Abstract
Background: There is limited information about how prescription drug cost-sharing effects change over time.
Aim: In a natural experiment, we examined annual changes in prescription drug expenditures associated with having a two-tier prescription drug co-payment over a 5-year period.
Methods: We examined 8224 adult Kaiser Permanente-Northern California commercially insured members with diabetes in January 2001 who were continuously enrolled from 2001–2005. In 2001, subjects paid $5 for prescription drugs. Starting in 2002, 66% of subjects paid $15 for brand-name drugs and $5 for generics (two-tier), while 34% of subjects had no changes (one-tier). There were no further changes in co-payments from 2002–2005. We used linear regression with a generalized estimation equations approach to examine annual prescription drug costs, truncated at the 99th percentile. We adjusted for patients’ characteristics (age, gender, membership tenure, race/ethnicity, neighborhood social economic status and comorbidity) and secular trends (yearly indicators). We examined group differences (two-tier vs. one-tier) over time using group-year interactions and estimated adjusted prescription drug costs based on the model coefficients and mean values of the covariates.
Results: The mean age of subjects was 51 years; 47% were female, and 40% were white. In the baseline year, 2001, the mean number of prescription fills was 26.5: 27.0 in the two-tier group and 25.6 in the one-tier group. The adjusted annual total prescription drug costs were $1342 and $1435 in the two-tier and one-tier groups, respectively (difference=−$93, 95% CI: −$168 to −$18). In 2002, the first year with a two-tier drug co-payment, annual drug costs decreased significantly among the two-tier group compared with the one-tier group (difference-in-difference =−$52; 95%CI: −$97 to −$8). In subsequent years, the magnitude of the changes were small and not statistically significant relative to the baseline year (difference-in-difference=−$38 for year two of the two-tier co-payment, 95%CI: −$85 to $10; −$38 for year three, 95%CI: −$87 to $11; and −$34, 95%CI: −$88 to $20).
Conclusions: Introduction of a two-tier co-payment structure resulted in decreases in total prescription drug costs in the first year as expected. In subsequent years, however, changes in annual costs relative to baseline levels were smaller in magnitude and not statistically significant.
- Received September 11, 2008.
