Many seniors quit taking drugs for chronic illnesses such as diabetes and high blood pressure when they exceed their drug plan's yearly spending limits, according to a new Rand Corporation study.
Even when drug benefits resume at the start of a new health plan year, a significant number of seniors do not resume their prescription medications, according to the findings published in the September/October edition of the journal Health Affairs.
The study, which examined the behavior of seniors enrolled in a national private health plan, provides insight into how seniors may act under provisions of Medicare's new drug benefit plan that will leave about one-third of enrollees without drug coverage for some part of each benefit year.
"Prescription use falls significantly as patients reach their benefit caps," said Geoffrey Joyce, the study's lead author and a senior economist at Rand, a nonprofit research organization. "Most of the drugs we studied help prevent long-term complications of chronic disease so there are likely to be adverse health consequences for seniors who hit their caps."
Rand Health researchers studied prescription drug use from 2003 to 2005 among more than 60,000 people enrolled in a health plan offered to retirees by a large national employer. Enrollees had a choice of two drug plans that offered annual drug benefit caps of $1,000 or $2,500 and one drug plan that had no spending limit. Participants had to pay a portion of individual drug purchases in each of the plans.
The study examined enrollees' use of drugs used to treat high blood pressure, drugs that target cardiac problems, diabetes drugs, ulcer treatments and antidepressants. They also studied prescription pain medications that have over-the-counter substitutes.
About 6 percent to 13 percent of the people enrolled in drug plans with caps reached their spending limits in each of the years studied, with about half the affected enrollees going without benefits for more than 90 days, according to the study.
High spenders in the capped plans were more likely to discontinue use of their medications than people enrolled in the plan with no spending limits, according to researchers. Discontinuation rates differed by type of drug, ranging from 15 percent for anti-cholesterol medication to 28 percent for cardiac drugs. Rates were higher for pain medications and anti-ulcer drugs where over-the-counter alternatives were available.
Researchers say they were surprised that more people did not switch to generic drugs, given they are generally cheaper than name-brand medications. While people were less likely to quit using generic drugs once they reached benefit caps, no widespread move to the lower-cost alternatives was noted.
Among patients who stopped taking a medication in the capped plan, more than half did not restart their prescriptions during the first three months after benefits resumed.
"Given the importance of these drugs, it's distressing that the resumption rates are not higher," said Dana Goldman, the study's senior author and director of health economics at Rand. "Drug caps are a cost-saving measure, but our findings raise the issue of whether in the long run they may lead to other medical costs such as increased hospitalizations."
Researchers said the study may help guide policymakers who are concerned with the so-called "doughnut hole" in Medicare prescription drug plans. Spending limits contained in the Medicare drug plan are expected to leave between 24 percent and 38 percent of enrollees without drug coverage for part of each benefit year.
"Caps on prescription drug spending are not a prudent way to restrict costs," Joyce said. "Cycling in and out of coverage is likely to have adverse health effects over time."
Other authors of the study are Pinar Karaca-Mandic and Yuhui Zheng of Rand.
Support for the study was provided by the National Institute on Aging through its support of the Rand Roybal Center for Health Policy Simulation and the Bing Center for Health Economics.
Materials provided by Rand Corporation. Note: Content may be edited for style and length.
Cite This Page: