While there are limited data on health disparities in cardiometabolic health in Caribbean populations, recent studies on disparities among Black Caribbean populations found that there is a higher prevalence of hypertension and diabetes among Caribbean Black populations compared to West African Black populations and European white populations as well as for those with lower socioeconomic status (SES), but lower prevalence of diabetes and hypertension when compared to African Americans (1, 2). Additionally, analyses using survey data from Jamaica have shown that the prevalence of diabetes was higher among men with less education and rates of hypertension, obesity and hypercholesterolemia were higher among less educated younger women (1–3). Individual-level socioeconomic factors alone do not explain the risk and there is increasing evidence that the social determinants of health including, economic stability, neighborhood and built environment, education access and quality, healthcare access and quality and social and community context, are major contributors to disease (4, 5).

Extensive data from the U.S. show that potential contributors to poor cardiometabolic health are neighborhood physical (housing, built environment, and environmental exposures) and social (community and society context, social cohesion, and social capital) environments specifically (6–8). However, only a few studies using Jamaican data have examined the neighborhood context and none have examined perceived community stressors. A recent study using data from a sample of residents in urban Jamaica evaluated SES determined by educational attainment, household assets and community property values with ideal cardiovascular health (9). Results showed that among men, those living in communities with lower median land values (MLV), property values as a proxy for neighborhood level socioeconomic status, had lower odds of ideal cardiovascular health compared to those living in communities with high MLV. The opposite was shown in women. Likewise, in older data (2008), poorer neighborhood infrastructure (as measured by distance to open space or public parks for physical activity) was associated with higher rates of overweight/obesity among men (10). In the US context, there is some evidence that community stressors can be associated with worse cardiometabolic outcomes (11). However, only one study in the Jamaican context evaluated interviewer perceptions of neighborhood disorder and home disorder, which were associated with lower levels of physical activity and more cumulative biological risk among women (12). This manuscript is the first study to explore associations of perceptions of community stressors as measured by community violence, victimization, and neighborhood disorder, with cardiometabolic outcomes (obesity, diabetes, hypertension, and high cholesterol) in urban Jamaica. We hypothesized that more community stressors would be associated with poorer cardiometabolic health.

Sociodemographic, community stressors and cardiometabolic characteristics of the study population are located in Table 1. p-values associated with t-tests for continuous variables and χ² tests for categorical variables were also reported. Of the 849 participants (M = 282; F = 567), mean age was 48 ± 18.5 years and most had at least a high school education (77.7%). Men were more likely to be current smokers (29.4 vs. 10.6%) and adequately physically active (53.2 vs. 42.0%). Men reported significantly higher scores on the community violence scale and women reported significantly higher scores on the neighborhood disorder scale. More women were obese (46.0 vs. 19.0%), more likely to have hypertension (52.9 vs. 45.4%) and have high cholesterol (34.2 vs. 21.6%) (all χ² p < 0.05).

Table 2 shows the adjusted prevalence ratios for community stressors with each cardiometabolic outcome for women. The middle tertile of community violence (PR: 1.65) was marginally associated with more diabetes (p = 0.058). The middle tertile of neighborhood disorder was significantly associated with high cholesterol [PR:1.72, 95% CI (1.20 to 2.47)] compared to the lowest tertile.

For men (Table 3), those in the middle tertile of neighborhood disorder had a significantly lower prevalence of obesity [PR:0.24, 95% CI (0.10 to 0.53)] compared to those in the lowest tertile. There was no significant association for those in the highest tertile of neighborhood disorder compared to the lowest tertile. Several other associations reached marginal significance (p > 0.05, p < 0.10). The highest tertile of community violence (PR: 0.55) was marginally associated with a lower prevalence of obesity, and the middle tertile of community violence (PR:1.51) was marginally associated with a higher prevalence of hypertension. Experiencing one event of victimization (PR: 2.20) was marginally associated with higher prevalence of diabetes. The middle tertile of neighborhood disorder (PR: 0.54) was marginally associated with lower prevalence of high cholesterol.

Sensitivity analyses were performed for participants who had high measured values for diabetes, hypertension, and high cholesterol, but did not have a doctor’s diagnosis and were labeled as undiagnosed. Cross tabulations with χ² tests comparing socio demographics and community stressors for undiagnosed participants with those who reported doctor diagnosis of the outcomes showed that the undiagnosed tended to be younger and included more current smokers (data not shown). Although the interaction terms between community stressors and sex were statistically significant for the obesity outcome only, analyses were stratified by sex for all outcomes.

To our knowledge, this is the first study to examine association of perceptions of community stressors with measured cardiometabolic outcomes in a robust sample of participants from urban Jamaica. While we hypothesized that higher scores of community stressors would show a linear association with poorer cardiometabolic health, we observed significant associations only for those in the middle tertile of neighborhood disorder with prevalence of higher cholesterol in women and lower prevalence of obesity in men.

Limited data on community-level stressors are available for comparison purposes in Jamaica or the Caribbean. Analyses using the Jamaican Health and Lifestyle Survey II (2008) have shown that more neighborhood disorder, as measured by interviewer perceptions of the conditions of home, streets and yards, amount of noise and air quality of neighborhoods, was associated with poorer cumulative biological risk (a combined marker of cardiometabolic health) among women (12). Further, in the same dataset, increased mean waist circumference (measure of central obesity) was higher for those living in areas with increased crime (22). Gender differences have been shown in prior studies conducted by the authors and have been different across several outcomes. In this analysis, men reported significantly higher levels of neighborhood disorder in the middle tertile than women, but interestingly not in the highest tertile. While these differences are not fully understood, our group has been stratifying data by gender and implementing sex-specific models to assure that we are documenting and actively exploring these relationships.

Our finding that the middle tertile of neighborhood disorder was associated with cholesterol and obesity and not the highest tertile did not support our hypothesis that this relationship would be linear. This finding is not unique to data from Jamaica, however, there are several other relationships that have shown a non-linear relationship in Jamaican datasets, including income with blood pressure (23) and obesity (22, 24), indicating higher levels for both high and low incomes. Given that these data are cross-sectional, causal inferences are not appropriate. However, it should be noted that Jamaica is going through an epidemiologic transition where there is a co-existence of obesity with undernutrition as is the case in many Low- and Middle-Income Countries (LMIC) (22). These changing population patterns indicate that the patterns may be shifting to the low socioeconomic status and worse chronic disease pattern that we see in US populations (4). Another possible explanation could be how residents experience stress at the neighborhood and individual level. There are no data to understand whether there were homogenous experiences of stressors, however, exploring this concept in a future study is warranted.

Despite its strengths, the analysis has several limitations including limited measures of socio-economic status as income is chronically underreported. Another limitation was the missing data (9%) for the community violence scale and finally, the study has no measure of alcohol intake which could serve as a potential confounder in our analyses (25). These analyses will contribute to the literature on neighborhood and health in the Jamaican context and allow for other studies to build on these measures which incorporate participant perceptions of their neighborhood.

Results suggest that higher, but not the highest level of neighborhood disorder was associated with higher cholesterol levels in women and lower obesity in men. Future work will explore additional approaches to measuring neighborhood characteristics in Jamaica and the mechanisms that may underlie any relationships that are identified. Future research from our team will explore novel measures of neighborhood-level SES (property sales and land values) and the relationship to other (individual and neighborhood-level) socioeconomic variables and with cardiometabolic outcomes. We also plan to further explore mechanisms contributing to neighborhoods and health in Jamaican datasets, and eventually compare these to cohort data for Black Americans.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the University of the West Indies (UWI) Mona Campus Research Ethics Committee (MCREC) (ECP89 16/17). The patients/participants provided their written informed consent to participate in this study.

TG-W and TF conceptualized and led the analysis and drafting of the manuscript. TF conducted the original survey along with many of the authors who are co-investigators. HD conducted statistical analysis under the advisement of TG-W, TF, and NY-C. All authors contributed to the article and approved the submitted version.

This research was supported by the Bernard Lown Scholars in Cardiovascular Health Program of the Department of Global Health and Population of the Harvard T.H. Chan School of Public Health (grant #: BLSCHP-1604). This work was also supported by a ‘Year of Global’ internal grant from the University of Pittsburgh.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.