Carlos Vallarino, United States of America

Eli Lilly and Company Diabetes Business Unit

Presenter of 1 Presentation

ORAL PRESENTATION SESSION

REAL WORLD COST OUTCOMES WITH DIABETES TECHNOLOGY USAGE AMONG ADULTS WITH TYPE 1 OR TYPE 2 DIABETES USING RAPID-ACTING INSULIN

Abstract

Background and Aims

We aimed to determine the 1-year medical costs associated with initiation of a continuous glucose monitor (CGM) or insulin pump.

Methods

Using the IBM MarketScan database, we included people with diabetes (PwD) ≥40 years old using rapid-acting insulin who initiated a CGM or a pump (index date) between 2015-2017. A third cohort, a control group, were blood glucose monitor (BGM) users. Continuous enrollment was required in the 12 months before (baseline) and after (follow-up) index date. Mean total medical costs, excluding device-related costs, are reported as unadjusted values. Adjusted follow-up costs were calculated by fitting a generalized linear model with gamma distribution and log link.

Results

table.jpgPwD meeting all criteria (n=7,700) were assigned to one of the 3 cohorts (table). For CGM and Pump cohorts, unadjusted follow-up costs were higher than baseline. After multivariable adjustment, in the first year after initiation CGM was associated with $2038 in savings compared to BGM (p=0.0076) while pump was associated with $1786 higher costs (p=0.0039). Significant covariates in the models included baseline cost, age, diabetes type (and interaction with age), and polypharmacy.

Conclusions

During the follow-up period, the unadjusted medical costs appear lower than BGM for both CGM and Pump, but multivariable adjustment yielded medical cost savings only for CGM use. Economic benefits might be observable sooner for CGMs than for pumps. Future work should extend the analyses to longer follow-up periods, characterize the source(s) of the incremental savings/costs, and evaluate whether the results differ for newer generations of devices.

Hide