Mitchel L. Zoler, PhD
February 18, 2021
Roughly 40% of all US cases of incident diabetes during 2013-2016 were directly attributable to obesity, a finding that further solidifies the major etiologic role for obesity in the current American diabetes epidemic.
Researchers used data from a diverse cohort of 4200 American adults in the MESA study during 2000-2017 to calculate a relative risk for developing diabetes of 2.7 in people with obesity compared with similar participants without obesity.
They then applied this relative risk estimate to obesity prevalence rates during serial iterations of NHANES, the recurring US-wide survey of vital statistics in a representative cross-sectional population.
Their calculations showed that, during 2013-2016, 41% of US adults who developed new onset diabetes did so because of obesity, after adjusting for potential confounders.
This "population attributable fraction," or disease burden attributable to obesity, varied somewhat by sex, and by racial and ethnic subgrouping. Obesity was linked with the highest attributable rate among non-Hispanic White women, a rate of 53%, and with the lowest rate among non-Hispanic Black men, with an attributable fraction of 30%, Natalie A. Cameron, MD, and colleagues report in their study, published online February 10 in the Journal of the American Heart Association.
Potential for 'Meaningful Impact' by Reducing Obesity
"Our study highlights the meaningful impact that reducing obesity could have on type 2 diabetes prevention in the United States. Decreasing obesity needs to be a priority," said Cameron, of the McGaw Medical Center of Northwestern University in Chicago, Illinois, in a statement issued by the American Heart Association.
"Public health efforts that support healthy lifestyles, such as increasing access to nutritious foods, promoting physical activity, and developing community programs to prevent obesity, could substantially reduce new cases of type 2 diabetes," she added.
MESA (Multi-Ethnic Study of Atherosclerosis) enrolled adults 45-84 years old and free from clinical cardiovascular disease at six US sites during 2000-2002, and then followed them with four additional examinations through 2017.
For the current study, researchers narrowed the cohort down to 4200 participants who were 45-79 years old and free from diabetes at entry, and also restricted this subgroup to participants classified as non-Hispanic White (54% of the cohort), non-Hispanic Black (33%), or Mexican American (13%). At entry, 34% of the cohort had obesity, with a body mass index of at least 30 kg/m2.
During a median follow-up of just over 9 years, 12% of the cohort developed incident diabetes. After adjusting for possible confounders, a hazard ratio model showed an overall 2.7-fold higher rate of incident diabetes among people with obesity compared to those without.
The researchers then applied this hazard ratio to obesity prevalence statistics from NHANES (National Health and Nutrition Examination Survey) during the same time period, with data from the biennial NHANES project collapsed into four time strata: 2001-2004, 2005-2008, 2009-2012, and 2013-2016. They again limited their analysis to NHANES data collected from people 45-79 years old who self-reported categorization as non-Hispanic White, non-Hispanic Black, or Mexican American.
During the period from 2001-2004 to 2013-2016, overall obesity prevalence tallied by NHANES data rose from 34% to 41%. Among people with type 2 diabetes during 2013-2016, obesity prevalence was 65%.
To calculate the population attributable fraction researchers combined the MESA and NHANES estimates and adjusted for potential confounders and found that, overall, in 41% of people with incident diabetes during 2013-2016, the disease was attributable to obesity.
The study received no commercial funding, and none of the authors had disclosures.