Objective: Evidence-based preventive care services are crucial for people with diabetes to improve their health outcomes, but the utilization of the services among older adults is suboptimal. Our objective is to compare relative performance of the Diabetes Belt and surrounding counties in the uptake of preventive services among Medicare beneficiaries with diabetes given available health care resources in the counties. The Diabetes Belt includes 644 counties in the Appalachian and deep south areas of the US that have high diabetes prevalence.

Methods: We used data envelopment analysis (DEA) to compute technical efficiency of counties for 2009-2015. Technical efficiency measures how many outputs a decision-making unit produces given a set of inputs. Inputs included the number of physicians and nurses per 1000 adults in the counties, and outputs included the number of foot exams, a1c tests, and microalbumin tests per 1000 Medicare beneficiaries with diabetes. The Diabetes Belt includes counties in the Appalachian and deep south areas of the US that have high diabetes prevalence. We performed t-tests to compare efficiency between the Diabetes Belt and surrounding counties. We excluded counties with 0 doctors or 0 nurses for >1 year in 2009-2015.

Results: Of 1057 counties, 861 (496 in the Belt and 365 in surrounding areas) were included. During the study period, the average technical efficiency of the counties ranged from 79% to 83%, with only 2.- 3.4% of counties achieving 100% technical efficiency. Diabetes Belt counties had consistently lower efficiency than surrounding counties during the study period (p <0.05) . The mean efficiency was 78-81% in the Diabetes Belt and 81-85% in the surrounding counties.

Discussion: Our results suggest that the degree of inefficiencies was higher in the Diabetes Belt compared to surrounding counties. Given relatively low available resources in these areas, more attention to improve diabetes preventive care is needed.


H.Kang: None. S.Kim: None. J.M.Lobo: n/a. M.Sohn: None.


National Institutes of Health (1R01DK113295-01A1)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.