Abstract
Keywords
Introduction
The Centers for Disease Control and Prevention (2018) reported that 700,000 people died in 2020 from heart disease, and according to the American Heart Association (2019), by 2035, close to 50% of the US population will have some form of cardiovascular disease (CVD). Research has shown that socioeconomic status (SES) has a measurable and important impact on CVD prevalence and outcomes (Schultz et al., 2018), with risk factors associated with CVD more prevalent in individuals from low socioeconomic groups (National Center of Health Statistics, 2012). While a considerable amount of research has examined the biological factors and health behaviors that contribute to CVD (Shiffman et al., 2020; Tran et al., 2021), there is a need for more research focused on the association between social determinants of health, such as neighborhood socioeconomic deprivation, with CVD risk factors and biological markers of CVD risk. Galobardes et al. (2006) define neighborhood socioeconomic deprivation as “the relatively low physical (e.g. houses for sale, graffiti), social (e.g. unemployment, non-voter), and economical position (e.g. income, education level) of a neighborhood.” Hypertension (HTN), the most common risk factor for CVD (Fuchs & Welton, 2020) has been associated with neighborhood deprivation (Xu et al., 2022). Other major CVD risk factors such as diabetes mellitus (DM), obesity, and dyslipidemia, have been also associated with neighborhood deprivation (Hu et al., 2021; Sheets et al., 2020; Shohaimi et al., 2014). The pathway between neighborhood deprivation and CVD is poorly understood. Several mechanisms have been studied: reduced access to behaviors that promote healthy lifestyles, and environmental conditions that do not facilitate physical activity (Calling et al., 2016); food insecurity, defined as the inability to access nutritious food, and relying in inexpensive high-calorie foods (Laraia, 2013); and environmental chronic stress as a result of high crime rate, abandoned buildings, physical incivilities, and litter (Augustin et al., 2008).
The Pew Research Center (2019) reported that an estimated 36.6 million Hispanics of Mexican origin live in the US, 26% in Texas, and about 20% of them live in poverty (Noe-Bustamante et al., 2019). Despite the size and growth of the Mexican American (MA) population, studies of the association between neighborhood deprivation and cardiovascular risk factors are scarce (Gallo et al., 2022; Savin et al., 2022).
The purpose of this study was to investigate if neighborhood deprivation, measured by the area deprivation index (ADI) (Kind et al., 2018), is associated with a higher prevalence of HTN, DM, dyslipidemia, and obesity. Also, we sought to evaluate if individuals living in a deprived neighborhood have increased odds of presenting with cardiovascular risk factors (CVRF). It is hypothesized that those living in neighborhoods with greater area deprivation index will have a higher prevalence and higher odds of having HTN, DM, dyslipidemia, and obesity. We also hypothesized that the relationship between neighborhood deprivation and CVRF will be impacted by ethnicity.
Methods
Study Design and Setting
Since 2017, the Health and Aging Brain Study: Health Disparities (HABS-HD) has been recruited participants. The HABS-HD study, previously known as the HABLE study, is a longitudinal community-based study. Subjects included for participation had an age of 50 and above and self-identified as Mexican American (MA), non-Hispanic White (NHW), or African American (AA). After signing a written consent form, HABS-HD participants undergo a clinical interview, neuropsychological exam battery, functional examination, MRI of the head, amyloid PET scan, and fasting blood draw for clinical and biomarker analysis. A comprehensive description of the HABS-HD study, including recruitment, inclusion-exclusion criteria, methodology, etc., has been published elsewhere (O’Bryant et al., 2021). The current study is a cross-sectional analysis of data from the HABS-HD study.
All HABS-HD data is available to the scientific community through the UNTHSC Institute for Translational Research (ITR) website at https://apps.unthsc.edu/itr/.
Study Population
From March 2017 until June 2022, 2,384 subjects were enrolled in the study. One thousand and forty-five participants self-identified as MA, 295 as AA, and 1,044 as NHW. One hundred and forty-six AA did not have enough information to obtain the ADI and/or a CVD diagnosis. Due to sample size limitations on statistical power, we excluded AA from the analysis. Of the remaining 2,089 participants, 137 were excluded due to lack of demographic information along with an additional 85 participants that did not have information to make an accurate diagnosis of HTN, DM, dyslipidemia, and obesity. All participants included in the final analysis (971 MA and 896 NHW) met the following criteria: (1) self-reported ethnicity as Mexican American or Non-Hispanic White, (2) have complete data to determine ADI, (3) have all data required for a diagnosis of HTN, DM, dyslipidemia, and obesity, (4) have complete data for all covariates (i.e., age, sex, education).
Area Deprivation Index
The ADI is a composite measure of neighborhood socioeconomic disadvantage (Kind et al., 2018) that uses 17 census variables including education, employment, family income, poverty level, occupation, unemployment rate, rate of home ownership, and access to telephones and vehicles to determine the level of disadvantage. The ADI allows for comparisons at the census-block group level and is publicly available at https://www.neigborhoodatlas.medicine.wisc.edu/. For analysis, we used the national ADI percentile ranking, with scores from 1 to 100. A score/ranking of 1 shows the lowest level of neighborhood disadvantage (least deprived), and a ranking/score of 100 is the highest level of neighborhood disadvantage (most deprived).
Cardiovascular Risk Factors
Hypertension
Hypertension was classified by self-reported medical history, use of blood pressure lowering drugs, and/or an average of two blood pressure measurements ≥140/≥90 mm Hg per the American Heart Association Guidelines.
Diabetes Mellitus Type 2
Diabetes was classified via self-reported medical diagnosis, current use of insulin or oral hypoglycemic agents, and/or HbA1c ≥6.5% per the American Diabetes Association guidelines.
Dyslipidemia
Participants were classified as having dyslipidemia if they had a medical diagnosis of high cholesterol and/or triglycerides, use of cholesterol-lowering drugs, and/or total cholesterol >200 mg/dL (National Heart, Lung and Blood Institute), and triglycerides >150 mg/dL (National Heart, Lung, and Blood Institute).
Obesity
To calculate obesity height (in inches) and weight (in pounds) were measured. Following the World Health Organization guidelines, participants with a BMI of 30 or higher were classified as obese.
Covariates
Demographic information (age, sex, and total years of education) were self-reported during the interview and entered as covariates.
Statistical Analysis
Data were analyzed with SPSS version 27 for Windows (SPSS INC., Chicago, IL). Chi-squared was used to assess differences in categorical variables and t-test for differences in the continuous variables. We also analyzed the prevalence of CVRF by ADI quartiles. The Cochran-Armitage test for trend allows for testing the difference between percentages or proportions with within-subject design for two or more levels. This test was used to evaluate trends in the association of HTN, DM, dyslipidemia, and obesity (0 = no, 1 = yes), and ADI quartiles 1 to 4 (1 = least deprived neighborhood, 4 = most deprived neighborhood). Finally, binary logistic regression was used to find the odds ratio of having the CVRF among participants in ADI quartiles 1 and 4. All the analyses were split by ethnicity while controlling for age, sex, and education. The one sided .05 level was used to determine statistical significance.
Results
Demographics
Table 1 presents the characteristics of the sample. The total sample of 1,867 participants had a mean age of 65.90 (SD 12.53); 61% of participants were female and with a mean of 12.53 (SD 4.80) years of education. When compared to Non-Hispanic Whites, MA were significantly younger and had fewer years of education (p ≤ .05). MA had a significantly higher prevalence of HTN, DM, and obesity. Only 12.3% of Non-Hispanic Whites lived in the most deprived neighborhoods (percentile 4), while 57.5% of Mexican Americans lived in the percentile 4 ranking areas (p ≤ .05).
Baseline Characteristics.
Note. MA = Mexican American; NHW = non-Hispanic Whites; SD = standard deviation.
Linear Association of Cardiovascular Risk Factors With ADI Quartiles
The Cochran-Armitage test was used to assess if the probability of having HTN, DM, dyslipidemia, or obesity increases as the ADI quartile increases. Table 2 shows that for MA, a significant linear increase from 30.1% to 41.8% was seen for DM. For NHW, a significant linear increase by ADI quartiles was observed for HTN (from 50% to 68.2%), DM (from 6.8% to 20%), and obesity (from 24.8% to 50%) (Table 3).
Cochran-Armitage Test for Trend—Mexican American.
Cochran-Armitage Test for Trend—Non-Hispanic White.
Association of Least and Most Deprived Neighborhoods With CVRF
As seen in Table 4, when compared with those in ADI quartile 1 (least deprived), MA living in the most deprived neighborhoods (quartile 4), showed no significant difference in the odds of having HTN, DM, dyslipidemia, or obesity when compared with those living in the least deprived neighborhoods. NHW participants living in the most deprived neighborhoods were 2.14 times more likely to have HTN (OR = 2.14, 95% CI [1.31, 3.48]), 3.42 times more likely to have a DM diagnosis (OR = 3.42; 95% CI [1.67, 7.01]), and 3.03 times more likely of being obese (OR = 3.03, 95% CI [1.86, 4.95]) than those leaving in the least deprived neighborhoods. There was no significant difference in the odds of having dyslipidemia between non-Hispanic Whites living in the ADI quartile 1 when compared to those living in the ADI quartile 4.
Odds Ratio. ADI Percentile 1 Versus Percentile 4.
Discussion
The present study examines whether the ADI is associated with HTN, DM, dyslipidemia, and obesity in MA and NHW enrolled in the HABS-HD study. Among middle age and older adults in our cohort, MA showed an increased probability of having DM as the ADI quartile increased. In NHW, the test for trend was significant for HTN, DM, and obesity. We also found that the odds of having HTN, DM, and obesity were higher for NHW living in the most deprived neighborhoods, while their MA counterparts showed no increased odds.
Despite living in similarly deprived neighborhoods (quartile 4), higher associations were found between neighborhood deprivation and CVRF in NHW than in MA. An explanation for these findings may be that cultural, behavioral, and socioeconomic factors associated with cardiovascular risk may have a different impact on Hispanics. Morales et al. (2002) did an extensive review and found that tobacco use, alcohol consumption, and diet have a greater impact on NHW health outcomes when compared to Hispanics. Dubowitz et al. (2008) stated that there is a possibility that MA are less susceptible to the effects of environmental factors because of the strong cultural influence on their diet. Becares et al. (2012), in a systematic review of the association of ethnic density (areas with a higher concentration of people from their own racial/ethnic group) with health, mortality, and health behaviors reported a protective effect of Hispanic ethnic density, especially for health behaviors. However, none of the studies in the review adjusted for area deprivation.
Studies in NHW cohorts have reported that living in more deprived neighborhoods is associated with higher prevalence and increased odds of having HTN (Keita et al., 2014; Xu et al., 2022) which is consistent with our findings. Contrary to our results, a study using data from the Multi-Ethnic Study of Atherosclerosis (MESA) (Mujahid et al., 2011) showed that Hispanics living in more stressful neighborhoods have a higher prevalence of HTN. One explanation may be the different Hispanic cohort background, 63.6% of Hispanics in the MESA study were foreign-born, and of those, 67.8% reported a place of origin in the Caribbean, while 100% of our Hispanic cohort is of Mexican origin. Recent research reporting results from 3,000 subjects (92.9% of Mexican origin) from the Hispanic Community Health Study/Study of Latinos and the SOL CASAS ancillary study (Savin et al., 2022), reported that socioeconomic deprivation was cross-sectionally and longitudinally associated with greater systolic blood pressure and the incidence of HTN. The difference in the findings may be at least in part due to the use of different methodologies to define HTN (e.g., using systolic blood pressure as a continuous variable, instead of HTN as a dichotomous variable).
Our findings suggest that neighborhood deprivation contributes to DM in both MA and NHW. Similar results were found in Sweden (Mezuk et al., 2013) and in a multiethnic population in the United States (Laraia et al., 2012). The lack of healthy foods, lack of recreational infrastructure, and air pollution, have been named as mechanisms that contribute to this link (Astell-Burt et al., 2014; Bilal et al., 2018; Morland & Evenson, 2009).
In a meta-analysis covering over 1 million subjects in high-income countries, Mohammed et al. (2019), concluded that living in low socioeconomic status neighborhoods was associated with higher odds of being overweight and obese. They consider the high availability of high-energy and junk food as the main contributing factor. For Hispanics, place of birth has been found to be an important factor when considering the link between segregation, neighborhood poverty level, and the risk of obesity. Phuong Do and Frank (2020) found no association between residing in poor neighborhoods and the risk of obesity in Hispanic immigrants, while the association exists in native Hispanics. Wen and Maloney (2011) also found that in Hispanics, there was a difference in obesity risk among US-born Hispanics and immigrants when compared to NHW, a difference that was not explained by neighborhood context. In our analyses, we did not consider nativity which may explain the lack of association between neighborhood socioeconomic status and obesity in our MA cohort. Future analyses with nativity variables are needed to clarify how neighborhood deprivation is related to obesity.
Contrary to other studies, we did not find a relationship between neighborhood deprivation and cholesterol. Jenkins and Ofstedal (2014) found that this association is driven by gender. Other authors reported similar results about sex differences in the association of lipid levels and socio-economic status (Shohaimi et al., 2014). Future analyses stratified by sex may help to clarify the differences.
Our study is subject to a number of limitations. The data are cross-sectional and causal inferences cannot be drawn from the results. Due to our Hispanic cohort being 100% Mexican Americans, the results may not be generalizable to other Hispanic populations. Factors such as duration of residence and changes in living environment along with other variables such as smoking status, physical activity, diet, chronic kidney disease, and medication use may have an impact and will need to be addressed in future research. Other factors like the different stress coping mechanisms, and family and religious support systems, may vary by ethnicity and need to be consider in future research. Even given these limitations, the current research shows the importance of ethnicity in evaluating the link between living in a deprived neighborhood and cardiovascular health.
Conclusion
The area deprivation index is associated with cardiovascular risk factors. Where we live matters. This study adds to the research that cardiovascular risk factors are not only a product of individual variables, but are related to social determinants of health. Analyzing these social determinants of health within and between ethnicity is an important contribution to the literature in the topic. Cardiovascular risk is attributed to modifiable factors with tremendous implications for intervention approaches tailored to individual and neighborhood determinants of health. Future studies of the difference in cardiovascular risk factors across socioeconomic levels are warranted.
Footnotes
Acknowledgements
The research team also thanks the local Fort Worth community and participants of the HABS-HD study.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: RV, AS, JH, declare that there is no conflict of interest. LJ has a financial interest in CX Precision Medicine, Inc. SO has multiple patents on precision medicine for neurodegenerative diseases and is the scientific founder of CX Precision Medicine, Inc.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported on this publication was supported by the National Institute on Aging of the National Institutes of Health under Award Numbers R01AG054073, R01AG058533, P41EB015922, U19AG078109, and R35AG071916. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Ethics Statement
The study has been approved by the North Texas Institutional Review Board (Prot. No. 2016-128).
Informed Consent
Written informed consent was obtained from each participant, and the study adhered to the guidelines outlined in the Declaration of Helsinki (World Medical Association, 2008).
