PURPOSE: To estimate state-level diabetes-attributable deaths and years of life lost (YLL) in the Unites States.
METHODS: We estimated diabetes-attributable all-cause and cardiovascular disease (CVD) deaths by age, sex, and state, using the attributable fraction approach. Data on diabetes prevalence were collected from Behavioral Risk Factor Surveillance System. Relative risks for people with and without diabetes were estimated using the National Health Interview Survey. State-sex-age-specific deaths were obtained from CDC WONDER. YLL were calculated by multiplying the number of people with diabetes by the difference in life expectancy between people with and without diabetes using the life table approach.
RESULTS: Nationally, estimated diabetes-attributable all-cause deaths and CVD deaths were 293,224 and 90,953, respectively. Diabetes resulted in a total of 109,707,000 YLL with an average 4.4 years of life lost per person with diabetes. Most state variation in total deaths was explained by state population size and diabetes prevalence. All-cause deaths ranged from 415 in Alaska to 28,538 in California, and CVD deaths ranged from 113 in Alaska to 8908 in California. Across all states, the average diabetes-attributable death rate per 100,000 was 125 for males and 105 for females for all-cause deaths and 40 for males and 31 for females for CVD deaths.
CONCLUSIONS: Mortality attributable to diabetes is greatly underestimated when looking only at diabetes listed as an underlying cause of death. These results can be used to track state differences in deaths due to diabetes and to monitor the success of public health activities.