Adult type 2 diabetes in the inland north west: A large regional study of the healthcare utilization, costs, and diabetes screening amongst diabetics from 2009 to 2011

Background: Over 10% of adults in the United States have type 2 diabetes and the estimated financial cost of diabetes is $174 billion. The substantial human and economic costs of diabetes suggest an important role for health policy in minimizing its burden on individuals, families, and society. This study examines individual and regional variation in healthcare delivery for adults with type 2 diabetes from 5 hospital referral regions (HRRs) of the Pacific Northwest.

Methods: We obtained 2009 and 2010 claims data for adults with type-2 diabetes from a third-party vendor. Data comprised separate commercial payer and Medicare samples from the 5 HRRs. Independent variables in ordinary least squares and logistic regression analyses were individual age, gender, proprietary diagnosis-based risk score, health plan type and geographic area characteristics. Area characteristics were measured in two ways: by Urban Influence Codes (capturing rurality) and the Health Resource and Services Administration published measure of medically underserved area/population (MUA/P) status to capture regional variation in availability of health care resources. Dependent measures were receipt of preventive screenings (HbA1c, LDL, retinopathy, nephropathy) and total all-cause allowed payments (“costs”) per person per year.

Results: In both commercial and Medicare samples across all years, between 19% and 29% of care-seeking beneficiaries showed no annual HbA1c screening test, between 39% and 47% showed no annual nephropathy test, and between 27% and 34% showed no annual LDL test. In total, only 47% to 55% of the patients received all three recommended screening tests at least once in a year, and only 10% -17% received all four recommended tests. However, screening rates were greater in 2010 than 2009. Among the commercially insured, lower screening rates were consistently associated positively with rurality and with having a commercial major medical health plan. For Medicare beneficiaries, rurality was not associated with undergoing screening; however, higher risk score and being female were positively related to screening rates. For both the commercial and Medicare samples, annual total costs per person were significantly and positively related to the individual risk score, but not to rurality.

Discussion/Conclusions: Regional rates of screening do not adhere to the American Diabetes Association recommendation that every diagnosed diabetic receive these screening tests annually. Commercial PPO and HMO health plans' benefit design is associated with increased odds of receiving recommended screening. Rural areas, in particular, face challenges in achieving recommended preventive screening for commercially insured persons.