Estimating State-Level Health Burden of Diabetes: Diabetes-Attributable Fractions for Diabetes Complications.

INTRODUCTION: Limited information is available on the health burden of diabetes at the state level. This study estimated state-specific attributable fractions and the number of cases attributable to diabetes for diabetes-related complications. METHODS: For each state, diabetes-attributable fractions for nine diabetes complications were estimated: three self-reported complications from the 2013 Behavioral Risk Factor Surveillance System, hospitalizations with three complications from 2011 to 2014 State Inpatient Databases, and three complications from 2013 Medicare data. Attributable fractions were calculated using RR and diabetes prevalence and the total number of cases using attributable fractions and total number of complications. Adjusted RR of each complication for people with and without diabetes by age and sex was estimated using a generalized linear model. Analyses were conducted in 2015-2016. RESULTS: Median state-level diabetes-attributable fractions for self-reported complications were 0.14 (range, 0.10-0.19) for mobility limitations; 0.13 (range, 0.04-0.21) for limitations in instrumental activities of daily living; and 0.12 (range, 0.06-0.20) for severe visual impairment or blindness. Median state-level diabetes-attributable fractions for diabetes-associated hospitalizations were 0.19 (range, 0.08-0.24) for congestive heart failure; 0.08 (range, 0.02-0.16) for myocardial infarction; and 0.62 (range, 0.46-0.73) for lower extremity amputations. Median state-level diabetes-attributable fractions for complications among Medicare beneficiaries were 0.17 (range, 0.14-0.23) for coronary heart disease; 0.28 (range, 0.24-0.33) for chronic kidney disease; and 0.22 (range, 0.08-0.32) for peripheral vascular disease. CONCLUSIONS: Diabetes carries a significant health burden, and results vary across states. Efforts to prevent or delay diabetes or to improve diabetes management could reduce the health burden because of diabetes.

Economic Costs Attributable to Diabetes in Each U.S. State.

OBJECTIVE: To estimate direct medical and indirect costs attributable to diabetes in each U.S. state in total and per person with diabetes. RESEARCH DESIGN AND METHODS: We used an attributable fraction approach to estimate direct medical costs using data from the 2013 State Health Expenditure Accounts, 2013 Behavioral Risk Factor Surveillance System, and the Centers for Medicare & Medicaid Services' 2013-2014 Minimum Data Set. We used a human capital approach to estimate indirect costs measured by lost productivity from morbidity (absenteeism, presenteeism, lost household productivity, and inability to work) and premature mortality, using the 2008-2013 National Health Interview Survey, 2013 daily housework value data, 2013 mortality data from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research, and mean wages from the 2014 Bureau of Labor Statistics. Costs were adjusted to 2017 U.S. dollars. RESULTS: The estimated median state economic cost was $5.9 billion, ranging from $694 million to $55.5 billion, in total and $18,248, ranging from $15,418 to $30,915, per person with diabetes. The corresponding estimates for direct medical costs were $2.8 billion (range $0.3-22.9) and $8,544 (range $6,591-12,953) and for indirect costs were $3.0 billion (range $0.4-32.6) and $9,672 (range $7,133-17,962). In general, the estimated state median indirect costs resulting from morbidity were larger than costs from mortality both in total and per person with diabetes. CONCLUSIONS: Economic costs attributable to diabetes were large and varied widely across states. Our comprehensive state-specific estimates provide essential information needed by state policymakers to monitor the economic burden of the disease and to better plan and evaluate interventions for preventing type 2 diabetes and managing diabetes in their states.

Diabetes-Attributable Nursing Home Costs for Each U.S. State.

OBJECTIVE: To estimate the diabetes-attributable nursing home costs for each state. RESEARCH DESIGN AND METHODS: We used a diabetes-attributable fraction (AF) approach to estimate nursing home costs attributable to diabetes (in 2013 dollars) in aggregate and per person with diabetes in each state. We calculated the AFs as the difference in diabetes prevalence between nursing homes and the community. We used the Centers for Medicare & Medicaid Services 2013-2015 Minimum Data Set to estimate the prevalence of diabetes in nursing homes and to adjust for the intensity of care among people with diabetes in nursing homes. Community prevalence was estimated using the Behavioral Risk Factor Surveillance System (BRFSS). State nursing home expenditures were from the 2013 State Health Expenditure Accounts. RESULTS: The fraction of total nursing home expenditures attributable to diabetes ranged from 12.3% (Illinois) to 22.5% (Washington, DC; median AF of 15.6%, New Jersey). The median AF was highest in the 19-64 years age-group and lowest in the 85 years or older age-group. Nationally, diabetes-attributable nursing home costs were $18.6 billion. State-level diabetes-attributable costs ranged from $21 million in Alaska to $2.0 billion in California. Diabetes-attributable nursing home costs per person ranged from $374 in New Mexico to $1,610 in Washington, DC (median of $799 in Maine). CONCLUSIONS: Our estimates provide state policymakers with an improved understanding of the economic burden of diabetes in each state's nursing homes. These estimates could serve as critical inputs for planning and evaluating diabetes prevention and management interventions that can keep people healthier and living longer in their communities.

Projected Costs of Informal Caregiving for Cardiovascular Disease: 2015 to 2035: A Policy Statement From the American Heart Association.

INTRODUCTION: In a recent report, the American Heart Association estimated that medical costs and productivity losses of cardiovascular disease (CVD) are expected to grow from $555 billion in 2015 to $1.1 trillion in 2035. Although the burden is significant, the estimate does not include the costs of family, informal, or unpaid caregiving provided to patients with CVD. In this analysis, we estimated projections of costs of informal caregiving attributable to CVD for 2015 to 2035. METHODS: We used data from the 2014 Health and Retirement Survey to estimate hours of informal caregiving for individuals with CVD by age/sex/race using a zero-inflated binomial model and controlling for sociodemographic factors and health conditions. Costs of informal caregiving were estimated separately for hypertension, coronary heart disease, heart failure, stroke, and other heart disease. We analyzed data from a nationally representative sample of 16 731 noninstitutionalized adults ≥54 years of age. The value of caregiving hours was monetized by the use of home health aide workers' wages. The per-person costs were multiplied by census population counts to estimate nation-level costs and to be consistent with other American Heart Association analyses of burden of CVD, and the costs were projected from 2015 through 2035, assuming that within each age/sex/racial group, CVD prevalence and caregiving hours remain constant. RESULTS: The costs of informal caregiving for patients with CVD were estimated to be $61 billion in 2015 and are projected to increase to $128 billion in 2035. Costs of informal caregiving of patients with stroke constitute more than half of the total costs of CVD informal caregiving ($31 billion in 2015 and $66 billion in 2035). By age, costs are the highest among those 65 to 79 years of age in 2015 but are expected to be surpassed by costs among those ≥80 years of age by 2035. Costs of informal caregiving for patients with CVD represent an additional 11% of medical and productivity costs attributable to CVD. CONCLUSIONS: The burden of informal caregiving for patients with CVD is significant; accounting for these costs increases total CVD costs to $616 billion in 2015 and $1.2 trillion in 2035. These estimates have important research and policy implications, and they may be used to guide policy development to reduce the burden of CVD on patients and their caregivers.