DEMENTIA & ETHNICITY IN THE U.S.: PREVALENCE AND INCIDENCE IN ETHNIC AND RACIAL POPULATIONS Gwen Yeo, PhD, AGSF Updates on Dementia Conference May 11, 2017 Systematic Review of Dementia Prevalence and Incidence of Dementiain United States Race/Ethnic Populations Kala Mehta and Gwen Yeo Alzheimer s & Dementia January 2017 1
How Common is a Disease in a Population? Two Measures Prevalence Number or percent of the population diagnosed with the disease at one point in time Gives an idea of the burden of the disease Incidence Number or rate of new cases of a disease during a specific time period. Gives an idea of the rate of change over time Can compare time periods or populations Search Data Base Analysis 1229 studies identified through PubMed reviewed 114 with appropriate recruitment and diagnostic methods were included in database Comparisons very difficult Different ages, methodologies, reporting styles 2
What Do We Know About Dementia Prevalence in Race/Ethnic Populations? African Americans (10 of the 19 studies) ~ ages of samples from 40+ to 100+ ~ prevalence ranges from 0.01% to 68% ~ four reported age 65: 7%, 8%, 16%, 22% ~ four reported age 85: 18%, 23%, 32%, 59% In the 8 studies who also had a non-latino white comparison group, all but one showed lower rates for whites. Latino/Hispanic Prevalence Mexican American (SALSA) 60: 5%; 85: 31% Puerto Rican (PR Veterans) 65: 13% Cuban American (women) 65: 13% Caribbean Hispanic (No. Manhattan Dominican, PR, & Cuban) 65-74: 8% 75-84: 28% 85: 63%. 3
Asian American Prevalence Japanese Americans Honolulu Heart Study/HAAS Men 65: 8%; 85: 33% Kame Study (Seattle area) 65: 6% Korean American MASK-MD (screening data only) 60: 7% NO DATA AVAILABLE ON OTHER ASIAN AMERICAN POPULATIONS American Indian Prevalence NO RELIABLE REPORTS AMONG AMERICAN INDIAN POPULATIONS Pacific Islander Prevalence Chamorros (indigenous population of Guam) 65: 12% (9% Guam Dementia) 4
AVERAGE ANNUAL INCIDENCE BY RACE/ETHNICITY 0 0.1 0.2 0.3 0.4 0.5 0.6 African American (Evans, 2003) African American (Gao, 2011) African American Average 0.26 African American (Muller, 2007) African American (Shadlen, 2006 African American(Gurland, 1999) African American(Luschinger, 2001) Ashkenazi Jewish (Sanders, 2010) Caribbean Hispanic (Louis, 2010) Caribbean Hispanic (Muller, 2007) Japanese American (Foley, 2001) Japanese American (Havlik 2001) Mexican American-Haan 2007 Japanese American Average 0.19 Mexican American Average 0.08 Caribbean Hispanic Average 0.40 White, Non Latino (Evans, 2003) White, Non Latino (Gurland, 1999) White Average 0.17 White, Non Latino (Luschinger, 2001) White, Non Latino (Shadlen, 2006) White, Non Latino,men only (Perkins, 1997) In follow up analysis, age and education had the strongest association with risk of dementia. Effects of Ethnicity & Education North Manhattan Study When differences in age 30 and education were controlled, there were 20 NO differences between 10 the ethnic groups in rates of dementia. 0 60 50 40 Prevalence Hisp. Af. Am. NH White Gurland et al., 1997 5
What Did We Learn? There are major gaps in the evidence for rates of prevalence or incidence among different race/ethnic populations. American Indians Most Asian American populations Chinese, Filipino, Asian Indian, Vietnamese Smaller Asian groups Most Pacific Islander populations What Did We Learn? It is impossible to compare results of studies fairly within or across population groups because of differences in: Age of subjects Recruitment and inclusion strategies Diagnostic methods How results are reported (by age or gender categories only) Result: Disparities cannot accurately be documented 6
What Did We Learn? It is important to disaggregate data on ethnic populations within the large race/ethnic minority categories used by the Census. For example, the lowest and highest rates are found among populations classified as Hispanic/Latino. If lumped together, the results are misleading. Recommendations to Increase our Knowledge about Disparities Develop standardized protocol for race/ethnic epidemiological research Prioritize funding for ethnic specific populations with little or no data Include ethnic community members in research teams 7
CLINICAL IMPLICATIONS LANGUAGE, EDUCATION, & LITERACY COMORBIDITY CAREGIVING Appropriate Assessment Use of Interpreters Diabetes Hypertension, Stroke Support Programs Caregiver Health NEXT STEPS KNOWLEDGE OF RISK FACTORS INTERVENTIONS DECREASE IN DISPARITIES 8
Thank You gwenyeo@stanford.edu 9