Racial and Ethnic Disparities in Alzheimer's Disease: A Literature Review. Evidence of Racial and Ethnic Disparities


Prevalence, Incidence, and Diagnosis
Prevalence describes the proportion of a population with a disease at a specific point in time, while incidence describes the proportion of the population that develops a disease over a specific time period. Reported Alzheimer's disease prevalence rates range from 14% to 500% higher among African Americans than among Whites (Froehlich et al., 2001). The most frequently cited estimates are that Blacks are about two times more likely than Whites to have Alzheimer's disease and Hispanics are about 1.5 times more likely than Whites to have Alzheimer's disease (Alzheimer's Association, 2011, 2014).
Prevalence studies in New Haven, Connecticut; Copiah County, Mississippi; Houston, Texas; New York, New York; the Piedmont region of North Carolina; and East Baltimore, Maryland, have all found higher rates of dementia among ethnic minority elderly--specifically African Americans and Hispanics--compared to Whites and Asians (Harwood & Ownby, 2000). Among African Americans, prevalence estimates have been as high as 56%, although rates of diagnosed disease have been more in the range of 5%-16% (Teresi et al., 2002; Gurland & Lantigua, 2002). Prevalence rates among Hispanics have been less commonly measured, but one study estimated dementia prevalence of 8% of Caribbean Latinos in the 65-74 age group, 28% in the 75-84 age group, and 63% in those aged 85 and older (Gurland et al., 1999). Another more recent study found that Mexican American men had significantly higher prevalence of cognitive impairment than did non-Hispanic White men (Samper-Ternent et al., 2012).
An analysis of 2006 Medicare claims data found that older African Americans and Hispanics were more likely than Whites to have a diagnosis of Alzheimer's disease. Rates were 14% for Hispanics, 13% for African Americans, 10% for Whites, 9% for Native Americans, and 8% for Asians (Alzheimer's Association, 2011). The report's authors cautioned that prevalence rates based on diagnosis codes may reflect varying levels of underdiagnosis across populations. They also pointed out that the Health and Retirement Study has found that among those with cognitive impairment, 46% of Whites over age 55 had been told by a physician that they had a memory-related disease, compared to 34% of Hispanics and 34% of African Americans.
In terms of the prevalence of cogntive impairment, the 2006 Health and Retirement Study shows a marked difference by age and race/ethnicity, as shown in Figure
1. Overall prevalence is about 105 per 1,000 Americans aged 55 or over. African Americans are approximately 2-3 times more likely than Whites to have cognitive impairment, and these differences are greater in the younger age groups (Alzheimer's Association, 2011).
FIGURE 1. Prevalence of Cognitive Impairment among Americans Aged 55 and Older (rate per 1,000 population) by Age and Race/Ethnicity, 2006 Health and Retirement Study 
FIGURE 1. Prevalence of Cognitive Impairment among Americans Aged 55 and Older (rate per 1,000 population) by Age and Race/Ethnicity, 2006 Health and Retirement Study
SOURCE: Adapted from Alzheimer's Association, 2010.

Incidence estimates also suggest higher risk of Alzheimer's disease among non-White populations. In a 7-year study in the Washington Heights and Inwood communities of New York City, overall adjusted incidence rates for probable and possible Alzheimer's disease (excluding vascular and other dementias) among Whites were 0.4% per person-year for ages 65-74, 2.6% for ages 75-84, and 4.2% for ages 85+ (Tang et al., 2001). Incidence was higher among African Americans (1.7%, 4.4%, and 11.4%) and Caribbean Hispanics (1.4%, 4.4%, and 8.8%) in the same community. One recent study found that the unadjusted hazard ratio for developing dementia over 12 years was 1.44 for Black participants compared to White, but after adjusting for demographics, apolipoprotein (APOE) E4, comorbidities, lifestyle factors, and socioeconomic status, the Black-White difference was reduced to 1.09 and was no longer significant (Yaffe et al., 2013).

Health Outcomes
Mortality and other health outcomes among people with Alzheimer's disease vary by race and ethnicity. Despite a lower prevalence and incidence rate, Whites have a higher overall mortality rate from Alzheimer's disease (Gillum & Obisesan, 2011; Mehta et al., 2008). In a recent study, Whites had dementia listed as the underlying or contributing cause of death on their death certificates more often than Blacks during 1999-2004: the age-adjusted rate per 100,000 was 647 for Whites versus 628 for Blacks (Gillum & Obisesan, 2011). Rates were higher among Whites than Blacks in New England, the Mid-Atlantic, and in the East and West South Central census regions. In the East and West North Central, South Atlantic, Mountain, and Pacific regions, rates were higher among Blacks than Whites. African Americans and Latinos with Alzheimer's disease had a lower adjusted risk of mortality than their White counterparts. Asians and American Indians had similar mortality risk as Whites (Mehta et al., 2008).
In another study, Blacks and Whites born in the so-called stroke belt states (Alabama, Arkansas, Georgia, Mississippi, North Carolina, South Carolina, and Tennessee) were at higher age-adjusted and sex-adjusted risk of dementia mortality--29% higher for Blacks and 19% higher for Whites--than those born in other states, even among those who did not live in those states at death (Glymour et al., 2011).
In a large epidemiological study aimed at identifying the prevalence of specific behaviors among patients with dementia, after adjusting for age, sex, education, income, Mini-Mental Status Examination (MMSE) score, activities of daily living, and caregiver characteristics, Blacks were at increased risk (relative to Whites) of being constantly talkative, having hallucinations, expressing unreasonable anger, wandering, and waking the caregiver. Hispanics were at increased risk (relative to Whites) of having hallucinations, expressing unreasonable anger, being combative, and wandering. Whites were at increased risk of being paranoid relative to Hispanics (Sink et al., 2004). Older African Americans with Alzheimer's disease consistently describe themselves as more impaired, with greater need for health services, than Whites--despite a lack of measured differences in the number of recent sick days, number of reported chronic conditions, cognitive status, and mortality (Froehlich et al., 2001). Whether this difference is a result of unmeasured comorbidities or different types of comorbidities, cultural differences in the perception and description of cognitive disability, or some other factor, is not known.
Participation in Clinical Trials
Minorities are underrepresented in clinical trials and other research studies (Cooper et al., 2010; Wendler et al., 2006). Some researchers attribute this underrepresentation to a higher level of comorbidity among minorities (disqualifying them from participating), as well as language requirements (many clinical trials require English fluency) (Cooper et al., 2010). Others believe that past abuses, such as the Tuskegee Syphilis Study, have soured the relationship between researchers and racial/ethnic minorities (Wendler et al., 2006).
A 2006 review of the literature investigated whether the underrepresentation of minorities in clinical research was related to a lower consent rate, and found that the differences between racial groups were small and non-significant, with no clear directionality or patterns (Wendler et al., 2006). The authors conclude that willingness to enroll in research studies is more a function of the characteristics of individual studies than a function of racial or ethnic attitudes as a whole. Examples of factors other than willingness to participate that may be important in determining multi-ethnic participation include the choice of study site and proximity of research locations to places where minorities live and work; recruitment disparities, such as differences in knowledge about trials and invitations to participate; language barriers; and other barriers to participation, such as child care or elder care and travel expenses.
A qualitative study to investigate barriers to participation in clinical research among Latinos found that they often do not meet clinical trial eligibility guidelines because of underdiagnosis, since many studies require participants to have a physician diagnosis of Alzheimer's disease to participate. Underdiagnosis stemmed from several factors, including a perception of Alzheimer's disease-related symptoms as natural part of aging, barriers to access, and a lack of information about Alzheimer's disease and available services. In addition, many caregivers may also be overburdened to the point that participating in a research study would be too onerous (Gelman, 2010).
Use of Long-Term Services and Supports
Approximately 40% of persons who died in a nursing home between 1999 and 2006 had dementia (22% with mild to moderately severe dementia and 20% with advanced dementia) (Miller et al., 2010). Historically, disparities in access to nursing homes and other formal long-term care services contributed to the lower rates of use of such services by elderly members of minority groups, compared to elderly Whites (Belgrave et al., 1993). However, as nursing home occupancy rates have declined in most markets, and assisted living and other home and community-based options have proliferated, access to nursing home care may be less of a problem for elderly minorities (Feng et al., 2011). In both 1999 and 2008, rates of nursing home residency were highest among Blacks, followed by Whites, Hispanics, and Asians (Figure 2). Between 1999 and 2008, the absolute number of elderly Hispanics and Asians living in United States nursing homes grew by 55% and 54%, respectively, while the number of elderly Black residents increased 11% (Feng et al., 2011). During the same period, the number of White nursing home residents declined by 10%. However, on a population basis, the rates per 1,000 persons declined in all four groups (Figure 2). A meta-analysis of seven studies showed that minorities were 40% less likely to enter a long-term care facility than Whites in the United States. Insurance status, poverty, level of impairment (activities of daily living, instrumental activities of daily living, gait, etc.) and social support/caregiver willingness to provide care are the main determinants of institutionalization (Cooper et al., 2010).
Although use of nursing homes by minorities has grown in recent years, there is evidence that minorities tend to live in lower quality facilities (Smith et al., 2007). For example, a national study of 516,082 nursing home residents found that residents in facilities with a higher proportion of African Americans had higher odds of being hospitalized, a possible marker of poor-quality (Gruneir et al., 2008).
A meta-analysis of seven studies found that the use of community social services did not differ by race/ethnicity after controlling for dementia severity and socioeconomic status (Cooper et al., 2010). On the other hand, in their study of respite care provided through the Alzheimer's Disease Demonstration Grants to States program, Montgomery et al. (2001) found that use patterns differed by racial and ethnic groups. For example, Blacks tended to use small amounts of care for a long period of time, while Hispanics used a lot of care for a short period of time. Moreover, in this study, respite staff working with Blacks and Hispanics found a strong sense of self-sufficiency and general distrust of government services.
FIGURE 2. Nursing Home Residents per 1,000 Over Age 65 by Race/Ethnicity, 1999 and 2008
FIGURE 2. Nursing Home Residents per 1,000 Over Age 65 by Race/Ethnicity, 1999 and 2008
SOURCE: Adapted from Feng et al., 2011.

A recent systematic review of 20 studies on ethnic and racial disparities at the end of life among people with dementia found that attitudes toward end-of-life care were similar in different ethnic groups, and there was conflicting evidence on hospice usage. Artificial nutrition and other life-sustaining treatments (such as ICU care and mechanical ventilation) were more frequently used, and decisions to withhold treatment less common, in African Americans and Asians (Connolly et al., 2012).
Use of Medications
Multiple studies have shown that Whites are generally more likely to be taking acetylcholinesterase inhibitors (AChEIs) and other dementia medications than are non-Whites. A meta-analysis and literature review by Cooper and colleagues (2010) that pooled the results of three studies found that African Americans were 30% less likely to be prescribed AChEIs than Whites, whereas Hispanics had a similar likelihood of prescriptions as Whites. Three other studies of the use of medications by race/ethnicity that were not included in the above meta-analysis also found that African Americans were less likely to use dementia medications (Gruber-Baldini et al., 2007; Hernandez et al., 2010; McClendon et al., 2009). One analysis of data using the 2000-2002 Medicare Current Beneficiary Survey found the opposite: community-dwelling African American and Hispanic Medicare beneficiaries were more likely to be prescribed Aricept than were Whites (Perryman et al., 2009). In an analysis of four state Medicaid populations, Gilligan and colleagues found significant disparities in medication (AChEIs and memantine) use between Hispanics and non-Hispanic Whites (in favor of Hispanics) in Florida, between non-Hispanic Blacks and Hispanics (in favor of Hispanics) in California and Florida, between non-Hispanic Blacks and non-Hispanic others (in favor of non-Hispanic others) in California and New York, and between Hispanics and non-Hispanic others (in favor of non-Hispanic others) in California and New York (Gilligan et al., 2012).
Health Care Expenditures
Published literature on disparities in health and long-term services and supports expenditures is very sparse. Only two peer-reviewed studies were found, one of which used data from the early 1990s, and one used Medicaid data from 2004. In the first, Blacks with vascular dementia in Tennessee had higher total expenditures than did Whites with the same diagnosis ($17,359 vs. $12,904) (Husaini et al., 2003). In the second study, Blacks had significantly higher expenditures than Whites, Hispanics, and those of other racial/ethnic backgrounds, and Hispanics and Whites had significantly higher expenditures than others (Gilligan et al., 

TABLE 1. Use of and Medicare Payments for Health Care Services among Medicare Beneficiaries with a Dementia Diagnosis, by Race/Ethnicity, 2006  

  Overall White African American Hispanic 
Total mean Medicare payment per beneficiary $15,333 $14,498 $21,044 $19,933
Mean Medicare payment for hospital care per beneficiary $4,964 $4,563 $7,687 $6,632
Mean hospital discharges per 1,000 beneficiaries 660 632 887 731
Mean number of hospital days per beneficiary 4.2 3.9 6.4 5.2
Mean Medicare payment for physician visits per beneficiary $1,018 $956 $1,390 $1,411
Mean number of physician visits per beneficiary 12.7 12.1 16.4 17.1
Mean Medicare payment for home health care per beneficiary $1,118 $1,025 $1,591 $2,453
Percentage of beneficiaries with at least one home health claim  18.8% 18.2% 22.1% 25.9%
Mean Medicare payment for hospice per beneficiary $1,732 $1,789 $1,514 $1,225
Percentage of beneficiaries with at least one hospice claim 13.7% 14.2% 11.3% 8.8%

SOURCE: Adapted from Alzheimer's Association, 2011.

In a study funded by the Alzheimer's Association, total Medicare costs were substantially (45% and 37%) higher for older African Americans and Hispanics diagnosed with Alzheimer's disease than for Whites with a similar diagnosis (Table 1). These estimates were not adjusted, and therefore may be biased because of group-level differences in age, sex, and comorbidity burden.
In a 2010 survey conducted on behalf of the Alzheimer's Association, 70% of caregivers were White, 15% were African American, 12% were Hispanic, 1% were Asian American, and 2% were from other ethnic groups (Alzheimer's Association, 2012). Compared to caregivers of other races, African American and Hispanic caregivers were more likely to be single/never married, and more likely to have incomes less than $50,000 per year. Asian American caregivers were as likely to be male as female, in contrast to caregivers of other races, who were predominantly female. Asian American caregivers were more likely to be highly educated.
The caregiver role is often assumed by an adult child, friend, or other family member in Black families and by an adult daughter in Hispanic families (Janevic & Connell, 2001). The concept of the "primary caregiver" may not be equally applicable in all cultures; therefore, interventions that are aimed at the family, rather than just the primary caregiver, may be more culturally appropriate for these cultures (Janevic & Connell, 2001). Korean and Korean American caregivers are less likely to be spouses and more likely to be another family member (Janevic & Connell, 2001). The tradition in cultures with Confucian values is for the daughter-in-law of the eldest son to provide care.
Compared with non-Hispanic Whites, Mexican American caregivers of persons with dementia report greater distress, poorer self-rated health, more somatic complaints, and increased levels of sensitivity to patient problem behaviors (Harwood & Ownby, 2000). Similarly, depression is more common in Hispanic family caregivers than in Whites (Janevic & Connell, 2001).
Some studies of caregiving interventions for people with dementia suggest that African Americans find some aspects of caregiving less stressful than Whites and get more emotional benefit from the experience (Janevic & Connell, 2001). For example, in the Resources for Enhancing Alzheimer's Caregiver Health (REACH) study, African Americans had higher scores than Whites on the Positive Aspects of Caregiving scale, and lower anxiety and fewer feelings of bother by the care recipient's behavior (Roff et al., 2004). The fact that caregiving is more expected in some cultures may be protective against stress. Baseline differences between groups in psychological distress may account for differences in the impact of Alzheimer's disease on caregivers (Janevic & Connell, 2001).

View full report


"RacEthDis.pdf" (pdf, 361.48Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®