Analog inulin is newer, costlier, and now more widely used than human insulin in the U.S. This study aimed to compare insulin adherence and healthcare utilization between users of analog and human insulin using real-world data.
This retrospective cohort study used 2016-2018 Tennessee Medicaid claims data to identify patients with type 2 diabetes and capture their insulin prescription fills and healthcare utilization. U.S. adults aged ≥ 18 years who were on analog insulin vial (AV), analog insulin pen (AP), or human insulin (HI) with ≥ 6 months of continuous enrollment after their first filled insulin prescription were included. Patients with proportion of days covered value ≥ 80% in 6-month baseline period were considered adherent. Healthcare utilization was assessed through numbers of ED visits and hospitalizations in 6-month follow-up period. Logistic regression and negative binomial regression determined associations between insulin type and insulin adherence or healthcare utilization, respectively.
The study included 2,763 individuals, including 69.8% female, 55.2% African-American, and 88.6% health professional shortage area residents. Insulin adherence was observed in 685 patients. AV users were less likely to be adherent compared to HI users (OR: 0.685, 95% CI: 0.527-0.889). Odds of adherence increased with age (OR: 1.028, 95% CI: 1.020-1.036) and use of additional antidiabetics (OR: 1.194, 95% CI: 1.080-1.319). AV users had higher odds of ED visits than HI users (OR: 1.288, 95% CI: 1.033-1.610). No significant associations with hospitalizations were seen. However, AP users were less likely to be hospitalized than HI users (OR: 0.618, 95% CI: 0.391-0.975). The odds of an ED visit and hospitalization decreased with age but increased with comorbidity burden.
HI and AP may have higher real-world benefit compared to AV in patients at risk for disparities. Providers and payers should consider the differences in real-world benefit in selecting the right insulin for patients.
© 2021 Published by Elsevier Inc.