ways
in which racism has manifested in the United States, has played a key
role in the existence and persistence of these disparities. It is now well
accepted that race is a social category without biologic foundation and
a product of historical racism. Nevertheless, it is clear that racism has a
biologic impact as a form of psychosocial stress. It is now well established
Racial and Ethnic Disparities in Health Care
61CHAPTER 10
Life Expectancy at birth (Quintiles)
Geographic areas with no data available are filled in gray
56.9–75.1 75.2–77.5 77.6–79.5 79.6–81.6 81.7–97.5
FIGURE 10-3 Life expectancy at birth for U.S. census tracts, 2010-2015. (A New View of Life Expectancy, Surveillance and Data - Blogs and Stories, Centers for Disease
Control and Prevention. Retrieved from https://www.cdc.gov/surveillance/blogs-stories/life-expectancy.html.)
100
80
60
40
20
0
59.6
80.3
57.9
82.2
40.3
68.9
40.6
67.9
Referred
for evaluation
Placed on waiting list
or received transplant
Percentage of patients
Black women
White women
Black men
White men
FIGURE 10-4 Referral for evaluation at a transplantation center or placement on a
waiting list/receipt of a renal transplant within 18 months after the start of dialysis
among patients who wanted a transplant, according to race and sex. The reference
population consisted of 239 black women, 280 white women, 271 black men, and
271 white men. Racial differences were statistically significant among both the
women and the men (p < .0001 for each comparison). (From JZ Ayanian, PD Cleary,
JS Weissman, AM Epstein: The effect of patients’ preferences on racial differences
in access to renal transplantation. N Engl J Med 341:1661,1999. Copyright © 1999
Massachusetts Medical Society. Reprinted with permission from Massachusetts
Medical Society.)
poorer adherence to medical regimens provide an additional important
pathway through which stressors influence disease risk. This accelerated disease risk, aging, and premature death has been termed the
weathering effect.
While most empiric research focuses on interpersonal racial/
ethnic discrimination, structural racism (sometimes called institutional
racism) provides a more holistic framework. Structural racism refers to
the totality of ways that a society fosters, sustains, and reinforces discrimination through sociopolitical, legal, economic, and health structures
that determine differential access to risks, opportunities, and resources
that drive health and health care disparities. Structural racism explains
how racism’s structure and ideology can persist in governmental and
institutional policies in the absence of individual actors who are explicitly
racially prejudiced. For example, the history of residential segregation
has had lasting negative effects generationally on equal access for racial/
ethnic minorities to employment, banking, earnings, high-quality education, and health care. Policies that do not address root structural causes
will not address health and health care inequities.
With the promise of individualizing clinical decisions, the use of
race in clinical and risk assessment algorithms has long been a part of
modern medicine. The evidence is now clear that race is not a reliable
proxy for genetic difference and that race adjustment has the potential
to create inadvertent disparities in health care. One clinical example is
from nephrology. Blacks have higher rates of end-stage kidney disease
and death due to kidney failure than the overall population. The most
widely used cohort-derived equation to estimate glomerular filtration
rate (GFR), the Chronic Kidney Disease Epidemiology Collaboration
(CKD-EPI) equation, has the limitation that it produces 80–90% estimated GFR (eGFR) values that are within ±30% of a patient’s measured
GFR. In addition, this equation uses a black race-related factor, which
increases eGFR for any given serum creatinine by 15.9% compared to
a nonblack patient with the same age, sex, and serum creatinine. The
increase in eGFR is likely to disadvantage blacks for early referral to a
nephrologist, early treatment of advanced chronic kidney disease, and
that psychosocial stress negatively impacts health through psychophysiologic reactivity causing hyperstimulation of the sympathetic-adrenalmedullary system and the hypothalamic-pituitary-adrenal axis, leading
to vascular inflammation, endothelial dysfunction, and neurohormonal dysregulation causing an acceleration of cardiovascular disease.
Behavioral changes occurring as adaptations or coping responses to
stressors such as increased smoking, decreased exercise and sleep, and
62PART 1 The Profession of Medicine
Better
Black (n=202) Asian (n=185) AI/AN (n=116) NHPI (n=72) Hispanic, all races
(n=177)
23
0%
20%
40%
60%
80%
100%
97
82
56
77
52
13
56
11
37
24 47
38
78
61
Same Worse
FIGURE 10-5 Number and percentage of quality measures for which members of selected groups experienced better,
same, or worse quality of care compared with reference group (white) for the most recent data year, 2014, 2016, 2017, or
2018. AI/AN, American Indian or Alaska Native; NHPI, Native Hawaiian/Pacific Islander (From 2019 National Healthcare
Quality and Disparities Report. Rockville, MD: Agency for Healthcare Research and Quality; December 2020. AHRQ Pub.
No. 20(21)-0045-EF.)
Black (n=58)
3
Improving Not changing Worsening
Asian (n=37)
1
AI/AN (n=34)
2
NHPI (n=16)
1
Hispanic, all races
(n=53)
5
55 35 32 15 48
0%
20%
40%
60%
80%
100%
FIGURE 10-6 Number and percentage of quality measures with disparity at baseline for which disparities related to
race and ethnicity were improving, not changing, or worsening over time, 2000 through 2014, 2015, 2016, 2017, or 2018. AI/
AN, American Indian or Alaska Native; NHPI, Native Hawaiian/Pacific Islander. (From 2019 National Healthcare Quality
and Disparities Report. Rockville, MD: Agency for Healthcare Research and Quality; December 2020. AHRQ Pub. No.
20(21)-0045-EF.)
kidney transplantation. It is also not clear how to apply the race factor
when the patient’s race is unknown and/or ambiguous, as in those who
are multiracial. This disparity-inducing scenario could be avoided
through the use of cystatin C–based eGFR estimation, which has been
demonstrated to be more accurate than the CKD-EPI equation and for
which race is not required in estimation.
The application of artificial intelligence (AI) analytics to large
amounts of clinical electronic data—big data—holds the promise to
better understand health care costs, utilization, resource allocation, and
population health monitoring. Machine learning models can identify the
statistical patterns in large amounts of historically collected data. These
data naturally contain the patterning of preexisting health care disparities created by socially and historically structured inequities. This biased
patterning can lead to incorrect predictions, withholding of resources,
and worse outcomes for vulnerable populations. Recently, analysis of
a commercial, national, proprietary prediction algorithm, affecting
millions of patients, exhibited racial bias. Historical cost data were used
to predict clinical risk and allocate additional clinical services for highcost patients. Algorithmic bias arose because black patients historically
have less access to health care and thus less money is spent on their care
compared to white patients. Thus, blacks, who tended to be sicker than
white patients, received lower clinical
risk scores and thus were less likely to
receive additional clinical services. The
observed allocation bias was remedied
using direct measures of illness and
illness severity. Thus, machine learning
algorithms are not inherently free of
bias and should be assessed for accuracy
and fairness.
In summary, there are many ways
in which racism has contributed and
does and will continue to contribute to
racial and ethnic disparities in health
and health care.
■ SOCIAL DETERMINANTS
OF HEALTH
Minority Americans have poorer health
outcomes than whites from preventable
and treatable conditions such as cardiovascular disease, diabetes, asthma,
cancer, and HIV/AIDS. Multiple factors contribute to these racial and ethnic disparities in health. The landmark
National Academy of Medicine (formerly, the Institute of Medicine [IOM])
report, Unequal Treatment: Confronting
Racial and Ethnic Disparities in Health
Care, published in 2002, summarized the
scientific evidence on health disparities
and provided an important framework
for conceptualizing and defining racial/
ethnic disparities. Since the Unequal
Treatment report, there has been a growing empiric evidence base on how racism
and the SDOH, often working in synergy,
create and sustain disparities. Mechanistically, the biopsychosocial model brings
together the social and physical characteristics of the environment with individual physical and psychological attributes.
These environmental and individual
characteristics, in turn, influence health
behaviors and stress-related physiologic
pathways that directly impact health.
The National Institute on Minority
Health and Health Disparities SDOH
model builds on prior models and adds
the time element across the life course of the individual in recognition of
the long-lasting health effects of socioeconomic exposures (Fig. 10-7).
The resulting matrix has the domains of influence of health (biological,
behavioral, physical and built environment, sociocultural environment,
health care system) along the y-axis and the levels of influence on health
(individual, interpersonal, community, societal) along the x-axis. Cells
are not mutually exclusive, and examples of factors within each cell are
illustrative and not comprehensive. This framework emphasizes the
complex multidomain etiologies of disparities across the factors in the
conceptual matrix thus highlighting the limitation of individual-level
focused research and policy.
In addition to race and racism, Unequal Treatment identified a set
of root causes that included health system, provider-level, and patientlevel factors.
Health System Factors • HEALTH SYSTEM COMPLEXITY Even
among persons who are insured and educated and who have a high
degree of health literacy, navigating the U.S. health care system can be
complicated and confusing. Some individuals may be at higher risk for
receiving substandard care because of their difficulty navigating the system’s complexities. These individuals may include those from cultures
Racial and Ethnic Disparities in Health Care
63CHAPTER 10
unfamiliar with the Western model of health care delivery, those with
limited English proficiency, those with low health literacy, and those
who are mistrustful of the health care system. These individuals may
have difficulty knowing how and where to go for a referral to a specialist;
how to prepare for a procedure such as a colonoscopy; or how to follow
up on an abnormal test result such as a mammogram. Since people of
color in the United States tend to be overrepresented among the groups
listed above, the inherent complexity of navigating the health care system
has been seen as a root cause for racial/ethnic disparities in health care.
OTHER HEALTH SYSTEM FACTORS Racial/ethnic disparities are due
not only to differences in care provided within hospitals but also to
where and from whom minorities receive their care; i.e., certain specific
providers, geographic regions, or hospitals are lower-performing on
certain aspects of quality. For example, one study showed that 25% of
hospitals cared for 90% of black Medicare patients in the United States
and that these hospitals tended to have lower performance scores on
certain quality measures than other hospitals. That said, health systems
generally are not well prepared to measure, report, and intervene to
reduce disparities in care. Few hospitals or health plans stratify their
quality data by race/ethnicity or language to measure disparities,
and even fewer use data of this type to develop disparity-targeted
interventions. Similarly, despite regulations concerning the need for
professional interpreters, research demonstrates that many health care
organizations and providers fail to routinely provide this service for
patients with limited English proficiency. Despite the link between
limited English proficiency and health care quality and safety, few providers or institutions monitor performance for patients in these areas.
Provider-Level Factors • PROVIDER–PATIENT COMMUNICATION
Significant evidence highlights the impact of sociocultural factors,
race, ethnicity, and limited English proficiency on health and clinical
care. Health care professionals frequently care for diverse populations
with varied perspectives, values, beliefs, and behaviors regarding health
and well-being. The differences include variations in the recognition of
symptoms, thresholds for seeking care, comprehension of management
strategies, expectations of care (including preferences for or against
diagnostic and therapeutic procedures), and adherence to preventive
measures and medications. In addition, sociocultural differences
between patient and provider influence communication and clinical
decision-making and are especially pertinent: evidence clearly links
provider–patient communication to improved patient satisfaction,
regimen adherence, and better health outcomes (Fig. 10-8). Thus,
when sociocultural differences between patient and provider are not
appreciated, explored, understood, or communicated effectively during
the medical encounter, patient dissatisfaction, poor adherence, poorer
health outcomes, and racial/ethnic disparities in care may result.
A survey of 6722 Americans ≥18 years of age is particularly relevant
to this important link between provider–patient communication and
health outcomes. Whites, African Americans, Hispanics/Latinos, and
Asian Americans who had made a medical visit in the past 2 years
were asked whether they had trouble understanding their doctors;
whether they felt the doctors did not listen; and whether they had
medical questions they were afraid to ask. The survey found that 19%
of all patients experienced one or more of these problems, yet whites
experienced them 16% of the time as opposed to 23% of the time for
African Americans, 33% for Hispanics/Latinos, and 27% for Asian
Americans (Fig. 10-9).
Biological
Behavioral
Physical/Built
Environment
Sociocultural
Environment
Health Care
System
Health Outcomes Individual Health
Individual Interpersonal
Levels of Influence*
Domains of Influence
(Over the Lifecourse)
Community Societal
Community
Health
Population
Health
Family/
Organizational
Health
Biological Vulnerability
and Mechanisms
Caregiver–Child Interaction
Family Microbiome
Community Illness
Exposure
Herd Immunity
Sanitation
Immunization
Pathogen Exposure
Policies and Laws
Societal Structure
Community Functioning
Community Environment
Community Resources
Community Norms
Local Structural
Discrimination
Social Norms
Societal Structural
Discrimination
Quality of Care
Health Care Policies
Availability of Services
Safety Net Services
Family Functioning
School/Work Functioning
Household Environment
School/Work Environment
Social Networks
Family/Peer Norms
Interpersonal Discrimination
Patient–Clinician Relationship
Medical Decision-Making
Health Behaviors
Coping Strategies
Personal Environment
Sociodemographics
Limited English
Cultural Identity
Response to Discrimination
Insurance Coverage
Health Literacy
Treatment Preferences
FIGURE 10-7 National Institute on Minority Health and Health Disparities social determinants research framework. *
Health disparity populations: race/ethnicity, low
socioeconomic status, rural, sexual and gender minority. Other fundamental characteristics: sex and gender, disability, geographic region. (From National Institute on
Minority Health and Health Disparities. NIMHD Research Framework. 2017. Retrieved from https://www.nimhd.nih.gov/about/overview/research-framework.html.)
How do we link communication to outcomes?
Communication
Patient satisfaction
Adherence
Health outcomes
FIGURE 10-8 The link between effective communication and patient satisfaction,
adherence, and health outcomes. (Institute of Medicine. 2003. Unequal
Treatment: Confronting Racial and Ethnic Disparities in Health Care. https://doi
.org/10.17226/12875. Adapted and reproduced with permission from the National
Academy of Sciences, Courtesy of the National Academies Press, Washington, D.C.)
No comments:
Post a Comment
اكتب تعليق حول الموضوع