Measured and Diagnosed Diabetes and Mortality in the United States

Daesung Choi , University of Pennsylvania

Objectives: This study investigates the impact of diabetes on all cause-mortality using two different criteria- one based on HbA1c-level and the other based on self-reported diagnosed diabetes. In specific, the current study examines how the relative risk of death associated with diabetes at baseline diagnosed based on HbA1c levels varies by a previous diagnosis of diabetes. Methods: Data are from the National Health and Nutrition Examination Survey 1997-2010 and Linked Mortality Files. Adults aged 40-84 are grouped by the HbA1c cutoff level of = 6.5% and self-reported diagnosed diabetes. Hazard ratios of death during follow-up period are estimated by diabetes status defined both by HbA1c and diagnosed diabetes. Results: Adults with diagnosed diabetes show a higher risk of dying than individuals without diagnosed diabetes, irrespective of their HbA1c level at baseline. Conclusions: Mortality differentials are more pronounced based on a self-reported previous diagnosis of diabetes rather than when based on baseline HbA1c level.

See paper

 Presented in Session 5. Health & Mortality 1