Clinical Outcomes and Incremental Costs From a Medication Adherence Pilot Intervention Targeting Low-Income Patients With Diabetes at Risk of Cost-Related Medication Nonadherence

Abstract

PurposeThe extent to which reducing cost-related barriers affects diabetes outcomes and medication adherence among uninsured patients is not known. The purpose of these analyses was to understand the clinical impact and cost considerations of a prescription assistance program targeting low-income, minority patients with diabetes and at high risk for cost-related medication nonadherence. MethodsPatients received diabetes medications without copayments for 12 months. Change in diabetes control was calculated by using glycosylated hemoglobin (HbA1c) level at follow-up compared with baseline. Clinical data were collected from the electronic health record. Medication adherence for diabetes medications was estimated by using proportion of days covered (PDC). Incremental acquisition and per-patient costs, based on actual hospital medication costs, were calculated for different baseline HbA1c levels. FindingsPatients with baseline HbA1c levels ≥7%, ≥8%, and ≥9% experienced mean HbA1c reductions of 0.82% (P = 0.008), 1.02% (P = 0.010), and 1.47% (P = 0.010), respectively, during the 12-month period. The average PDC was 70.55%; 45.24% had a PDC ≥80%, indicating an adequate level of medication adherence. Medication adherence ≥80% was associated with ethnicity (P = 0.015), whereas mean PDC was associated with number of diabetes medication classes used (P = 0.031). Acquisition cost for 1242 prescriptions filled by 103 patients was $13,365.82, representing per-patient costs of $132.39; however, as baseline targets increased, acquisition costs decreased and per-patient costs increased from $10,682.59 and $169.56 to $6509.91 and $192.27, respectively. ImplicationsClinically significant reductions in HbA1c levels were achieved for all patients, although greater reductions were achieved with modest per-patient cost increases when considering patients with uncontrolled diabetes. Incorporating a multifactorial intervention to address cost-related medication nonadherence with a behavior change component may yield greater reductions in HbA1c with improved diabetes outcomes and meaningful hospital-based cost savings.