Past research suggests that Hispanic ethnic concentration can have varying effects on health and mortality outcomes [10
]. Some evidence suggests that it may be protective from early mortality, some forms of cancer, disability, depression and cognitive decline in Mexican Americans. Nevertheless, research also suggests that Hispanic ethnic concentration is a risk factor for late-stage cancer diagnosis, diabetes, and obesity [14
]. We expected that the association of Hispanic ethnic concentration and county-level obesity varies by socioeconomic environment. The findings from this support our expectations in a number of ways. First, we observed differentials in the relationship between % Hispanic and % obese by county level educational attainment. A high proportion of residents with a bachelor’s degree or more was associated with a low prevalence of county-level obesity. Counties with both high % Hispanic and high % with bachelor’s degrees or more had the lowest prevalence of obesity. At the same time, highest prevalence was in high % Hispanic/ low % with bachelor’s degrees. The regression analysis shed light on this association by demonstrating that % bachelor’s degree in a county has greater explanatory effect on the relationship between % Hispanic and % obese than does percent poverty. Still, when stratifying the analysis by high/low % bachelor’s degree and % poverty, we find that % Hispanic is only significantly negative associated with obesity prevalence in communities that more than 50% of the population has a bachelor’s degree and less than 50% of the population live in poverty. Findings that would indicate obesity rates in Hispanic communities are affected by differential socioeconomic conditions in the local environment.
Do et al
] found that % Hispanic was associated with a higher prevalence of obesity in individuals in the NHANES study. Findings from our study were mixed in their consistency with the results of Do et al
.’s study. First, % Hispanic was associated with lower county level obesity prevalence in counties where less than 50% of the population were living in poverty or more than 50% of the population had a bachelor’s degree. Second, this study revealed that counties that had 50% or more of the population below the poverty line, there were no differences in Hispanic ethnic composition, but the relationship was positive. The Do et al
. study used county-level characteristics to predict individual level outcomes, which is different that looking at the county-level as the unit of analysis for the outcome, therefore could explain the differentials in findings. Additionally, the Do et al
. study did not compare by different socioeconomic levels.
The relationship between Hispanic ethnic concentration and county-level obesity could also be the result of the cultural tendencies a community overall. To explore this possibility further we conducted a path analysis using % immigrant, % who speak English in the home, and % Hispanics who speak English less than well. Our findings yielded an interesting association that may provide insight into health outcomes associated with obesity in the Southwest [15
]. While % Hispanic had a positive effect on % obese, % immigrant had a negative direct effect. When taking into consideration other factors associated with both at a county level, such as % who speak English in the home, the relationships reversed in direction. Therefore, while immigrant concentration may serve as a protective factor potentially for the prevalence of obesity in a county, it depends on other related factors. From similar future analyses, we can further appreciate the complexity of this relationship and looking at multiple factors associated with the cultural environment may provide insight into the pathways at a community or environmental level responsible for inconsistencies in Hispanic disease and comorbidity prevalence’s across the nation.
At the individual level, in Mexican Americans, greater acculturation to the U.S. mainstream has been associated with lower fruit and vegetable intake [28
] and immigrants tend to have greater perceived control over their dietary intake [29
]. These differences in behaviors by level of acculturation may explain the differences we observed in the path analysis. Immigrant enclaves in New York City, where there is greater linguistic isolation (i.e.
, households of respondents where the primary language is other than English, however the dominant language of the surrounding neighborhood is English), have been observed as being associated with healthier eating habits [30
]. Additionally, while proportion immigrant was not associated with Body Mass Index (BMI), in New York City, linguistic isolation was [30
]. In the path analysis aspect of our study, while we found opposite effects of % Hispanic and % Immigrant, these effects were indirectly influenced by language use and educational attainment. Moreover, the influence of proportion of immigrants in a county contributed to a greater positive association of % Hispanic with county obesity prevalence. A similar finding was observed in Utah using driver’s license data linked to census tract % Hispanic (Latino) and % immigrant [31
]. While immigrant concentration was associated with a reduced risk of obesity in the state of Utah, % immigrant amplified this effect. Although the New York City and Utah studies are conducted at different ecological levels compared to our study, they all provide much needed information about the diversity of environments Hispanics live in and how this may exert inconsistencies from one level and one region to the next. In immigrant enclaves, traditions maintained by its inhabitants may be protective from the negative effects of poverty, since immigrant enclaves are often located in the poorest neighborhoods or areas [32
In the current study, the relationship between Hispanic ethnic concentration was associated with a higher prevalence of county-level obesity in communities with lower educational attainment and greater poverty. While most Hispanic communities in Texas are comprised of primarily U.S. born Mexican Americans, bilingualism is high, whereas Hispanics are fluent in both English and Spanish [22
]. Results from our study suggest that the health benefits of living in Hispanic counties may be most beneficial in places with higher educational attainment (e.g., where residents have more knowledge and awareness of both healthy behaviors and the adverse consequences of an unhealthy lifestyle).
Previous national and international studies that have evaluated the impact of socioeconomic environment on obesity have used just income [33
] or constructed artificial indices to evaluate risk. This study examined the effects of poverty and education separately and, as a result, found differential effects on Hispanic ethnic concentration and county-level obesity prevalence. Our findings on county-level educational attainment explained the effect of Hispanic ethnic concentration and accounted for a large portion of the effect of poverty. Poverty did not have as strong an effect on the relationship between % Hispanic and % obese as did education in this study. In fact, contrary to what was observed with % with a bachelor’s degree or more, prevalence of obesity did not vary significantly between poor counties that were either high % Hispanic or low % Hispanic. While poverty was significant in the regression model that did not include education, the inclusion of % bachelor’s degree in the final model accounted these effects. Results suggest that poverty may influence access to income-based resources that promote good health and healthy habits [34
], alternatively it may be that the effect of socioeconomic disadvantage is more complex than income alone and requires separate analysis for education and income.
Relatively few studies have looked at the relationship between obesity and educational attainment at a population level. A study in Cairo, Egypt, found that average neighborhood educational level was associated with BMI [36
]. At the individual level, education has been found to be associated with eating behaviors and levels of physical activity [37
]. The educational level of a population or community may encourage healthier living through the demand for healthier places to eat, gyms, and outdoor recreational areas and/or greater support for health living policies [38
]. This topic deserves further research attention because educational attainment at county-, state- and even country-levels has been associated with many health and mortality outcomes. Improving health education in obesity-prone communities may be an effective, cost efficient way to reverse the obesity epidemic in the U.S.