摘要：Objectives. We examined the association between Black ethnic density and depressive symptoms among African Americans. We sought to ascertain whether a threshold exists in the association between Black ethnic density and an important mental health outcome, and to identify differential effects of this association across social, economic, and demographic subpopulations. Methods. We analyzed the African American sample (n = 3570) from the National Survey of American Life, which we geocoded to the 2000 US Census. We determined the threshold with a multivariable regression spline model. We examined differential effects of ethnic density with random-effects multilevel linear regressions stratified by sociodemographic characteristics. Results. The protective association between Black ethnic density and depressive symptoms changed direction, becoming a detrimental effect, when ethnic density reached 85%. Black ethnic density was protective for lower socioeconomic positions and detrimental for the better-off categories. The masking effects of area deprivation were stronger in the highest levels of Black ethnic density. Conclusions. Addressing racism, racial discrimination, economic deprivation, and poor services—the main drivers differentiating ethnic density from residential segregation—will help to ensure that the racial/ethnic composition of a neighborhood is not a risk factor for poor mental health. Recent years have seen an increase in the number of studies examining the association between the residential concentration of racial/ethnic minorities (ethnic density) and health, with increasingly sophisticated statistical techniques and theoretical frameworks helping to identify the relevance of ethnic density effects. Despite these improvements, the association between ethnic density and health, given the concentration of poverty in areas of higher ethnic density, is still a puzzling phenomenon. The literature is characterized by inconclusive findings in both the direction and the size of ethnic density effects. Reviews have asserted that ethnic density effects are stronger for mental health 1 than for physical health, mortality, and health behaviors, 2 but even among the latter set of outcomes, protective ethnic density effects are more common than adverse associations. 2 One common finding among ethnic density studies, regardless of health outcome, is the variation in results across and within racial/ethnic groups. For example, US studies often report protective associations among Latinos but mostly detrimental associations for African Americans, 2 and the few studies that have examined subgroups among broad “US Black” ethnic categories have found differences by age, 3 gender, 4,5 and nativity. 6,7 Determining the specific populations for which ethnic density effects are protective or detrimental can help in achieving a greater understanding of the potential mechanisms by which ethnic density is associated with health. Another methodological improvement that would clarify the association between ethnic density and health is adequate adjustment for area resource deprivation. The positive correlation that exists between ethnic density and deprivation, and the established association between area deprivation and poor health, 8 may have a twofold effect in concealing ethnic density effects: first, by overriding protective effects of ethnic density; second, by complicating analytical attempts at disentangling harmful deprivation effects from protective ethnic density benefits, even with the use of multilevel methods. Reviews of the literature have highlighted the inadequate adjustment for area deprivation as one of the main limitations in current studies, most of which control for only 1 measure of area deprivation (e.g., median income) or, in some cases, do not adjust for any relevant confounders. 1,2 Although the appropriate adjustment for area deprivation is critical for detecting ethnic density effects, it is not sufficient. To properly capture the associations between ethnic density, area resource deprivation, and health, the potential suppressing effects of area deprivation in the association between ethnic density and health should be modeled. Detrimental ethnic density effects may not be due to the concentration of ethnic minorities in an area but to the concurrent concentration of poverty and social adversity, 9 and appropriate modeling can portray the relative contribution of ethnic density and area deprivation to health. In addition to differentiating ethnic density effects between subgroups and accurately modeling and adjusting for area deprivation, the possible nonlinearity in the association between ethnic density and health, and the potential thresholds at which ethnic density exerts protective or nonprotective effects on health, need to be addressed. 10 The combination of methods and theoretical frameworks aiming to understand the importance of concentrated poverty and threshold effects for ethnic density might also be useful in clarifying the difference between ethnic density and residential segregation. Although ethnic density is framed in terms of social support, racial/ethnic diversity, and a stronger sense of community, residential segregation is a direct consequence of current and historical racism and discrimination, and is recognized as a determinant of racial/ethnic health inequalities. 11 However, both ethnic density and residential segregation are conceptualized through use of a measure of racial/ethnic residential concentration, and it is unclear at what point the hypothesized protective benefits of ethnic density are overcome by the pernicious effects of racial residential segregation. Understanding this difference and its drivers has important implications for social and public health policy, as it would allow the promotion of factors that harness the protective effects of ethnic density while targeting the factors related to racial residential segregation. We examined the association between Black ethnic density and depressive symptoms among African Americans in the National Survey of American Life (NSAL), to ascertain (1) the differential effects of ethnic density across subgroups of African American NSAL respondents and (2) the protective or detrimental thresholds of Black ethnic density. We addressed these 2 study aims while accounting for, and adequately modeling, the potential suppressing effects of area resource deprivation on the association between ethnic density and an important mental health outcome. We selected depressive symptoms as the mental health outcome in this study because the literature is consistent regarding the ethnic density effects of outcomes such as psychoses, but not about the association between Black ethnic density and depression. 5 Although the prevalence of major depression is lower among African Americans than among the White majority, the prevalence of depressive symptoms and chronic low mood is high among this population, 12–14 and understanding any protective or risk factors of psychological distress, including at the neighborhood level, remains a priority. We focused on African Americans because most ethnic density studies have been conducted in this population, and it is the group in which ethnic density effects have been found to be the most detrimental. 1,2 Some previous studies have modeled nonlinear associations between Black ethnic density and several physical health indicators, 15–18 but not mental health ones. In all of these prior studies, potential cutoff points have not been based on formal threshold examinations. In addition, prior studies have analyzed non-Hispanic Black respondents. In this study, we focused specifically on African Americans because of the documented heterogeneity of sociodemographic characteristics, 19 health profiles, 20–22 and ethnic density effects 23 in the non-Hispanic Black population.
