Retrospective, long-term follow-up study of the effect of a three-tier prescription drug copayment system on pharmaceutical and other medical utilization and costs

Background: Previous research has suggested that 3-tier prescription drug copayment systems produce drug cost savings without affecting the use of other medical services during the first 12 months after implementation. Assessment of such systems with a longer follow-up period has been needed. Objective: This study examined the effect of a 3-tier copayment system on pharmaceutical and medical utilization and cost for 30 months after implementation in a population of commercially insured, preferred-provider organization members. Methods: This was a quasi-experimental, pre-post with comparison group design that gathered data retrospectively from the claims database of a preferred-provider organization in the Midwestern United States. The intervention group comprised members whose employer switched from a 2-tier (generic/brand copayment) plan to a 3-tier (generic/formulary/nonformulary) plan. The comparison group comprised members whose employer retained the 2-tier plan. Employers did not offer a choice between the 2- and 3-tier plans. Outcome measures included total drug cost; net insurer cost (drug cost minus copayment); number of prescription claims; numbers of office visits, inpatient hospitalizations, and emergency department visits; and rates of continuation with chronic medication therapy. Results: Relative to the comparison group (n = 4132), the intervention group (n = 3577) showed reduced growth in net cost and lower utilization of third-tier (nonformulary) medications (P < 0.001 and P < 0.01, respectively). The intervention and comparison groups did not differ significantly with respect to numbers of office visits, emergency department visits, or inpatient hospitalizations. Medication continuation rates were lower for the intervention than the comparison group at 6 months for oral contraceptives (P < 0.05), but chronic medication therapy continuation rates did not differ significantly at any other time point or for estrogens, antihypertensives, or antihyperlipidemics. Conclusion: In the population studied, previous research findings were confirmed over a longer time period. Copyright (C) 2003 Excerpta Medica, Inc.