Despite experiencing a wide range of stressors and economic and social adversities, non-Hispanic Blacks (NHBs) have a lower risk of depression than non-Hispanic Whites (NHWs) [ 1,2,3,4,5,6,7,8,9,10,11,12,13]. This paradox is one of the unexplained mysteries of research in the field of race and mental health in the United States [ 4,5,6,7,814]. Some researchers attribute this phenomenon to the fact that NHBs have a greater tendency to experience flourishing, psychological growth, and transcendence when faced with social and economic adversities than NHWs [ 4].

Another possible explanation for the Black-White mental health paradox, known as the Differential Effects Hypothesis [ 15,16], is that NHBs and NHWs differently gain or lose health due to the risk factors and resources in their social environment [ 16,17]. In this view, NHBs show a weaker slope of the effects of socioeconomic status (SES) resources than NHWs [ 18,19,20,21,22]. Similarly, compared to NHWs, NHBs tend to rely on non-economic resources such as social support, family, and religion [ 23]. Social support, family, and religion have been freely available to NHBs, even under slavery. Thus, religion and non-religious social support have become central elements of survival for NHBs [ 24]. In this view, NHBs not only report higher levels of religious and non-religious social interaction than NHWs, they also more effectively mobilize their social relations both inside and outside of church to better cope with harsh environments [ 25,26].

Cohen’s social support theory [ 40] has emphasized that positive and supportive social relations are protective on health. Social support is one the core elements of maintaining mental health, as it buffers the effects of stress [ 27,28]. There are many empirical [ 29,30,31] and theoretical [ 28] studies on the protective effects of positive social relations. For example, Krause [ 32,33,34], and Taylor and Chatters [ 35,36,37,38,39] have shown that social support is a main mechanism that NHBs use to maintain their mental health. However, many of these effects may not be equal for NHBs and NHWs [ 25,26,41,42].

An explanation for the Black-White mental health paradox is the higher availability and salience of positive social resources in the life of NHBs compared to NHWs [ 4,5,6,7,8]. Mouzen et al. have used the National Survey of American Life data and examined whether social relations in the family and church explain the lower prevalence of depression in NHBs than NHWs [ 7,8]. Their research could not provide evidence supporting the hypothesis that differences in the quality and the quantity of social relations (friends, fictive kinship, relatives, and family members) in the lives of NHBs and NHWs can explain the Black-White mental health paradox [ 7,8].

In several studies, Assari has also tried to understand the Black-White mental health paradox [ 5,43,44,45,46,47,48,49]. Various studies have documented differences between NHBs and NHWs in the psychosocial and medical correlates of depression [ 5,43,44,45,46,47,48,49]. This includes evidence suggesting that depression predicts chronic medical conditions [ 6,52,53] and mortality [ 50,51] of NHWs but not NHBs. Similarly, baseline depressive symptoms predicted subsequent depression for NHWs but not NHBs [ 54]. This weaker predictive role of risk factors on psychosocial and physical health outcomes in NHBs compared to NHWs are not limited to depression and shown for several other economic, social, and psychological resources [ 6,18,40,50,52,53,54,55,56,57]. These patterns can be either seen as a systemic resilience or diminished returns of resources in NHBs [ 4,5].

One potential hypothesis that may justify the weaker-than expected effects of social and medical risk factors in the NHB community is higher availability of social support and religiosity [ 7,8,25,41]. Social support and religiosity are both more common and more effective among NHBs than NHWs [ 41,42]. In other terms, NHBs might be more resilient to the harmful effects of societal adversities compared to NHWs, simply because social relations and religion are more salient for them. Assari has conducted a few studies showing that social support may have a more salient role in protecting NHB than NHW mental health [ 25,118]. In another study, Krause showed that church-based support shows stress-buffering effects for NHBs but not for NHWs [ 42]. Similarly, Lincoln, Chatters and Taylor used data from the National Comorbidity Survey (NCS) and showed that social support and lack of it are more closely associated with psychological distress in NHBs compared to NHWs [ 41]. Again, these findings are indicative of NHBs’ relative advantage to NHWs in receiving some mental health effects of social relations [ 26].

This study compared NHB and NHW adults in California for the effects of losing a loved one on depression. In line with previous studies on the Black-White mental health paradox [ 25,41,42], we expected a weaker effect of losing a loved one on depression in NHB than in NHW California adults. However, since social support has greater salience for the mental health of NHBs than for NHWs, loss of a loved one may exact a greater psychological toll on NHBs than on NHWs.

The Survey of California Adults on Serious Illness and End-of-Life is a cross-sectional representative survey of residents of California in 2019. The survey was conducted on the field June 6, 2019 through July 2, 2019. The study was conducted for the California Health Care Foundation. The survey was administered primarily using Ipsos’ KnowledgePanel.

This study included 2,588 California adults 18 years of age and older. The eligibility criteria were being a California resident and having an age of 18 or older. Participants were drawn from Ipsos’ KnowledgePanel, the first representative online research panel in the US. Panel members are randomly recruited through address-based sampling (ABS) methods. All households are provided with Internet access as well as hardware that might be needed. The survey also increased the number of Black respondents by using supplemental, nonprobability sampling.

Overall, the sample for this survey was designed to target the following numbers of respondents: 1) 800 Californians under 150% federal poverty level (FPL), 2) 800 Californians between 150% and 399% FPL, and 3) 800 Californians at 400% FPL and above. After the survey data were collected, cleaned, and processed, design weights were calculated to account for nonresponse as well as stratification. We applied weights to generate results that would be representative of California adults.

The survey was conducted in English and in Spanish. Participants answered the questions of the survey in their home online.

Loss of a loved one. The study measured loss of a loved one over the past two months using the following single item measure: “In the past two years, have you lost a loved one - that is, a member of your family or a close personal friend - who passed away?”. Responses were 0 (no) and 1 (yes).

Self-Rated Health (SRH). SRH was measured using a single item [ 74,75,76,77]. Participants reported their overall health. Response options included poor. The IOM, the Institute of Medicine, has recommended monitoring the health of the US general population by using single item SRH measures. As is common in the literature, we treated SRH as a dichotomous variable. To accomplish this, we combined the poor and fair categories, and compared it to other responses. Poor or fair SRH was coded as 1. Excellent, very good or good health were coded as 0. Poor SRH is highly valid, as it independently predicts the risk of mortality [ 74,75,76,77].

Race. Race and ethnicity, self-identified: [NHBs = 1, NHWs = 0 (reference category)].

Sociodemographic Factors. Sociodemographic control variables included gender, age, education, poverty status, household head, marital status, and employment. Gender was a dichotomous measure [male =1, female = 0 (reference group)]. Age was a continuous variable. Education was a four-level categorical variable: 1) <12 years, 2) 12 years, 3) 13-15 years, and 4) 16+ years. Poverty status was measured based on the income per household member. This variable was a three-level categorical variable: 1) under 150% of the federal poverty line, 2) 150% - 399% of the federal poverty line, and 3) 400% + of the federal poverty line. A dichotomous variable measured if the individual is the household head (0 = no, 1=yes). Marital status, a dichotomous variable, was coded as married = 1, other = 0. Employment was coded as employed = 1, unemployed, not in labor market, searching for job, retired, or disabled = 0. Employed included those who conducted any work for pay or those who were self-employed.

We applied SPSS 23.0 (IBM, New York, U.S.A.) to perform our data analysis. For descriptive statistics, we reported means and proportions (frequencies). For bivariate analysis, we used the Chi Square or independent sample t test. For multivariable models, we ran four logistic regression models. In all of these models, loss of a loved one was the independent variable, and self-reported depression was the dependent variable. Demographics, education, poverty status, household head, marital status, and employment were the control variables, and race was the focal moderator. First, we ran logistic regression models in the total sample that included NHBs and NHWs. The initial model did not include race by loss interaction term. In the next step, we ran a model with race by loss interaction term. Then, in two separate models, we ran race/specific models in NHWs and NHBs. Adjusted Odds Ratios (ORs), 95% Confidence Intervals (9% CI), and p-values were reported. p-values equal or less than 0.05 was significant.

Table 1 summarizes the descriptive statistics of the participants overall. Overall, 1603 people entered our study. We compared 901 (56.2%) NHB and 702 (43.8%) NHW adults (age 18 and older). Compared to NHWs, NHBs were younger, were more likely to be women, had lower education, had lower income, were less likely to be married and employed, and had higher prevalence of losing a loved one.

Table 2 provides a summary of the results of our binary logistic regression models that were conducted in the total sample in which loss of a loved one was the independent variable and self-reported depression was the dependent variable. Loss of a loved one was not associated with depression in the pooled sample. We found an interaction between race and loss of a loved one on depression. This finding suggested that the effect of loss of a loved one on depression was larger for NHBs than for NHWs (Table 2).

Table 3 provides a summary of the results of our binary logistic regression models that were conducted in NHWs and NHBs. In these models, loss of a loved one was the independent variable and self-reported depression was the dependent variable. Loss of a loved one was not associated with depression in NHWs. We found an association between loss of a loved one and depression in NHBs (Table 3).

We found that NHBs and NHWs differ in the effects of losing a loved one on depression. To be more specific, NHBs showed an association between losing a loved one and depression, an association which could not be found in NHWs.

Although the literature says loss of a loved one is a risk factor for depression [ 119,120,121], we did not see any association between losing a loved one and depression in the pooled sample and NHWs. Thus, our finding was not in line with the existing literature on the association between stress and depression. For NHBs, however, we did find an association between losing a loved one and depression. Within NHBs, this link may be explained via many mechanisms. First, loss of a loved one means a decline in social support and social relations. Thus, it operates as a source of stress. Loved ones may provide support that has direct protective effects or that buffers the effects of stress on health. A decline in social support via loss of a loved one is a predictor of an increase in perceived stress [ 69].

Our result on the interaction of race and loss of a loved one on depression was indicative of the greater salience for NHBs of social relations for mental health than for NHWs. Assari et al., [ 25,118], Lincoln et al., [ 41] and Krause [ 70] have shown that race alters the mental health benefits associated with social support both inside and outside of church. Most of these studies have shown that NHBs gain greater health from their social relations and religion than do NHWs [ 70]. As Skarupski has argued, this is indicative of an advantage of NHBs relative to NHWs [ 26].

Compared to NHWs, NHBs have different transactions with their social network [ 71,72,73,74]. This may be in part because NHBs and NHWs have differently composed social networks [ 75,74]. It is likely that in response to discrimination and racism, NHBs have turned to social relations and other organizations such as church to cope with adversity. As a result, the social relations-mental health link may vary between NHBs and NHWs [ 25,41]. Such historical adaptations may have resulted in more effective social relations required for dealing with adversity [ 2,78].

Another reason NHBs and NHWs may differ in the salience of loss of a loved one is self-construal. NHBs, like many other racial and ethnic minority groups, value social relations such as friends and family more than NHWs. Thus, NHBs may be more likely than NHWs to cultivate deep social relationships. This may be because they place a higher value on the family, collectivism, and interdependence in life [ 122,123]. NHBs are more likely than NHWs to define life and being as something interconnected, which is a part of their cultural beliefs. Such attitudes and values may result in NHBs placing a higher importance on social relations and social support [ 80]. Difference in the self-construal may result in a difference between NHWs and NHBs in how they approach their social environment and how much a loved one matters [ 81]. NHB culture is interdependent and communal, and highly values harmony via the establishment of social support networks [ 82,83]. As a result, compared to NHWs, NHBs may place a higher value on collectivistic attitudes, which includes a positive relationship with a loved one [ 81]. There are a few studies comparing NHBs and NHWs on their levels of social interaction and the salience of others [ 81,82,84]. Compared to NHW cultures, NHB culture is more communal or interdependent, which sees others as a part of the self. NHWs&#x02019; culture, in contrast, endorses independent attitudes. Such world-view values uniqueness and distinction of self from others even from significant and loved ones [ 81].

In a recent study by Assari, the effects of religious and secular social support on depression differed between NHWs and NHBs, with NHBs benefiting more from social support, across all sources, compared to NHWs [ 118]. Similar patterns were shown before by Assari [ 25], Krause [ 42], and Lincoln [ 41]. All these studies support the results of our study on the more significant role of social relations for NHBs&#x02019; than NHWs&#x02019; depression.

We found that NHWs and NHBs differ in the effect of loss of a loved one on depression. NHWs and NHBs differ in many health effects of risk factors [ 18,53,101,102]. Similarly, race alters the links between obesity and overweight [ 43,44,104,105], perceived health [ 6,45,50], chronic illness [ 14,46,52] and mortality [ 50,51] with depression. This suggests that race may also alter sensitivity to a wide range of social risks and protective factors [ 50,55,56,57]. Race interacts with a wide range of risk and protective factors in shaping mental health [ 18,45,48,104,108,109,110,111,112]. Thus, racial differences in health are not the sum of race and risk factors but their multiplicative effects.

Our study is not without limitations. First, we studied self-reported depression rather than clinical depression diagnosed by a psychiatrist. We also did not include measures of depressive symptoms or use of anti-depressant medications. Self-reported depression is not ideal to measure depression, and NHBs and NHW may have different levels of awareness of their mental health. They may also experience different levels of stigma around mental health, which could affect their willingness to report depression. Thus, our variable may differently reflect depression for NHWs and NHBs [ 56,108]. Third, we did not control for all possible confounders such as the details of participants&#x02019; social relations. Fourth, this study did not explore how loss of a loved one operates through stress, reduced emotional social support, or negative thoughts. Emotional social support may be a mechanism by which loss of a loved one is associated with depression [ 90]. We measured our variables as single items. Research is still needed to explore why racial differences exist in the effects of loss on mental health. Despite the above limitations, using a large and a representative sample of California adults was a strength of this study.

We found that NHB and NHW adults in California how different susceptibility to the effects of losing a love one to depression. Differential susceptibility of racial groups to losing a love one is not in line with NHB mental health paradox or NHB resilience but higher salience of social support and social relations in the life of NHBs compared to NHWs.

Ethics

All participants provided written consent. The current analysis is exempt from s separate IRB review because we only used fully de-identified data.

Acknowledgment and Funding

This study was commissioned by the California Black Health Network (CBHN) with funding from the California Health Care Foundation (CHCF).&#x000a0;In addition, Sharon Cobb and Mohsen Bazargan are supported by 5R25MD007610-14 (NIH) through the CDU-Clinical Research Education and Career Development Program.

Conflicts of Interest

The authors declare no conflicts of interest.