Effect on Drug Utilization and Expenditures of a Cost-Share Change From Copayment to Coinsurance

BACKGROUND: While increases in prescription drug spending have moderated in recent years, drug spending is still a concern among managed care organizations and health plan administrators. In order to minimize cost increases from year to year, many health care plans have shifted more of the cost of medications to the member-consumer. Coinsurance, a benefit design in which the patient pays a percentage of the cost of the medication, is garnering more attention as a type of cost-sharing that differs from the traditional copayment model. OBJECTIVES: To estimate the impact on medication expenditures and utilization of a pharmacy benefit design change from 3-tier copayment to coinsurance. METHODS: Drug expenditures and utilization of beneficiaries aged e18 years and continuously enrolled in 2 privately insured groups were compared before and after a benefit design change in 1 of the groups. For the 12 months before the benefit design change, both groups had a 3-tier, fixed-dollar copayment structure with identical cost-sharing per 30-day supply: $10 tier-1 copayment for generic drugs, $25 tier-2 copayment for preferred brand drugs, and $40 tier-3 copayment for non-preferred brand drugs. On September 1, 2005, a 4-tier coinsurance benefit design (25% for all tiers except tier-3 [non-preferred] drugs at 50%, with minimum and maximum patient out-of-pocket [OOP] cost applied to each tier) was implemented in the intervention group (N=46,311). The 3-tier copayment design was maintained in the comparison group (N=7,916). A difference in- difference analysis was used to estimate the effect of the benefit design change on expenditures and utilization, overall (for all prescription drugs), and for 3 classes of essential medications: antihypertensives, antidepressants, and statins. Analyses measured changes in outcomes from 6 months pre-change (October 1, 2004, through March 31, 2005) through 6 months post-change (October 1, 2005, through March 31, 2006). In the overall (all drug) analyses, per member per month (PMPM) outcome measures were total pharmacy claims and cost, beneficiary (patient OOP) cost, and employer (plan sponsor) cost. Analyses of the 3 essential drug classes were limited to members with at least 1 claim in the drug class in both the pre-change and post-change periods (N=11,917, intervention group; 1,792, comparison group), and assessed per patient per month (PPPM) days supply, beneficiary cost, employer cost, and total cost. RESULTS: Beneficiaries in the intervention group paid 31.8% of total pharmacy benefit cost at the point of care versus 31.5% in the comparison group in the post-change period. The increases in beneficiary cost from the pre-change period to the post-change period were not significantly different for the intervention (7.5%) and comparison (3.0%) groups (P=0.983). From the pre-change period to the post-change period, total spending per member increased $4.57 PMPM (6.3%), from $72.29 to $76.87 in the intervention group versus a $5.87 PMPM increase (9.5%), from $61.54 to $67.41, in the comparison group, a relative difference of $1.30 PMPM (P=0.013). The increases in utilization from the pre-change period to the post-change period were not significantly different in the intervention group (2.4%) versus the comparison group (4.6%, P=0.189). Utilization per patient in the 3 essential drug classes increased 4.1% (1.59 days PPPM) in the intervention group versus 9.0% (3.23 days PPPM) in the comparison group (P=0.004). Total expenditures in the 3 classes for the intervention and comparison groups increased 8.2% ($5.07 PPPM) and 13.3% ($7.80 PPPM), respectively, a difference of $2.73 PPPM (P=0.003). Beneficiary cost for all 3 drug classes increased $2.20 PPPM (9.2%) in the intervention group versus $2.12 PPPM (9.1%) in the comparison group (P=0.032). The increases in employer cost for the 3 essential drug classes in the intervention group (7.5%, $2.86 PPPM) and comparison group (16.1%, $5.67 PPPM) did not significantly differ (P=0.057). CONCLUSIONS: A pharmacy benefit design change from tiered copayment to tiered coinsurance, without a significant increase in beneficiary OOP costs, was associated with a lower rate of increase in total pharmacy benefit cost and no significant reduction in utilization. For utilizers in 3 essential drug classes, drug utilization and total spending increased in the coinsurance group but at a lower rate of increase compared with the copayment group. The coinsurance design provides another approach for controlling prescription utilization and spending for certain medication classes.