The relationship between fat distribution and diabetes in US adults by race/ethnicity

Abstract

Introduction: This study examined the relationship between fat distribution and diabetes by sex-specific racial/ethnic groups.

Methods: A secondary data analysis of National Health and Nutrition Examination Survey 2011-2018 data (n = 11,972) was completed. Key variables examined were visceral adipose tissue area (VATA), subcutaneous fat area (SFA), diabetes prevalence, and race/ethnicity. The association of VATA and SFA and diabetes prevalence was examined separately and simultaneously using multiple logistic regression. Bonferroni corrections were applied to all multiple comparisons between racial/ethnic groups. All analyses were adjusted for demographics and muscle mass.

Results: VATA was positively associated with diabetes in both sexes (p < 0.001) and across all racial/ethnic groups (p < 0.05) except Black females. No statistically significant relationships were observed between SFA and diabetes while accounting for VATA with the exception of White females (p = 0.032). When comparing racial/ethnic groups, the relationship between VATA and diabetes was stronger in White and Hispanic females than in Black females (p < 0.005) while the relationship between SFA and diabetes did not differ between any racial/ethnic groups.

Conclusion: This study found that VATA is associated with diabetes for both sexes across almost all racial/ethnic groups independent of SFA whereas the only significant relationship between SFA and diabetes, independent of VATA, was observed in White females. The findings indicated that visceral fat was more strongly associated with diabetes than subcutaneous. Additionally, there are health disparities in sex-specific racial/ethnic groups thus further study is warranted.

Keywords: adults; diabetes; subcutaneous fat area; type 2 diabetes; visceral fat.

Figure 1

Venn diagram of overlap for diabetes among diagnosed, medication taken, and blood glucose level. Out of 11,972 respondents, 418 was not included in this figure due to missing blood test. As the result of that, 11,554 were used for this figure.