Abstract
Introduction:
The HCHS/SOL demonstrated that diabetes mellitus (DM) was associated with elevated risk for CVD in a diverse Hispanic/Latino (H/L) cohort. However, the nature of these relationships as a function of H/L background, sex and other relevant variables is still unknown.
Objective:
We examined across approximately 6 years, the prevalence and incidence of DM and CVD and the CVD-DM relationship in diverse H/L.
Methods:
Participants at Visit 1:V1 (2008-2011) were 16,386 individuals with DM relevant data, whereas participants at Visit 2:V2 (2014-2016) were 8,401 individuals with similar data who attended the ongoing examination, constituting approximately 60% of the cohort to be studied. Descriptive characteristics were age-standardized to the 2010 U.S. population, and stratified by sex and H/L background. Prevalence estimates were weighted to the known population distribution, adjusting for sampling probability and nonresponse, and trimmed to handle extreme values of weights. Age-adjusted incidence rates /100 person years were estimated across Visit 1, based upon Poisson regression with robust variance taking into account the complex survey design. Both prevalence and incidence values are presented as % (95% CI). Prevalence and incidence of DM were examined by sex, age, H/L background, field center and BMI. We also examined the prevalence and incidence of CVD in those with and without DM by sex, H/L background, age and BMI.
Results:
Overall prevalence of DM was 17.8 (17.0, 18.6) at V1 and 19.4 (18.3, 20.5) at V2. The prevalence of DM at V2 was lowest, 11.2 (8.2, 15.3) for those of South American and highest for those of Puerto Rican, 22.5 (19.5, 25.8) background. While the prevalence of DM did not differ between women and men, the overall incidence rate for DM was significantly higher for men, 1.53 (1.32, 1.76) than for women, 1.06 (0.94, 1.18). The overall prevalence of CVD was significantly higher for DM than for non-DM individuals at V2: 9.2 (7.9, 10.7) vs. 4.5 (3.9, 5.2). The incidence rate across Visits, 0.71 (0.55, 0.92) vs. 0.20 (0.15, 0.27) was also higher for DM individuals. At V2 the CVD prevalence for DM men, 12.0 (9.7, 14.6) was greater than for DM women, 7.2 (5.6, 9.2). The relationship of CVD prevalence to DM status revealed different patterns among H/L background groups. At V2, for example, those of South American background showed relatively low CVD prevalence: 5.3 (2.6, 10.4) with DM vs. 4.2 (2.3, 7.5) without DM. In contrast, those of Puerto Rican background showed relatively high CVD prevalence: 15.6 (11.0, 21.6) with DM vs. 5.7 (4.1, 8.0) without DM.
Conclusions:
Overall prevalence and incidence of CVD was significantly higher for DM than for non-DM individuals and these CVD-DM relationships varied markedly across H/L background groups.