LAUSR

AIMS To investigate the association between nonadherence to diabetes treatment and the occurrence of diabetes complications. METHODS Our study retrospectively identified adherence and nonadherence to diabetes treatment in patients during the first year of observation after new diagnoses of type 2 diabetes enrolled in commercial database from 52 health insurers in Japan. Participants were insurance enrollees with type 2 diabetes who received healthcare between 2005 and 2013, and who could be tracked for more than 12months from the initiation of diabetes treatment. We compared the occurrence of diabetes-related complications (retinopathy, nephropathy, neuropathy, ischemic heart disease, cerebrovascular disease, and chronic arterial occlusion) and all-cause healthcare expenditure during the second to eighth years. RESULTS We identified 1784 nonadherent patients and 9547 adherent patients. Cox proportional hazard models showed that the occurrence of microvascular complications was significantly higher in the nonadherent group: the hazard ratios (95% confidence intervals) for retinopathy, nephropathy, and neuropathy were 2.04 (1.57-2.66), 1.91 (1.35-2.72), and 1.83 (1.02-3.27), respectively. However, no significant differences were observed between the adherent and nonadherent groups for the macrovascular complications (ischemic heart disease, cerebrovascular disease, and chronic arterial occlusion). In addition, the nonadherent group had a significantly higher cumulative healthcare expenditure than the adherent group during the second-to-fifth-year period (p=0.029) and the second-to-sixth-year period (p=0.009) after treatment initiation. CONCLUSIONS Nonadherence in the first year of diabetes may increase the incidence of complications and result in higher expenditures for patients and payers.