Systematic Review and Meta-Analysis of Racial and Ethnic Differences in Dementia Caregivers' Well-Being.

ABSTRACT

BACKGROUND AND OBJECTIVES: Studies comparing racial/ethnic differences on measures of psychological and physical well-being for dementia caregivers have reported differences between minority and white caregivers. Recruitment methods often differ for minority and white participants due to enrollment targets and may lead to biased comparisons, especially in convenience samples. We aimed to examine racial/ethnic differences in dementia caregiver outcomes and to determine whether differences vary between studies with population-based or convenience samples. RESEARCH DESIGN AND METHODS: We systematically reviewed articles with primary data from PubMed, Google Scholar, and PsycINFO. We included studies comparing African American or Hispanic/Latino to white dementia caregivers on measures of psychological well-being or physical well-being. Reviewers screened titles and abstracts, reviewed full texts and conducted risk-of-bias assessments. Meta-analyses were conducted to assess effects by race/ethnicity and study bias. RESULTS: A total of 159 effects were extracted from 38 studies, 2 of which were population based. Random-effects models revealed small but statistically significant effects with better psychological well-being in African American caregivers compared with white caregivers in both population-based (d = -0.22) and convenience sample studies (d = -0.21). Hispanics/Latino caregivers reported lower levels of physical well-being than white caregivers (d = 0.12), though these effects varied by level of rated study bias. DISCUSSION AND IMPLICATIONS Consistency across study methods raises confidence in the validity of previous reports of better psychological well-being in African American caregivers. Future studies should use population-based samples with subgroups of Hispanic/Latino, Asian American, and American Indian caregivers that are culturally distinct on factors such as country of origin and tribe.