Are perceptions of community safety associated with respiratory illness among a low-income, minority adult population?

Exposure to social stressors and environmental hazards are more common and are elevated in low-income, minority, urban communities [1‐5]. While it is well established that higher polluting industries are more likely to settle near low income communities (or that low income communities develop nearby due to lower housing costs) [6, 7], there is also growing evidence that suggests that social disadvantage (e.g. exposure to community violence) magnifies the effects from environmental hazards on adverse health outcomes [8‐15].

Stress appears to not only have a direct effect on health outcomes, but may also influence susceptibility to or be influenced by other determinants of health [16]. Animal studies support the biologic plausibility that chronic social stress leads to greater susceptibility of respiratory health issues due to air pollution exposure [17]. Because the effects of air pollution on respiratory health, especially asthma, have been found to be greater in low socio-economic status (SES) populations, psychosocial stress has been suggested to also have an effect on this relationship [12, 18]; this suggests that those who are exposed to both air pollution and social stress are more susceptible to the adverse health effects of air pollution than those who are singly exposed [11, 19].

Residential community violence/crime or an overall feeling of living in an unsafe neighborhood are measurable factors of social disadvantage. Neighborhood-level violence/crime, as a measure for psychosocial stress, is a suggested risk factor in the pathophysiology of asthma incidence and morbidity [20, 21]. Perceived stress may also be a risk factor for adult-onset asthma [22] and has been strongly associated with increased asthma morbidity [23]. Exposure to violence, measured individually, has been consistently associated with worsening asthma morbidity in children [10, 24, 25], risk of asthma among children [11, 26], and asthma symptom severity in adults [27, 28]. In addition, the Adverse Childhood Experiences Study has found a graded and significant relationship between exposure to violence/trauma during childhood and incidence and morbidity of chronic obstructive pulmonary disease (COPD) as an adult [29].

Within the context of environmental hazards research, the effects on adverse respiratory health from air pollution may have become diluted in analyses when whole communities with healthy individuals are investigated. For instance, a greater and more pertinent effect on adverse health may be found when instead investigating vulnerable populations from both perspectives - either as environmentally vulnerable (e.g. those living very near high traffic roads or goods movement traffic) or as socially vulnerable (e.g. those who are more exposed to psychosocial stressors) [30, 31]. Research on the effects of social disadvantage and ambient air pollution on respiratory health is specifically lacking in adult populations living in the context of this type of intersection (social and environmental vulnerability), as is the case for minority populations living near a major pollutant emission source, such as a freight railyard.

The inland region of Southern California offers such a yet to be studied context. In a region well known for routine poor air quality [32], it is also home to the San Bernardino Railyard (SBR), a major stationary, diesel particulate emitting matter source [7], located adjacent to a densely populated low income community. This SBR region is on a shortlist of widely recognized full-fledged inland ports in the U.S. with diesel-powered locomotives and trucks operating 24/7 [7, 33]. According to the California Air Resources Board Health Risk Assessment report, the diesel particulate matter emissions within one mile from the SBR are estimated at about 22 tons per year, which represent 66% of the total on-site and off-site emissions combined [33]. Of 18 railyards in California, the SBR has the highest population exposure to railyard emissions due to the highest residential density near the railyard [7]. Moreover, San Bernardino County residents continuously exhibit higher chronic disease morbidity than California state counterparts, including asthma [34]. San Bernardino County is one of the most underfunded regions in the state [35]. It’s economic hardship became well known after the city filed for bankruptcy in 2012 [36]. Nearer to the SBR, the city of San Bernardino has the third largest number of gang members in the U.S. [35]. In 2012, metropolitan San Bernardino had more than twice the violent crime and murder offenses per population size compared to neighboring Los Angeles [37] and the city in past decades has consistently been in the top 25 most violent cities in the U.S. [38]. Many other social determinants from economic to neighborhood segregation support evidence that this is a socially and environmentally vulnerable population exhibiting varying levels of social deprivation [35].

Households from the San Bernardino city area directly adjacent to, and at varying distances from, this goods movement network were included in the Environmental Railyard Research Impacting Community Health (ENRRICH) Project, a mixed-methods, community-based participatory research study. During the community engagement phase of the research project, community members expressed that most of their immediate concerns centered on issues related to law enforcement, street lighting and repair, trees and greenery, and a violence and unemployment ripple effect [39]; in other words, social and neighborhood health factors, as opposed to personal health concerns such as asthma due to the nearby emitting railyard. This paper explores follow-up survey results of adult residents. The goal was to explore if the social stressor related concerns community members had (i.e. perceived community safety) were related to adverse respiratory health outcomes. Thus, in a predominantly low-income, minority, environmentally vulnerable adult population, this paper aims to determine (1) the extent to which this population perceives their community as unsafe and (2) to what degree such perceptions affect respiratory illness.

To our knowledge, this is the first study to provide insight into the possible effect of community perceptions of safety on respiratory illness among adult, low-income, minority community members who were also environmentally vulnerable due to living in a region with routine poor air quality and with the additional exposure to a major pollutant source. We found a strong independent association between lack of perceived community safety and doctor diagnosed respiratory illness. Perceiving your community to be unsafe, arguably a stressful feeling, increased the likelihood of having been diagnosed with a respiratory illness by 40%. This effect was independent from other known factors associated with respiratory illness, such as age, tobacco exposure, low socio-economic status, or even distance from a diesel emission particulate matter air polluter. Although no association was found between perceived community safety and respiratory symptoms among the un-diagnosed population, these 166 participants with symptoms but no respiratory illness (21% of the total population) may represent future diagnosed respiratory illness cases in this population. Our findings are in line with others who have found that psychosocial stressors, such as perceived community safety, may contribute to chronic stress which in turn may contribute to the development [29, 46] and/or exacerbations [23, 28] of chronic respiratory illnesses, including asthma and COPD. Chronic stress is known to affect the immune and pro-inflammatory systems, which are suspected to be a part of the etiology of asthma [17, 22, 47‐49].

Studies of the effects of psychosocial stress, including exposure to violence [11, 24, 26, 28], on respiratory health have largely been conducted in children [10, 12, 13]. To our knowledge, this rigorously designed cross-sectional study of individually-measured, perceived lack of community safety on respiratory health among a socially- and environmentally- vulnerable adult population is the first of its kind. Our results are in line with a broader body of literature that supports evidence for the association between past exposure to trauma or violence/adversity, often measured by adverse childhood experiences, and chronic respiratory illness among adults [29, 46, 50‐52]; however, these did not adjust for ambient air pollution indicators as was did by adjusting for residential proximity to the railyard.

In addition, in line with our previously published qualitative findings [39] we found that a perception that one’s residential community is unsafe was highly prevalent. More than one in four participants reported an unsafe walking environment and more than one in three participants perceived a problem of community violence/crime demonstrating this population to be socially vulnerable. Perception of community safety was, not surprisingly, place-based; the closer the residential region was to the environmental hazard the higher the proportion of participants who reported lack of perceived community safety. Others have reported evidence supporting the causal relationship between upstream, place-based factors and health disparities [20, 31, 49, 53, 54]. Place of residence affects one’s perceptions of their surroundings [55, 56] and those perceptions may, in turn, affect the individual’s health [23, 57‐61].

Despite sampling methods to ensure that study participants were selected at varying distances from the railyard, minimal heterogeneity among the various respiratory health outcomes was found. This provided a unique setting to determine the relationship between a place-based, psychosocial determinant and respiratory health in a socially- and environmentally- vulnerable population. The minimal heterogeneity in the respiratory effects from the different environmental hazard (residential sampling) regions allowed us to examine how a psychosocial stressor may contribute to a well-supported [30, 62, 63] causal relationship between ambient air pollution exposure and adverse respiratory health.

The lack of heterogeneity of the respiratory health outcome across the three environmental hazard regions, in addition to the smaller sample size within each region, may have contributed to our null findings when determining if differences exist in the strengths of associations between perceived community safety and respiratory illness in our region-specific sensitivity analyses. On the other hand, these findings could also be due to residual confounding. These unaccounted-for factors (e.g. neighborhood green space or social cohesion) may be different within each residential region, as it has been demonstrated that many low-income, minority study populations have diversity in the range of social disadvantages experienced [20, 64]. Although non-significant, there exists a clear trend in the strengthening of the association between perceived community safety and respiratory illness as residential distance to the environmental hazard gets smaller. The individual associations within region A and region B are borderline significant. In Region A, the prevalence of diagnosed respiratory illness was 80% higher among the participants who reported lack of perceived community safety compared to those who did not report a lack of perceived community safety.

Like many other environmental epidemiology studies, our research had some limitations that should be noted. The ENRRICH study was cross-sectional and temporality between psychosocial stressors and respiratory outcomes cannot be determined. More prospective studies in adult, vulnerable populations in this field of research are needed. However, such studies are difficult to conduct due to financial and methodological constraints (tracking bi- and monolingual low-income populations is challenging). Lack of trust in low income neighborhoods may have affected participation; however, given our response rate (87%) we believe that our CBPR approach may have ameliorated this issue to a large degree. Also potentially affecting participation, some residents may not have been available for interviewing during daylight (working) hours or during early evening daylight hours we used for interviewing. Nevertheless, persons who had spent more time in the environment (resulting in higher risk exposure), and thus were more likely to be available to be interviewed, were of particular interest to this study. Another possible limitation is that both respiratory illness and respiratory symptoms were self-reported and thus subject to information bias; however, a compilation of survey items was explored to capture participants who may have underreported due to health care illiteracy or who may not have had a proper diagnosis due to lack of healthcare access [65].

Despite our limitations, our study has notable strengths. Sampled households were randomly selected. Data was collected by trained bi-lingual community members using a community based participatory research approach [40], which contributed to a high response rate, a notable strength, in a majorly Latino low income target population, with many facing immigration challenges. In addition, it is an advantage to be able to assess psychosocial stressors measured at an individual-level rather than solely relying on census block-level data or use of administrative data. Because the impact of a stressor depends on how one experiences, perceives, and/or interprets the event [47, 61], measuring social stressors at a group-level can result in imprecise assessments of psychosocial stress [10].

Furthermore, the consideration of the community member’s concerns as a guide for this analysis, by use of the ENRRICH project’s mixed-methods design, lends itself to two strengths. First, we analyzed an identified salient social stressor in this community across a spatially heterogeneous area, which reduced confounding by other hypothesized, less-salient stressors [10]. Secondly, because the experience and ‘voices’ of community members are crucial for successful place-based interventions [66], this study will be more easily translatable to public health practice in populations that are socially-, economically- and environmentally-vulnerable to adverse health outcomes.

Current research and environmental policy has focused too narrowly on air pollutants alone and should be broadened to take into account the cumulative impact of exposures and vulnerabilities encountered by people who live in low-SES, minority neighborhoods [67]. Our study demonstrates that heterogeneity with regards to social vulnerability (i.e. perceptions of community safety) exists within a low-income, minority population and that a non-chemical, yet place-based factor (i.e. perceptions of community safety) may contribute to adverse chronic respiratory health, aside from exposure to a stationary air pollution emission source alone. Future steps for analyses using the ENRRICH study population will be to test interactions between place-based factors and individual perceptions and to determine relevant social buffers for future intervention strategies.

In conclusion, when living in a low-income community near a goods movement network with high exposure to diesel emission and in a region with routine poor air quality, the added psychosocial stressor of perceiving your residential community as unsafe increases the likelihood of having a doctor diagnosed chronic respiratory illness. Our finding further supports that when trying to elucidate the effect of air pollution on respiratory health, public health professionals and policy makers must take into account a communities’ social, as well as environmental risk, context.