OBJECTIVE To update estimates of the economic burden of undiagnosed diabetes, prediabetes, and gestational diabetes mellitus in 2012 in the U.S. and to present state-level estimates. Combined with published estimates for diagnosed diabetes, these statistics provide a detailed picture of the economic costs associated with elevated glucose levels. RESEARCH DESIGN AND METHODS This study estimated health care use and medical expenditures in excess of expected levels occurring in the absence of diabetes or prediabetes. Data sources that were analyzed include Optum medical claims for ∼4.9 million commercially insured patients who were continuously enrolled from 2010 to 2012, Medicare Standard Analytical Files containing medical claims for ∼2.6 million Medicare patients in 2011, and the 2010 Nationwide Inpatient Sample containing ∼7.8 million hospital discharge records. The indirect economic burden includes reduced labor force participation, missed workdays, and reduced productivity. State-level estimates reflect geographic variation in prevalence, risk factors, and prices. RESULTS The economic burden associated with diagnosed diabetes (all ages) and undiagnosed diabetes, gestational diabetes, and prediabetes (adults) exceeded $322 billion in 2012, consisting of $244 billion in excess medical costs and $78 billion in reduced productivity. Combined, this amounts to an economic burden exceeding $1,000 for each American in 2012. This national estimate is 48% higher than the $218 billion estimate for 2007. The burden per case averaged $10,970 for diagnosed diabetes, $5,800 for gestational diabetes, $4,030 for undiagnosed diabetes, and $510 for prediabetes. CONCLUSIONS These statistics underscore the importance of finding ways to reduce the burden of prediabetes and diabetes through prevention and treatment.

BackgroundThe Centers for Disease Control and Prevention estimates that 28.9 million adults had diabetes in 2012 in the US, though many patients are undiagnosed or not managing their condition. This study provides US national and state estimates of insured adults with type 2 diabetes who are diagnosed, receiving exams and medication, managing glycemic levels, with diabetes complications, and their health expenditures. Such information can be used for benchmarking and to identify gaps in diabetes detection and management.MethodsThe study combines analysis of survey data with medical claims analysis for the commercially insured, Medicare, and Medicaid populations to estimate the number of adults with diagnosed type 2 diabetes and undiagnosed diabetes by insurance type, age, and sex. Medical claims analysis used the 2012 de-identified Normative Health Information database covering a nationally representative commercially insured population, the 2011 Medicare 5% Sample, and the 2008 Medicaid Mini-Max.ResultsAmong insured adults in 2012, approximately 16.9 million had diagnosed type 2 diabetes, 1.45 million had diagnosed type 1 diabetes, and 6.9 million had undiagnosed diabetes. Of those with diagnosed type 2, approximately 13.0 million (77%) received diabetes medication-ranging from 70% in New Jersey to 82% in Utah. Suboptimal percentages had claims indicating recommended exams were performed. Of those receiving diabetes medication, 43% (5.6 million) had medical claims indicating poorly controlled diabetes-ranging from 29% with poor control in Minnesota and Iowa to 53% in Texas. Poor control was correlated with higher prevalence of neurological complications (+14%), renal complications (+14%), and peripheral vascular disease (+11%). Patients with poor control averaged $4,860 higher average annual health care expenditures-ranging from $6,680 for commercially insured patients to $4,360 for Medicaid and $3,430 for Medicare patients.ConclusionsThis study highlights the large number of insured adults with undiagnosed type 2 diabetes by insurance type and state. Furthermore, this study sheds light on other gaps in diabetes care quality among patients with diagnosed diabetes and corresponding poorly controlled diabetes. These findings underscore the need for improvements in data collection and diabetes screening and management, along with policies that support these improvements.Electronic supplementary materialThe online version of this article (doi:10.1186/s12963-016-0110-4) contains supplementary material, which is available to authorized users.