A Comparison of Medical and Prescription Costs Between Disease Specific and Standard Benefit Plans Over Time: A Natural Experiment for Translation in Diabetes: A (NEXT-D) Study
Authors: Tannaz Moin, MD, MBA, MSHS, W. Neil Steers, Ph.D., O. Kenrik Duru, MD, MSHS, Norman Turk, MS, Charles Chan, MS, Abigail Keckhafer MBA, MPH, Robert Luchs, MD, Sam Ho, MD, Carol M. Mangione, MD, MSPH, and Susan L. Ettner, PhD
Background:
Diabetes-related health care costs account for 20% of all health care spending in the U.S. The Diabetes Health Plan (DHP) is an innovative disease-specific health plan for patients with diabetes and pre-diabetes offered by the nations’ largest private insurer , UnitedHealthCare (UHC). The DHP provides enhanced access to medications and primary care, resources for chronic disease management and patient clinical care reminders. We examined changes in total, out-of-pocket (OOP) and plan , medical, and prescription costs associated with employers’ purchase of the DHP, using employers that did not purchase DHP as concurrent control groups.
Methods:
We linked UHC claims and eligibility data to conduct an interrupted time series (ITS) analysis of mean employer-level costs per member per month (PMPM) between 2009-2013 (12 months of pre-DHP data to conduct a baseline assessment and define the study sample and 36 months of post-DHP implementation data). We included data from all employees/dependents who were continuously enrolled over 4 years, 19-62 years old, and diagnosed with diabetes or pre-diabetes at baseline (N= 2454 DHP enrollees). To identify comparable control employers offering standard benefit plans, we conducted an employer-level propensity match using mean age of employees/dependents, mean income, race/ethnicity, proportion employees/dependents who were female, 14 comorbidity indicators, proportion employees/dependents in high-deductible plans, plan anniversary month, a proprietary estimate of plan benefit generosity and quarterly costs at baseline. Final sample sizes were 8 DHP employers and 45 37-51 control employers. Separate ITS models were estimated for total, OOP, and plan prescription costs and for total, OOP, and plan medical costs, all reported on a PMPM basis. All expenditure measures were inflation and geography-adjusted. Results are presented as differences-in-post-intervention monthly changes-in-expenditures (the slope measuring the monthly post-intervention changes-in-expenditures among the DHP minus the comparable slope among controls).
Results:
After propensity score matching, the DHP and control employers balanced on all baseline employer-level characteristics.). Averaged over 3 years of follow-up, DHP was associated with significantly lower monthly changes in total prescription costs (-$1.64, p<.0001), OOP prescription costs (-$0.20, p<.0001) and plan prescription costs (-$1.33, p<.0001). DHP was also associated with significantly lower monthly changes in total medical costs (-$9.20, p=001), OOP medical costs (-$0.73, p=0.003) and plan medical costs (-$9.07, p=0.002).
Conclusions:
We found that employer groups who purchased the DHP had lower monthly changes in per-member-per-month total, OOP and plan prescription and medical costs over 3 years of follow-up after DHP roll-out, compared with concurrent controls offering standard benefit plans. These findings suggest that disease-specific insurance benefit designs that enhance access to care may play an important role in decreasing prescription and medical costs over time.