Neighborhood Characteristics and Inflammation Among Older Black Americans: The Moderating Effects of Hopelessness and Pessimism

Abstract

Background

Research documents the adverse health effects of systemic inflammation. Overall, older Black Americans tend to have higher inflammation than older non-Hispanic White adults. Given that inflammation is related to a range of chronic health problems that disproportionately affect Blacks compared to Whites, this racial disparity in inflammation may contribute to racial disparities in particular chronic health problems. Thus, a better understanding of its determinants in the older Black population is of critical importance. This analysis examined the association between neighborhood characteristics and inflammation in a national sample of older non-Hispanic Black Americans. An additional aim of this study was to determine whether hopelessness and pessimism moderate the association between neighborhood characteristics and inflammation.

Methods

A sample of older non-Hispanic Black Americans aged 60+ were drawn from the Health and Retirement Study (N = 1004). Neighborhood characteristics included neighborhood physical disadvantage and neighborhood social cohesion. Inflammation was assessed by C-reactive protein.

Results

The analyses indicated that neighborhood physical disadvantage and social cohesion were not associated with C-reactive protein. Hopelessness and pessimism moderated the association between neighborhood physical disadvantage and C-reactive protein.

Conclusions

Knowledge regarding the role of hopelessness and pessimism as moderator in the neighborhood–inflammation association can inform cognitive-behavioral interventions targeted at changes in cognition patterns.

A growing body of research documents the adverse health effects of systemic inflammation (1). Inflammation is the body’s immunological response to injury, infection, and threat. Acute increases in inflammation are short-lived and generally protective. However, prolonged inflammatory responses or chronically elevated inflammation levels can adversely affect health and may be a sign of immune dysregulation. Repeated or sustained exposure to physical and psychosocial stressors can contribute to immune dysregulation and increase risk for poor mental and physical health (2). For instance, elevated C-reactive protein (CRP), a marker of systemic inflammation, is associated with hypertension (3), metabolic syndrome (4), cognitive function (5), and mortality (6). CRP is also predictive of cardiovascular events (7,8). Older racial and ethnic minority adults tend to have higher levels of CRP compared to older non-Hispanic White adults (9). Given that inflammation is related to a range of chronic health problems, this racial disparity in inflammation may contribute to racial disparities in these chronic health problems. Thus, a better understanding of its determinants in the older Black population is of critical importance.

Neighborhood disadvantage (ie, neighborhood disorder) and associated stressors experienced by residents may contribute to immune dysregulation and elevated inflammation, particularly among older Black Americans, who may have spent a substantial portion of their lives living in disadvantaged neighborhoods. At the same time, socially cohesive neighborhoods may protect against the development of chronically elevated inflammation levels by regulating emotional and cognitive responses to stressful neighborhood conditions. Ross and Mirowski’s (2001) theory of neighborhood disadvantage and health hypothesizes that neighborhoods that are high in social and physical disadvantage are perceived as stressful environments by their residents (10). The chronic stress of living in these disadvantaged environments can lead to heightened stress reactivity, which then results in compromised health, such as increased inflammation.

Despite theoretical support for the role of stressful neighborhood characteristics in health, only a few studies have examined their associations with inflammation. Collectively, the findings from these studies have been mixed, varying across age, gender, neighborhood characteristics, and indicators of inflammation. Some studies have found that CRP is not associated with neighborhood socioeconomic disadvantage, perceived neighborhood safety, and other neighborhood characteristics (11,12), while other studies have found that neighborhood socioeconomic deprivation is positively associated with CRP (13). Several studies have documented race and gender differences in the associations between inflammation and neighborhood characteristics (14–17). Research on the effects of neighborhood social cohesion demonstrates that it is associated with lower levels of interleukin-6 but was not associated with CRP (18). Altogether, research on the effects of neighborhood characteristics on inflammation is inconclusive and there currently is no information available regarding older Black Americans in this area, which indicates a need for further investigation.

Research on factors that moderate the associations between neighborhood characteristics and inflammation is scant. Nevertheless, the cognitive diathesis–stress model (19), which postulates that the interaction between predisposed cognitive vulnerabilities and stressors lead to mental and physical health problems, suggests that negative cognitive dispositions could moderate these relationships. Evidence indicating that hopelessness and pessimism—generalized negative outcome expectancies that represent cognitive diatheses—are associated with inflammation and health conditions linked to inflammation suggests that they may moderate the neighborhoods–inflammation connection (20–22).

To address the critical knowledge gaps detailed above, the goal of this analysis was to examine the association between neighborhood characteristics (neighborhood physical and social environments) and increases in inflammation in a national sample of non-Hispanic Black Americans aged 60 and older. An additional goal of this study was to determine whether pessimism and hopelessness—negative cognitive dispositions—moderate the association between neighborhood characteristics and increases in inflammation. These psychosocial measures were evaluated as nontraditional risk factors for elevated inflammation. This analysis will focus specifically on non-Hispanic Black Americans aged 60 and older because neighborhood disadvantage disproportionately affects this population and challenges their ability to live safely and independently in their communities (23–25). This cross-sectional analysis will use data from the 2010 and 2012 waves of the Health and Retirement Study (HRS).

Method

Sample

Data from the current study came from the HRS. The HRS is a nationally representative panel study of adults aged 51 and older living in the United States. Data for the HRS have been collected biannually since 1992. The HRS sample is replenished once every 6 years. Respondents are selected for the HRS from a multistage probability sample design, and the HRS oversamples for Black adults and older adults living in Florida (26). Starting in 2006, HRS initiated the Enhanced Face-to-Face (EFTF) interview. In addition to the core interviews, the EFTF interviews included blood samples and self-administered questionnaires on psychosocial topics from a rotating random half-sample of the noninstitutionalized older adults (26). Specifically, the Psychosocial and Lifestyle Questionnaires (LBQ) were left with respondents upon completion of the in-person core interview and mailed back to the HRS (27). For biomarker data, special informed consent was obtained before the blood acquisition process. In 2010 and 2012, blood-based biomarkers were assayed in laboratories at the University of Washington (28). Data for the current study come from the RAND HRS Fat Files (version P), the RAND HRS Longitudinal File, and the HRS Sensitive Health Data data sets (which contain the biomarker data used for the current study). In the present study, the data from the 2010 and 2012 waves were concatenated to acquire data for the complete sample. Study inclusion criteria included community-dwelling respondents, who were aged 60 or older, self-identified as Black or African American, and did not self-identify as Hispanic or Latino. Respondents who reported multiple races were included in the study if they self-identified that Black was their primary race (N = 21). The analytic sample was limited to older Black Americans because historical and contemporary forms of racial residential segregation have led to concentrated social and economic disadvantage in Black communities (23,29), so these communities are more likely to be under-resourced (30). As a result, older Black Americans tend to experience more neighborhood disadvantages than other older racial and ethnic groups, which can affect their ability to age in place. Respondents who completed the 2010 or 2012 waves of the HRS Core interview and EFTF interview (including the LBQ and biomarker collection) were included in the analytic sample (N = 1004).

Variables

CRP levels were obtained from dried blood spots for all HRS respondents who participated in the EFTF interview. Dried blood spots are blood drops deposited on specially manufactured paper after a finger prick and can be collected as part of a regular interview without medically trained personnel, which makes collecting blood samples among a large population feasible (31). As recommended by Crimmins and colleagues (28), we utilized the National Health and Nutrition Examination Survey equivalent assay values for our measure of CRP. Due to its skewed distribution, CRP was log-transformed for the multivariable analyses. For more information on CRP and the HRS data collection procedures, please see Crimmins et al (28).

Neighborhood characteristic variables included neighborhood social cohesion and neighborhood physical disadvantage. Neighborhood social cohesion was assessed with 4 items, which measured the extent in which respondents felt like (i) they belonged in their area; (ii) could trust others in their neighborhood; (iii) other people living in their neighborhood were friendly; and (iv) other people living in their neighborhood would help them out if they were in trouble. These items were answered on a scale ranging from 1 to 7 and were reverse-coded. The reverse-coded items were averaged together to produce a neighborhood social cohesion score. The final scores were set to missing if there were more than 2 items with missing data (Cronbach’s α = 0.84). Higher neighborhood social cohesion scores represented greater neighborhood social cohesion. Neighborhood physical disadvantage was assessed with 4 items, which measured the extent of the following neighborhood problems: (i) vandalism and graffiti; (ii) fear of walking home after dark; (iii) problems with cleanliness of the area; and (iv) quantity of vacant/deserted homes or storefronts in the area. These items used a similar response scale to the neighborhood social cohesion items, and a method similar to the one described for neighborhood social cohesion was used to derive a neighborhood physical disadvantage score (Cronbach’s α = 0.81). Higher scores represented higher levels of neighborhood physical disadvantage.

Hopelessness was operationalized by a 4-item scale. This scale comprised 2 items from Everson et al’s (32) hopelessness scale and 2 items from Beck et al’s (33) hopelessness scale. Respondents indicated, on a scale of 1 (strongly disagree) to 6 (strongly agree), the extent to which they agreed or disagreed with the following statements: (i) I feel it is impossible for me to reach the goals that I would like to strive for; (ii) The future seems hopeless to me, and I can’t believe that things are changing for the better; (iii) I don’t expect to get what I really want; and (iv) There’s no use in really trying to get something I want because I probably won’t get it. The 4 items were averaged together with higher scores representing higher levels of hopelessness (Cronbach’s alpha = 0.83).

Pessimism was measured by the 3 pessimism items of the Life Orientation Test-Revised (34). The Life Orientation Test-Revised is used to assess respondents’ dispositional optimism and pessimism. The 3 pessimism items were: (i) If something can go wrong for me, it will; (ii) I hardly ever expect things to go my way; and (iii) I rarely count on good things happening to me. Respondents were asked to rate these statements on a strongly disagree (1) to strongly agree (6) scale. A pessimism score was created by averaging the responses across the 3 items, with higher scores representing higher levels of pessimism. The final score was set to missing if there was more than 1 item missing. The Cronbach’s alpha for the pessimism scale was 0.68.

Covariates included sociodemographic and physical and psychological health variables. Gender was dummy-coded (men = 1, women = 2). Education was measured by the following categories: less than high school (reference category), high school/general equivalency diploma (GED), some college, and bachelor’s degree or higher. Marital status differentiated between respondents who were married or cohabitating (reference category); separated, divorced, or widowed; and never married. Net household wealth was measured continuously and was imputed by RAND to account for missing values (35). Net household wealth was the sum of respondents’ and their spouses’ wealth minus all debt. We applied the inverse hyperbolic sine transformation to the net household wealth variable in the multivariable analyses because it included negative and zero values. The inverse hyperbolic sine transformation is preferable over the more traditional logarithmic transformation, as it can handle the extreme positive skewness of wealth while including negative and zero values (36). Nativity differentiated between respondents who were born in the United States and respondents who were born outside of the United States (reference group). Current smoking status, blood pressure medication usage, cholesterol-lowering medication use, and diabetes medication usage (both oral medications and insulation) were assessed dichotomously (yes/no). Alcohol consumption was coded categorically based on drinking levels established by the Dietary Guidelines for Americans and compared nondrinkers to moderate and heavy drinkers. The guidelines are gender-specific and define moderate drinking as 1–3 drinks/day and less than 8 drinks/week for women and 1–4 drinks/day and less than 15 drinks/week for men. Heavy drinking is 4 or more drinks/day or 8 or more drinks/week for women and 5 or more drinks/day or 15 or more drinks/week for men. Nondrinkers are individuals who do not drink alcohol at all. Depressive symptoms were measured by the 8-item version of the Center for Epidemiological Studies-Depression scale (37). Respondents noted how many depressive symptoms they had in the previous week, including whether respondents felt depressed, everything they did was an effort, if their sleep was restless, happy, lonely, they enjoyed life, sad, and they could not get going for much of the time during the past week. Two items from the Center for Epidemiological Studies-Depression 8-item scale were positively worded and were reverse-coded. All scale items were summed together to produce a count of depressive symptoms (range of 0–8). The KR-20 score for this scale was 0.78. Number of chronic health conditions was a count measure of the number of chronic diseases and conditions (ie, high blood pressure, diabetes, cancer, lung disease, heart problems, stroke, arthritis, and psychiatric problems) that respondents reported. Waist-to-height ratio was calculated by dividing waist circumference by height.

Analysis

All analyses utilized respondent-level weights for the biomarker sample from the HRS to account for the complex sampling design of the survey. Listwise deletion was used to handle missing data. Descriptive statistics for respondents with complete data versus missing data are presented in Supplementary Table 2. We utilized a series of multivariable linear regression models to test the relationship between neighborhood characteristics and CRP and the moderating effects of hopelessness and pessimism on these associations. We constructed interaction terms between hopelessness and neighborhood variables and pessimism and neighborhood variables to test the moderating effects of hopelessness and pessimism. These interaction terms were individually tested in separate regression models, and interactions that were significant at the p < .05 level were retained for the final regression models. Model 1 tested the effects of neighborhood characteristics and covariates on CRP. Model 2 included all variables from Model 1 and the addition of the neighborhood physical disadvantage * hopelessness interaction term. Model 3 included all variables from Model 1 and the addition of the neighborhood physical disadvantage * pessimism interaction term. Significant interactions are depicted using estimated values for CRP. Although hopelessness and pessimism were treated as continuous variables in the analysis, each is depicted categorically (low vs high hopelessness/pessimism) in the figures for ease of interpretation. The low and high hopelessness/pessimism groups were represented by respondents with a hopelessness/pessimism score of 1 SD below and above the mean, respectively. All multivariable analyses accounted for sociodemographic, health, and psychological characteristics. Analyses were conducted using Stata v16.1.

Results

Table 1 presents the characteristics of the sample and distribution of the study variables. Women comprised the majority of the sample (62%), and the mean age of respondents was 70 years. Close to one-third of all respondents reported an educational attainment level of a high school degree or GED, and nearly another third of all respondents reported an educational attainment level of less than high school. The mean net household wealth was $120 993. About 2 out of 5 respondents were either married or cohabiting, and a little less than half of respondents were either separated, divorced, or widowed. Most respondents were born in the United States (95%). Regarding health measures, close to 4 out of 5 respondents reported that they did not smoke, and the vast majority of the sample (94%) reported that they were taking blood pressure medication. Nearly one-third of the sample indicated that they either take oral diabetes medication or insulin. On average, respondents reported 2.5 chronic health conditions and 1.6 depressive symptoms. The mean CRP value was 4.9 mg/L, which is relatively high; CRP values at or above 3.0 mg/L are considered high-risk (38). The mean waist-to-height ratio was 0.61. Respondents reported relatively low levels of hopelessness and pessimism, moderate levels of neighborhood physical disadvantage, and high levels of neighborhood social cohesion. Supplementary Table 1 presents bivariate relationships between study variables and low- versus high-risk CRP.

Table 1.

Demographic Characteristics of the Sample and Distribution of Study Variables

	N (%)	Mean (SD)	Min	Max
Gender
Men	388 (37.58)
Women	716 (62.42)
Education
Less than high school	331 (30.76)
High school/GED	377 (34.43)
Some college	251 (21.33)
Bachelor’s or greater	144 (13.48)
Marital status
Married/cohabiting	501 (42.96)
Separated/divorced/widowed	541 (46.64)
Never married	62 (10.40)
Nativity
Born outside of the United States	56 (4.71)
Born in the United States	1046 (95.29)
Currently smokes
No	940 (86.35)
Yes	157 (13.65)
Blood pressure medication
No	49 (5.88)
Yes	830 (94.12)
Alcohol consumption
None	839 (76.04)
Moderate	227 (20.58)
Heavy	38 (3.38)
Cholesterol medication
No	393 (42.35)
Yes	571 (57.65)
Diabetes medication
No	756 (70.97)
Yes	339 (29.03)
Age		70.16 (7.79)	60	98
Net household wealth		120 993.20 (243 651)	−132 940	3 345 000
Depressive symptoms		1.62 (1.98)	0	8
Waist-to-height ratio		0.61 (0.09)	0.02	0.97
Chronic health conditions		2.54 (1.46)	0	8
Neighborhood physical disadvantage		3.24 (1.59)	1	7
Neighborhood social cohesion		4.90 (1.51)	1	7
Hopelessness		2.44 (1.34)	1	6
Pessimism		2.79 (1.33)	1	6
C-reactive protein		4.89 (8.11)	0.05	121.91

Notes: SD = standard deviation. Percents and N are presented for categorical variables and means and SDs are presented for continuous variables. Percentages are weighted and frequencies are unweighted.

Results from the multiple regression analysis indicated that neither neighborhood physical disadvantage nor neighborhood social cohesion was associated with CRP (Table 2, Model 1). The moderating effects of hopelessness and pessimism on the association between neighborhood physical disadvantage and CRP were tested in Models 2 and 3, respectively. A statistically significant interaction between neighborhood physical disadvantage and hopelessness (Model 2, B = 0.05, 95% CI: 0.01–0.10) revealed that among respondents with low levels of hopelessness, neighborhood physical disadvantage was marginally associated with higher CRP (Figure 1). However, among respondents with high levels of hopelessness, the positive association between neighborhood physical disadvantage and CRP was substantially stronger. As a result, at the lowest level of neighborhood physical disadvantage, respondents with high levels of hopelessness had similar CRP levels as respondents with low levels of hopelessness. In contrast, at high levels of neighborhood physical disadvantage, respondents with high hopelessness scores had substantially higher CRP than respondents with low hopelessness scores. The significant interaction between neighborhood physical disadvantage and pessimism (Model 3) indicated a similar pattern (B = 0.06, 95% CI: 0.01–0.11). Among respondents with low levels of pessimism, neighborhood physical disadvantage was unrelated to CRP (Figure 2). Conversely, among respondents with high levels of pessimism, neighborhood physical disadvantage was associated with high levels of CRP.

Table 2.

Multiple Linear Regression Analyses for the Association Between Neighborhood Characteristics and Log-CRP Among Older Non-Hispanic Black Americans

	Model 1	Model 2	Model 3
	B (95% CI)	B (95% CI)	B (95% CI)
Neighborhood physical disadvantage	0.07 (−0.01, 0.15)	−0.04 (−0.14, 0.06)	−0.08 (−0.21, 0.04)
Neighborhood social cohesion	0.02 (−0.06, 0.11)	0.03 (−0.06, 0.13)	0.02 (−0.07, 0.11)
Hopelessness	—	−0.13 (−0.32, 0.06)	—
Neighborhood physical disadvantage * hopelessness	—	0.05 (0.00, 0.10)*	—
Pessimism	—	—	−0.10 (−0.27, 0.07)
Neighborhood physical disadvantage * pessimism	—	—	0.06 (0.00, 0.11)*
Age	0.00 (−0.01, 0.02)	0.00 (−0.01, 0.02)	0.00 (−0.01, 0.02)
Gender
Men†	0	0	0
Women	0.07 (−0.22, 0.35)	0.07 (−0.20, 0.34)	0.07 (−0.21, 0.34)
Education
Less than high school†	0	0	0
High school/GED	−0.22 (−0.47, 0.03)	−0.19 (−0.45, 0.07)	−0.19 (−0.45, 0.07)
Some college	−0.23 (−0.49, 0.04)	−0.23 (−0.50, 0.04)	−0.21 (−0.49, 0.07)
Bachelor’s or higher	0.06 (−0.24, 0.37)	0.02 (−0.27, 0.31)	0.07 (−0.22, 0.35)
Marital status
Married/cohabiting†	0	0	0
Separated/divorced/widowed	0.10 (−0.10, 0.31)	0.11 (−0.10, 0.32)	0.10 (−0.11, 0.32)
Never married	−0.29 (−0.74, 0.17)	−0.27 (−0.72, 0.19)	−0.32 (−0.80, 0.17)
Net household wealth	−0.01 (−0.02, 0.01)	−0.01 (−0.02, 0.01)	−0.01 (−0.02, 0.01)
Nativity
Born outside of the United States†	0	0	0
Born in the United States	0.24 (−0.24, 0.72)	0.22 (−0.23, 0.67)	0.25 (−0.20, 0.70)
Depressive symptoms	0.02 (−0.03, 0.06)	0.01 (−0.04, 0.07)	0.01 (−0.04, 0.06)
Waist-to-height ratio	3.90 (2.45, 5.34)***	3.91 (2.51, 5.30)***	4.12 (2.79, 5.45)***
Currently smokes
No†	0	0	0
Yes	0.03 (−0.27, 0.32)	0.03 (−0.22, 0.29)	0.02 (−0.26, 0.30)
Chronic health conditions	0.00 (−0.09, 0.10)	0.00 (−0.10, 0.11)	−0.01 (−0.11, 0.09)
Blood pressure medication
No†	0	0	0
Yes	−0.08 (−0.85, 0.68)	−0.14 (−0.89, 0.61)	−0.14 (−0.88, 0.60)
Diabetes medication
No†	0	0	0
Yes	−0.07 (−0.33, 0.19)	−0.09 (−0.35, 0.18)	−0.05 (−0.32, 0.21)
Cholesterol medication
No†	0	0	0
Yes	−0.19 (−0.43, 0.06)	−0.21 (−0.46, 0.04)	−0.21 (−0.44, 0.02)
Alcohol consumptions
None†	0	0	0
Moderate	0.02 (−0.27, 0.30)	−0.03 (−0.29, 0.24)	−0.04 (−0.31, 0.23)
High	0.27 (−0.16, 0.70)	0.28 (−0.16, 0.72)	0.30 (−0.12, 0.72)
Intercept	−1.99 (−3.70, −0.27)*	−1.73 (−3.51, 0.04)	−1.89 (−3.62, −0.17)*
F	3.01	2.96	3.96
Complex design df	49	49	49
N	729	721	716

Notes: Model 1—neighborhood characteristics and covariates; Model 2—neighborhood characteristics, covariates, and neighborhood physical disadvantage * hopelessness interaction term; Model 3—neighborhood characteristics, covariates, and neighborhood physical disadvantage * pessimism interaction term. B = regression coefficient; 95% CI = 95% confidence interval; CRP = C-reactive protein.

^†Reference category.

*p < .05. ***p < .001.

Figure 1.

Estimated log-CRP values by neighborhood physical disadvantage and hopelessness among older non-Hispanic Black Americans. CRP = C-reactive protein.

Figure 2.

Estimated log-CRP values by neighborhood physical disadvantage and pessimism among older non-Hispanic Black Americans. CRP = C-reactive protein.

Discussion

The purpose of this study was to advance knowledge on the relationship between neighborhood characteristics—neighborhood social cohesion and physical disadvantage—and inflammation among older non-Hispanic Black Americans. The link between neighborhood characteristics and health conditions is well-established. However, the biological explanatory mechanisms by which neighborhood characteristics connect to health, such as inflammation, are largely unknown. Further, few studies on the physiological effects of neighborhood characteristics focus specifically on older Black Americans. Neighborhood characteristics are particularly important determinants of health for Black Americans, as this population is more likely to reside in under-resourced neighborhoods and experience the negative consequences of living in such neighborhoods (29). The current analysis is the first to use a national sample of older non-Hispanic Black Americans to examine the extent to which neighborhood social cohesion and physical disadvantage were associated with inflammation. Additionally, this study tested the moderating effects of negative cognitive dispositions (ie, hopelessness and pessimism) on the relationship between neighborhood characteristics and inflammation.

Our analysis indicated that neighborhood physical disadvantage and social cohesion were unrelated to inflammation among older non-Hispanic Black adults. Extant findings in this area are equivocal, with some studies indicating associations between neighborhood social cohesion and disadvantage and inflammation (11,12,18) and other studies indicating no association between neighborhood characteristics and inflammation (13,18). Our null findings are consistent with the body of literature indicating no direct relationships between neighborhood characteristics and inflammation. Research on the social determinants of inflammation indicates that, on average, social support is a less reliable predictor of inflammation than indicators of stress (39). This may explain why neighborhood social cohesion, which taps into social support at the neighborhood level, was unrelated to inflammation.

Findings from our study indicated that hopelessness and pessimism moderate the association between neighborhood physical disadvantage and inflammation. In particular, among older non-Hispanic Black adults who reported high levels of hopelessness and pessimism, greater neighborhood physical disadvantage was associated with more inflammation. In contrast, among older non-Hispanic Black adults who reported low levels of hopelessness and pessimism, neighborhood physical disadvantage was only marginally related to inflammation. These findings support the cognitive diathesis–stress model, which suggests that the combination of cognitive diathesis (eg, negative cognitive dispositions) and stress lead to compromised health and physiological functioning. In the current analysis, with regards to inflammation, older non-Hispanic Black adults who had higher levels of pessimism and hopelessness, perhaps due to their disposition, were more sensitive to the effects of neighborhood physical disadvantage, which is a chronic stressor. That is, among older non-Hispanic Black adults who have higher levels of pessimism and hopelessness, perceptions of their neighborhood may be influenced by their disposition, which may be shaped by experiences that explain their level of hopelessness and pessimism (eg, lack of change in economic, political, social, and personal conditions). In contrast, inflammation levels among older non-Hispanic Black adults who had lower levels of pessimism and hopelessness were more resilient to the effects of stressful neighborhood characteristics. It is possible that older non-Hispanic Black respondents who had lower levels of pessimism and hopelessness (or higher levels of optimism and hopefulness) may perceive their neighborhoods differently than older non-Hispanic Black respondents who had higher levels of pessimism and hopelessness. This difference in perception may contribute to their resiliency and lower inflammation levels. These findings are concordant with evidence showing that pessimism and hopelessness are predictive of greater levels of inflammation and risk for health problems associated with inflammation, such as cardiovascular disease (40,41). The moderating effects of pessimism and hopelessness demonstrate the multiplicative effects of neighborhood chronic stress and negative cognitive dispositions on physiological functioning. These findings further indicate the importance of considering the moderating role of psychological/cognitive responses to environmental stimuli on physiological responses.

The study findings should be interpreted within the context of its limitations. Several of the study measures were self-reported, which are subjected to social desirability and recall biases. Future studies may consider using census block and track data to derive objective measures of neighborhood characteristics and safety. Moreover, this analysis excluded respondents who were institutionalized. Consequently, the study findings can only be generalized to community-dwelling older non-Hispanic Black adults. Additionally, this study was cross-sectional, and we are unable to determine the temporal ordering between hopelessness/pessimism and neighborhood characteristics. Future studies should leverage the available longitudinal HRS data and confirm these associations longitudinally. Lastly, missing data in the current analysis should be considered within the context of the generalizability of the study findings. However, ancillary analysis indicated that the data were missing completely at random, which should not bias parameter estimates or affect generalizability.

There are several notable strengths and contributions of this study. The current analysis is focused specifically on older non-Hispanic Black adults and used a large, national sample of this population. Previous studies have paid little attention to the role of neighborhood characteristics and negative cognitive dispositions in inflammation among older and racially/ethnically diverse populations. Due to historical and contemporaneous racial segregation in housing, Black Americans are more likely to live in under-resourced neighborhoods that have more stress-inducing characteristics than non-Hispanic White Americans and, thus, are more likely to suffer the health effects of living in such environments (29). Black Americans are also more likely to have elevated inflammation levels than their non-Hispanic White American counterparts (9). Inflammation is associated with a range of health conditions that disproportionately affect Black Americans compared to non-Hispanic White Americans, including cardiovascular disease and diabetes, and may explain these racial disparities in health. This study provides information on factors that influence inflammation among older Black Americans, which contributes to insights into the biopsychosocial explanatory mechanisms for racial disparities in health. Additionally, the within-group analytic approach (as opposed to a race comparative approach) contributes to our nuanced understanding of the multiplicative effects of neighborhood physical disadvantage and negative cognitive dispositions on the physiological functioning of older Black Americans. Another strength of this study is its focus on inflammation rather than specific health conditions (eg, cardiovascular disease, metabolic syndrome), as is the approach of most studies of neighborhood characteristics and health. The focus on inflammation, which is associated with a range of health conditions that are also linked to neighborhood characteristics, can contribute to knowledge on possible mechanisms by which neighborhood physical disadvantage increases the risk for certain morbidities.

Older Black adults are more likely to have higher CRP than older non-Hispanic White adults, which puts them at greater risk of developing cardiovascular diseases and other health conditions. They are also disproportionately exposed to neighborhood physical disadvantage and may experience greater barriers to social cohesion. Taken together, the current findings can improve our understanding of how the physical environment in which people live interact with negative cognitive disposition to contribute to health and health disparities. Knowledge regarding the role of hopelessness and pessimism as potential risk factors in the neighborhood–inflammation association can inform cognitive-behavioral interventions targeted at changes in cognition patterns and factors that could mitigate the effects of residing in neighborhoods that have high levels of physical disadvantage. Given the null main effect findings for neighborhood social cohesion and physical disadvantage in the current analysis and the mixed findings in the extant literature, future studies should clarify the relationship between neighborhood social cohesion and inflammation.

Funding

The preparation of this article was supported by a grant from the National Heart, Lung, and Blood Institute to A.W.N. (5R25HL105444-11).

Conflict of Interest

None declared.

Author Contributions

A.W.N. designed the study and conducted the data analyses. W.Q. performed data management. A.W.N., H.O.T., K.D.L., W.Q., T.H., F.W., and U.A.M. wrote the manuscript.