Background

Hopelessness refers to a negative expectation about the future, coupled with a sense of personal futility and loss of motivation [1]. The risk of hopelessness is notably high in older age, partly because aging-associated challenges, such as reduced mobility and independence from declines in physical functioning or chronic health conditions, financial strain from reduced income or increased medical expenses, and social isolation from loss of close relationships and social support [2, 3]. These challenges may make older adults feel they lack abilities and resources to adjust to the changes that accompany the aging process, thereby diminishing motivation and undermining a positive outlook toward the future. According to Hamzaoglu and colleagues [4], 56% of adults aged 65 and above experienced feelings of hopelessness, a rate nearly three times higher than the 18% observed among those aged 15–24. The impact of hopelessness extends across various health and well-being outcomes in later life. Empirical evidence suggests that middle-aged and older adults who feel hopeless are at an elevated risk for cognitive impairment [5], cardiovascular disease [6], loneliness [7], depression [8], and mortality [9]. Furthermore, hopelessness has emerged as a strong predictor of suicidal ideation [10, 11] and a pronounced unwillingness to receive life-sustaining medical interventions [12] among older adults. Given its importance for health and well-being and its differential impact on older adults, identifying factors to reduce hopelessness in the older population represents a public health priority.

The learned hopelessness theory [13] offers a useful theoretical foundation for identifying potential risk factors. It posits that consistent exposure to uncontrollable and adverse environmental conditions gradually fosters a belief that such adverse situations are inescapable, culminating in a sense of helplessness [13]. Prior research has primarily focused on individual-level socioeconomic disadvantage and shown a strong association between educational attainment [14] and income [15] with a subjective sense of failure and subsequent hopelessness. However, little attention has been given to neighborhood socioeconomic disadvantage despite its potentially crucial role in the development of sense of hopelessness.

Neighborhood poverty and hopelessness

Neighborhoods can serve as a key upstream source, influencing the degree of stressors and access to opportunities [16]. Impoverished neighborhoods are often exposed to more chronic stressors, such as unemployment, violence, and crime [17], and offer fewer resources, including limited access to healthcare, opportunities for physical activity, and social interactions, than affluent neighborhoods [18–20]. In prior research, neighborhood poverty has been found to be associated with worse psychological well-being, including greater levels of depression and loneliness [20–22]. Similarly, area-level concentration of economic and social challenges can lower residents’ hope for future opportunities. In a sample of young adults aged 18–24 in greater Seattle area, Snedker & Hooven [23] explored the role of neighborhood stressors in psychological well-being and found that perceived neighborhood poverty was associated with an increased sense of hopelessness. While this first and only study on neighborhood poverty and hopelessness provides initial evidence suggesting the potential of neighborhood poverty to shape one’s negative perceptions of the future, it is based on a young adult sample in a highly selected geographic area, and therefore the role of neighborhood poverty in hopelessness in older adulthood remains unclear. The effects of neighborhood context are likely to be accentuated among older adults due to age-related functional limitations, mobility decline, and reduction in social networks that can lead to greater reliance on immediate residential environments [24]. Thus, we expect that greater neighborhood poverty will be associated with a greater sense of hopelessness among older adults (Hypothesis 1).

Neighborhood disorder as a mediator

Neighborhoods with higher levels of poverty may be harmful for psychological well-being due to the presence of stressors with known relationships to health. Social disorganization theory [25] provides a framework for conceptualizing the underlying process through which structural factors of neighborhoods, such as poverty rates, shape the larger social context. This theory posits that structural disadvantages, exemplified by high poverty rates, create a stressful social environment by disrupting the development and maintenance of social controls and community ties that are essential for addressing communal challenges [25]. Impoverished areas often lack resources and investments for maintenance, repair, and other essential services needed to uphold public order and social control [26, 27], which consequently contribute to shifting social norms toward non-conventional behaviors and incivilities. Such disruptions result in social decline and community decay, which is operationalized as neighborhood disorder. Neighborhood disorder is defined as “the physical and social features of neighborhoods that may signal the breakdown of order and social control (p. 4325)” [28], and is represented by features such as litter, graffiti, vandalism, noise, rundown and poorly maintained buildings and dwellings, drug and alcohol use, and conflicts among neighbors [26, 29, 30]. Empirical studies substantiate this theory, demonstrating that impoverished neighborhoods exhibit reduced social control and heightened physical disorder [29–31]. This degradation of neighborhood social conditions, in turn, can exert downstream effects on residents’ health and well-being [32–34]. Previous research has reported the detrimental impact of neighborhood disorder on poor health and psychological well-being outcomes of older adults, including anxiety, depression, cognition, and physical functioning [35–38]. Moreover, Ross and Mirowsky [26] emphasized the cascading psychological effects of neighborhood disorder, highlighting how its prevalence heightens fear while diminishing the perception of control over one’s life, fostering a sense of powerlessness to escape adverse circumstances. Empirical studies have provided evidence that neighborhood disorder acts as a mediator through which neighborhood disadvantage operates on residents’ psychological well-being. For instance, both Ross [26] and Kim [39] found that the association between neighborhood disadvantage–an index created based on the prevalence of poverty and mother-only households–and depression was largely explained by perceived neighborhood disorder. Based on the theoretical and empirical evidence, we anticipate that neighborhood disorder would mediate the association between neighborhood poverty and a sense of hopelessness (Hypothesis 2).

Present study

This study leverages data from a nationally representative sample of middle-aged and older adults to examine the association between neighborhood poverty and hopelessness. Furthermore, we conduct mediation analysis to assess neighborhood disorder as a potential intermediate pathway linking neighborhood poverty to hopelessness. Elucidating the underlying mechanisms of this association can contribute to a better understanding of how neighborhood socioeconomic disadvantage influences feelings of hopelessness, but also may guide intervention strategies. If the relationship between neighborhood poverty and hopelessness is mediated by increased disorder, targeted interventions to reduce such disorder may mitigate the risk of hopelessness.

Data

Individual-level data came from the Health and Retirement Study (HRS), a nationally representative, longitudinal survey of US adults aged 50 years and older. HRS respondents are asked about a wide range of information such as sociodemographic factors, health status, and neighborhood characteristics. In 2006, a random half of the HRS sample was administered the Psychological and Lifestyle Self-Administrated Questionnaire (SAQ), which includes items measuring sense of hopelessness and perceived neighborhood disorder, with the other half completing the SAQ in 2008 [40]. The most recent SAQ panel was 2018/2020, which includes respondents surveyed during the pandemic, a time when older adults may have felt hopeless for reasons related to the global pandemic. Therefore, we used the next most recent data from the 2014/2016 SAQ. There were 6,972 age-eligible respondents in 2014 and 6,102 in 2016 who completed and returned the SAQ, respectively. Among the respondents who completed the SAQ in 2014 or 2016 (N = 13,074), we excluded 618 (4.7%) who have missing information on the following individual-level variables (missing may overlap): perceived neighborhood disorder (n = 296), hopelessness (n = 292), education (n = 54), race/ethnicity (n = 44), and marital status (n = 9), and activity of daily limitations (n = 1).

Data on neighborhood poverty come from the U.S. Census Bureau’s American Community Survey (ACS) tract-level 5-year estimates data, which have been linked to HRS respondent information and are available from the HRS Contextual Data Resource (HRS-CDR) [41]. HRS geocodes respondent addresses and uses that information to determine the census tract in which they reside. We linked respondents from the 2014 HRS survey to 2012–2016 ACS poverty data, and the 2016 HRS respondents to 2014–2018 poverty data. We excluded respondents who did not have valid census tract identifiers (n = 162) or neighborhood poverty data (n = 10). The final analytical sample consisted of 12,284 adults aged 50 years and older.

Measures

Analytical strategy

We estimated a series of linear regression models to examine the association between neighborhood poverty and a sense of hopelessness. In the initial model, we assessed the association of neighborhood poverty with hopelessness, adjusting for individual-level sociodemographic characteristics. In the second model, we examined the link between perceived neighborhood disorder and hopelessness, The final model included both neighborhood poverty and perceived disorder to determine whether perceived disorder could explain the association between neighborhood poverty and hopelessness observed in the first model.

To further quantify the mediating role of perceived neighborhood disorder, we conducted a formal mediation analysis using structural equation modeling (SEM). SEM estimates all model parameters simultaneously, producing more accurate standard errors for the path coefficients compared to the regression technique [46, 47]. We first conducted a confirmatory factor analysis to validate the measurement model for our latent variables–perceived neighborhood disorder and hopelessness. Subsequently, in the structural model, we performed the effect decomposition. Here, the direct effect was represented as the path from poverty to hopelessness while controlling for neighborhood disorder and other covariates. The indirect effect was the path from poverty to hopelessness through neighborhood disorder, independent of other covariates. The total effect was the sum of the direct and indirect effects. Individual-level confounders were included as predictors of neighborhood disorder and hopelessness. The overall fit of the model was evaluated using the following criteria: a root mean square error of approximation (RMSEA) below 0.05, a comparative fit index (CFI) above 0.95, and a standardized root mean square residual (SRMR) under 0.08 [48, 49]. We assessed mediated effects based on 95% bootstrapped confidence intervals (1,000 iterations). Effect sizes were quantified in proportion to the total effects mediated by neighborhood disorder [50].

All analyses incorporated sampling weights to account for nonresponse and differential probabilities of selection and to make the sample representative of the U.S. population. We used single-level regression models given minimal clustering at the census tract level in our sample, where the average number of respondents per tract was only 2.2. A significant portion of the sample lived in tracts with few individuals (22% in a singleton tract, 24% in tracts with two people, 9% in tracts with three people, and only 17% in tracts with more than ten individuals). Analyses using the multilevel models produced largely similar estimates to the single level models. Nevertheless, we adjusted the standard errors for within-cluster correlation to mitigate the risk of underestimating standard errors in single-level models. Regression analyses were performed using Stata 17.0 [51], and mediation analyses were computed with Mplus 8.0 [52].

Ethical approval

We used data from the Health and Retirement Study (HRS), which is sponsored by the National Institute on Aging (NIA U01AG009740) and conducted by the University of Michigan following written informed consent under a protocol approved by an IRB at the institution. While most data are publicly available, access to study participants’ geographic location and the HRS-CDR used for this study is regulated and requires a formal application process due to privacy and confidentiality requirements. Approval for this study was obtained from the Institutional Review Board (UP-13-00397-CR006) at the University of Southern California.

Limitation

This study has several limitations worth noting for interpretations of the findings. First, our use of cross-sectional data precludes the possibility of drawing causal inferences. Although we conducted additional analyses adjusting for baseline levels of hopelessness to minimize the potential reverse causality issues, the establishment of causal associations remains beyond the scope of this study. Second, while we controlled for key individual-level covariates, the potential for bias arising from residual confounding cannot be entirely ruled out. Lastly, our use of subjective measures to assess neighborhood disorder and hopelessness might have introduced bias related to individual variations in perception.