Exhaustion of Prescription Benefits and Medicare Beneficiaries' Disenrollment From Managed Care
- Emily R. Cox, PhD;
- Brenda R. Motheral, PhD;
- Kathi Fairman, MA
To the Editor: Rector1 recently reported that enrollees in Medicare health maintenance organizations who had exhausted their capped prescription benefits were more likely than those who had not exhausted their capped benefit to disenroll from their health plan. We reexamined this relationship during a 2-year period for 3 plans varying in cap amount and administration of benefits.
Methods
Prescription and eligibility data were obtained from January 1, 1997, through December 31, 1998, for 3 Medicare health maintenance organization plans with capped prescription benefits. All plans were located in markets with moderate to high Medicare managed care penetration. Plan markets were located in West South Central (2 plans) and South Atlantic states (1 plan).
In 1997, annual capped benefits were $600 (plan A), $1000 (plan B), and $1500 (plan C) and were administered on a quarterly basis (ie, only one fourth of the annual cap amount was available each quarter). In 1998, all plans had annual capped benefits of $1000, administered quarterly in plans A and C and annually in plan B. Enrollment in 1997 and 1998 for plan A was 5434 and 6769, respectively; for plan B, it was 8667 and 10,372; and for plan C, it was 2961 and 3731.
Individuals with enrollment in the first 3 months of each year were included in the analysis. For all beneficiaries who reached the cap, whether in quarterly or annually capped plans, we identified the first month of the year in which the capped limit was exceeded. Unlike Rector, we were not able to identify and exclude members who disenrolled nonvoluntarily. Like Rector, we used an extended Cox model with the internally defined time-dependent variable of reaching the cap to analyze the relationship between reaching the cap and disenrollment from the health plan.2 Models were estimated for each plan and each year controlling for participant age, sex, and chronic disease score.3
Results
The percentages of members reaching their annual prescription cap for plans A, B, and C, respectively, were 22.6%, 0.7%, and 1.6% in 1997 and 12%, 4.1%, and 3.9% in 1998. Disenrollment rates among those enrolled in the first 3 months of each year for plans A, B, and C, respectively, were 19.3%, 28.9%, and 6.8% in 1997 and 10.4%, 22.9%, and 14.0% in 1998. Among those disenrolling in 1997, 21%, 7%, and 7%, respectively, reenrolled in 1998.
The risk of disenrollment across all plans and both years was significantly associated with older age, greater disease burden (ie, higher chronic disease score), and reaching the cap. In 1997, the relative risks (RRs) of disenrollment in any given month for those reaching the cap for the 3 plans were 2.62 (95% confidence interval [CI], 2.15-3.19), 2.21 (95% CI, 1.70-2.88), and 2.24 (95% CI, 1.43-3.50); in 1998, the RRs of disenrollment were 3.04 (95% CI, 2.40-3.86), 1.79 (95% CI, 1.12-2.86), and 2.30 (95% CI, 1.86-2.86) in plans A, B, and C, respectively.
Comment
Exhaustion of prescription coverage, whether administered on a quarterly or annual basis, was associated with a 2- to 3-fold increase in the RR of disenrollment. These findings expand on those of Rector and suggest that this relationship holds under various scenarios including variation in underlying use, cap amounts, and cap administration.
Acknowledgments
Funding/Support: This research was funded by The Robert Wood Johnson Foundation's Changes in Health Care Financing and Organization (HCFO) Initiative.
