The medical expenditures associated with diabetes in general and by diabetes treatment mode among U.S. youth are less known. Using administrative claims data from >3 million youth, we estimated that the excess mean total medical expenditure associated with diabetes among youth was $7,593 per year. Estimated mean annual total medical expenditures were $9,333 for ITDM youth and $5,683 for NITDM youth, an excess of $3,650 for those with ITDM. Our ITDM group included all youth with type 1 diabetes but probably also some with type 2. Youth with NITDM most likely have type 2 diabetes (4
), therefore, our estimated medical expenditures for NITDM youth likely represent a lower bound for youth with type 2 diabetes receiving medications in the U.S.
We found that excess expenditures on prescription drugs contributed the most to the total excess medical expenditures associated with diabetes in youth. This contrasts with findings for the entire diabetic population, where hospitalization or inpatient expenditures contributed the most (5
). However, our results of health care spending across components are consistent with the estimates for a closely comparable age group (≤14 years) in Sweden (8
). Our relatively large expenditure on prescription drugs was mainly driven by a greater proportion of diabetic youth who required treatment with insulin, thus their greater expenses on insulin and diabetes supplies. This is consistent with findings in Sweden (8
We found that among diabetic youth and regardless of treatment mode, the drug expenses are mainly driven by the expenses for medications. The expenses for diabetic supplies are likely underestimated because not all health insurance plans cover the costs of the diabetes supplies.
Our estimated level of per capita total medical expenditure attributable to diabetes of $7,593 is greater than a corresponding per capita estimate of $6,649 for the entire U.S. diabetic population in 2007 (5
). Our estimated total medical expenditures ratio of those with diabetes compared with those without diabetes is also larger than that estimated for adults, which ranges from 2 to 5 (5
). However, our estimated total medical expenditures ratio is lower than in the previously mentioned Swedish study (8
). Our results are generally consistent with what has been previously reported: excess expenditure is higher in younger age groups, and the expenditures ratio between persons with or without diabetes tends to decrease with age (7
). The higher excess expenditure in younger age groups may have been driven by costs for specialist visits, medications, and diabetes testing supplies (8
). The higher medical expenditures ratio for youth may also have been driven by relatively lower medical expenditures for youth without diabetes than that for older populations without diabetes.
Our study has several limitations. First, our medical claim data are from enrollees in employer-sponsored health plans and without uncommon chronic conditions. The study population did not include individuals without health insurance coverage, those on Medicaid, and those identified with uncommon chronic conditions. Hence, the results may not be generalized to the U.S. youth population. Our sample only represents those enrolled in FFS plans and, therefore, may not reflect medical expenditures for those enrolled in capitated plans.
Second, the accuracy of our estimates is subject to diagnostic coding errors; however, the MarketScan database undergoes rigorous data quality checks, and less than 1% of all claims could not be verified (14
Last, we could not estimate the excess expenditures by diabetes type because the MarketScan database does not provide sufficient information to differentiate type 1 from type 2 diabetes. We were also unable to correctly identify diabetic youth who were treated with diet and exercise only; therefore, our sample does not include this population.
Our study showed that the excess medical expenditures associated with diabetes in youth were substantial. Our estimates associated with diabetes (both ITDM and NITDM) can be used to estimate the economic burden of diabetes in youth in the U.S. The estimates of expenditure for NITDM can also serve as a lower bound to evaluate the benefits of type 2 diabetes prevention programs. Future research on estimation of costs among youth with diabetes not on medications would provide additional insights into the expenditure associated with diabetes among youth.