Previous estimates based on cause of death on death certificates underestimated the problem.
Results of a new study published in PLOS One estimated that about 12% of deaths in the United States are attributable to diabetes, making the disease the third leading cause of death.
“It has not been generally recognized that diabetes is involved in this high of a fraction of deaths,” Samuel H. Preston, PhD, department of sociology and population studies center at the University of Pennsylvania, told Medical Economics.
Prior estimates of the effects of diabetes listed the disease as the seventh leading cause of death nationwide; however, these estimates used the cause of death listed on death certificates.
“Using death certificates is quite difficult because people who die with diabetes typically have other complications that are present and may contribute to death,” Preston said. “It is somewhat arbitrary at the time what gets assigned at the underlying cause of death.”
For this study, Preston and colleagues instead estimated the population attributable fractions (PAF) for cohorts aged 30 to 84 who were surveyed in the National Health Interview Survey (NHIS) between 1997 and 2009 and the in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2010.
Members of these two cohorts were followed prospectively for mortality through 2011. The researchers identified diabetes status using self-reported diagnoses in both surveys and using HbA1c in NHANES. They compared the prevalence of diabetes among those surveyed and among the subsequent deaths to those surveyed.
The researchers found there was a high degree of consistency between the two data sets and definitions of diabetes in the hazard ratios, estimates of diabetes prevalence and estimates of the proportion of deaths attributable to diabetes.
Specifically, they found the proportion of death attributable to diabetes was estimated at 11.5% using self-reports from NHIS, 11.7% using self-reports from NHANES, and 11.8% using HbA1c from NHANES.
“We looked at the excess mortality of people with diabetes,” Preston explained. “That excess mortality amounts to about 90%. People with diabetes have about 90% higher death rates than people without diabetes. Here we got rid of that excess 90% so that people with diabetes had the same death rate as people without diabetes, and that is where we got the 12% figure.”
According to Preston, a similar study that used data from the 1980s with an estimated 90% excess mortality estimated that diabetes was responsible for about 4% of deaths.
“What has happened in the meantime though is that the prevalence of diabetes has skyrocketed and, as a result, it is contributing to a higher fraction of deaths,” Preston said.
Among the subgroups examined in the study, the PAF was highest among obese persons at 19.4%.
“Diabetes is dangerous and growing, and is the leading risk factor for diabetes is obesity,” Preston said. “These two are marching hand-in-hand.”