Publications

Abstract

Purpose: The purpose of this study is to assess the weighted prevalence and odds ratios of obesity, prediabetes, and diabetes by (1) female sexual orientation (lesbian, bisexual, and straight) with racial/ethnic (Hispanic, non-Hispanic black, and non-Hispanic white) groups combined and (2) across and within racial/ethnic groups by sexual orientation. Methods: A secondary analysis of pooled 2014-2015 Behavioral Risk Factor Surveillance System data from 28 states (N = 136 878) was conducted. Rao-Scott chi-square test statistics were computed and logistic regression models were developed to assess weighted prevalence and odds ratios of obesity, prediabetes, and diabetes with adjustments for demographics (age, income, and education), depression, and health care access factors. Results: With racial/ethnic groups combined, lesbian and bisexual women, relative to straight women, had a significantly increased likelihood for obesity when controlling for demographics. Bisexual women were found to have significantly reduced odds for diabetes, compared with straight women, with adjustments for demographics, depression, and health care access factors. Compared with their non-Hispanic white counterparts, Hispanic lesbian women had significantly increased odds for obesity and diabetes, while non-Hispanic black bisexual women had a significantly greater likelihood for obesity, holding demographics, depression, and health care access factors constant. Non-Hispanic white lesbian women had an increased likelihood for obesity relative to their straight, ethnic/racial counterparts. Prediabetes subsample analysis revealed the prevalence was low across all female sexual orientation groups. Conclusion: Sexual minority women, particularly those of color, may be at increased risk for obesity and diabetes. Research is needed to confirm the findings.

Abstract

Background/Objective: Racial/ethnic disparities in type 2 diabetes (T2D) outcomes exist, and could be explained by nutrition- and inflammation-related differences. The objective of this study is to identify associations between race/ethnicity and glucose control among participants from NHANES 2007–2010, as influenced by diet quality, body mass, and inflammation and grouped by T2D status. Subjects/Methods: The following is a cross-sectional, secondary data analysis of two NHANES data cycles spanning 2007–2010. The association between race/ethnicity and hemoglobin A1c (HbA1c) as mediated by dietary intake score, body mass index (BMI), and C-reactive protein (CRP) was assessed, as was the strength of the difference of that association, or moderation, by T2D status. The sample included n = 7850 non-pregnant adult participants ≥ 20 years of age who had two days of reliable dietary recall data, and no missing data on key variables included in the analysis. The primary outcome examined was HbA1c. Results: The model accurately explained the variation in HbA1c measures in participants without T2D, as mediated by diet quality, BMI, and CRP. However, significant variation in HbA1c remained after accounting for aforementioned mediators when contrasting non-Hispanic White to non-Hispanic Black participants without T2D. The model was not a good fit for explaining racial/ethnic disparities in HbA1c in participants with T2D. A test of the index of moderated mediation for this model was not significant for the differences in the effect of race/ethnicity on HbA1c by T2D status (moderator). Conclusions: This study demonstrated that diet quality, BMI, and CRP mediated the effect of race/ethnicity on HbA1c in persons without T2D, but not in persons with T2D. Further research should include additional inflammatory markers, and other inflammation- and T2D-related health outcomes, and their association with racial/ethnic disparities in diabetes.