Cost-Shifting and Medication Adherence

By David B. Nash, M.D., M.B.A., F.A.C.P.; Brian Leas,, MS, MA | June 16, 2008

A broad array of new medications have become available over the past decade to help patients control elevated cholesterol, hypertension, diabetes, depression, and a host of other serious chronic conditions. These drugs have expanded the available tools in the physician’s arsenal and are key ingredients in maintaining the health of millions of people worldwide.
But no medication — whether an old standby or a new blockbuster — will work unless patients fill their prescriptions and take the right dose at the right time.

In reality, prescriptions often go unfilled and pills are not taken on schedule. Studies consistently show that 25% to 40% of medications are not taken as prescribed.

Among the factors that contribute to poor medication adherence are the impact of side effects, the challenge of managing a complex regimen of multiple medications, and simple forgetfulness.

One factor that warrants special attention in the current healthcare environment is the patient’s out-of-pocket cost for prescription drugs. Today, nearly all insured patients face co-pays or co-insurance, and those who don’t have insurance foot the entire retail price.

Not surprisingly, research shows that medication costs influence patient behavior.

The landmark RAND Health Insurance Experiment, conducted more than 20 years ago, demonstrated that increasing a patient’s share of medical costs leads to a reduction in the use of services, including prescription drugs.

Subsequent studies confirmed the connection between the level of cost incurred by patients and their likelihood to adhere fully to a medication regimen.

Patients faced with high medication costs may decide to take smaller doses, or to take doses less frequently so that they can extend the time between refills.

Those who need multiple medications — half of all Americans over age 60 take at least 3 medicines every day, and 10% take at least 7 — may be faced with choosing which drugs they can afford to take and which prescriptions to leave unfilled.

Some patients may alternate pills from day to day to make each prescription last longer; others will opt to take a medication only when they experience symptoms.

These cost-saving strategies are ineffective at best and represent dangerous behavior for many diseases.

Poor medication adherence has broad health and economic consequences. Patients who don’t take their medications as prescribed are more likely to experience poor health and increased risk of comorbidities.

On a population level, improper management of chronic disease can result in greater direct healthcare costs arising from preventable hospitalizations or visits to the emergency department.

Researchers have found that improving medication adherence can cut medical costs in half for patients with diabetes and high cholesterol. Likewise, children with asthma are far less likely to need emergency care or be hospitalized when they use medications appropriately.

Finally, poorly controlled chronic conditions are associated with lost productivity and other indirect economic costs.

Although the twin issues of medication adherence and prescription drug costs have been a subject of research and policy debates for a long time, there is now a heightened sense of significance as employers and insurers seek to use cost-shifting as an important tool for managing their spiraling health care costs.

Rising co-pays, restrictive formularies, and the expansion of high-deductible health plans all serve to increase the financial burden on patients. It will be important to understand the impact of these strategies on patient utilization of necessary medications.

Several recent initiatives have sought to improve medication adherence by tempering the financial barrier of medications. The city of Asheville, N.C., often cited as a model in this area, reduced or eliminated co-pays for city employees diagnosed with certain chronic diseases. Pitney-Bowes has emerged as a leader in the corporate world for a similar type of program.

The researchers monitoring these programs have concluded that reducing the costs of key drugs for selected conditions (along with educational and medication management programs) has resulted in improved medication adherence and lowered overall healthcare costs.

As the retail pharmacy market continues to experiment with highly discounted generic drugs, and as Medicare begins to reflect on the early impact of Part D coverage, it will be important to conduct an ongoing assessment of the real impact of direct drug costs on patient adherence.

The debate over shifting costs to patients remains to be settled.

Want More on Health Policy?

Read David Nash’s blog at http://departmentofhealthpolicy.blogspot.com.

Three weeks ago I took the same pills as I had a bronchal infection and didn’t want to see a doctor. Two days later I began to suffer from RLS and was amazed. I ended up getting a Z pack and blowing it out of my system. Here’s some proof and information on genetic coding. All of the doctors mentioned have some financial links to pharmaceutical companies.

By Michael Smith, Senior Staff Writer, MedPage Today
Reviewed by Zalman S. Agus, MD; Emeritus Professor at the University of Pennsylvania School of Medicine.

REYKJAVIK, Iceland, July 18 — A major symptom of restless legs syndrome has been associated with a genetic variation on chromosome six, two research groups reported.

The variation has been linked to about half the risk of one of the major symptoms of restless legs syndrome, defined as the periodic limb movements in sleep that are commonly but not exclusively seen in RLS, according to Kari Stefansson, M.D., Ph.D., of deCODE Genetics here.

Using genome-wide association scanning to study patients with RLS, Dr. Stefansson and colleagues found that a genetic variation called rs3923809 accounts for about half the population attributable risk of having periodic limb movements in sleep.

On the other hand, it was not linked to RLS if patients did not also have the periodic motions when they were asleep, he and colleagues reported online in the New England Journal of Medicine.

The study is one of three reports in the journal and in Nature Genetics that detailed results of genome-wide association scanning as a method of linking genetics to disease.

Researchers led by Thomas Meitinger, M.D., of the Institute of Human genetics in Munch, Germany, reported in Nature Genetics that they had used the same method to study RLS and come up with complementary results.

But in addition to the chromosome six result, Dr. Metinger’s group linked two other genetic regions to the risk of RLS.

In genome-wide association scanning, researchers use data on genetic variation — so-called single nucleotide polymorphisms, or SNPs — amassed by the Human Genome Project and the international HapMap collaboration.

Then, using microarrays that recognize up to 500,000 SNPs, they look for polymorphisms that are more common in people with a disease than in healthy controls.

In a discovery phase look at RLS, Dr. Stefansson and colleagues first tested 306 Icelanders who had RLS with periodic limb movements in sleep, and compared their genetics with 15,664 healthy controls.

That analysis led to a single genetic variation in an intron of the BTB (POZ) domain-containing 9 (BTBD9) gene, which is expressed in many tissues, but whose functions are not well known.

The researchers then replicated the findings using two other cohorts — one in Iceland with 123 cases and 1,233 controls and one in the U.S. with 188 cases subjects and 662 controls.

All the cases had RLS with periodic limb movements in sleep, and carrying the variant rs3923809 was significantly associated with the condition, Dr. Stefansson and colleagues found.

Indeed, for the groups combined, the odds ratio for the condition was 1.7 for those carrying the variant, which was significant at P=3X10-14.

But when the researchers looked at people who had RLS without the movements in sleep, there was no association with the variant. On the other hand, the risk remained for those with movements but without the other symptoms of RLS, the researchers said.

In other words, they said, “we have identified a genetic determinant of periodic limb movements in sleep” but further research is needed.

Dr. Meitinger’s group, on the other hand, did not make the same distinction, but also found a link between RLS and chromosome six. In addition, they found links between the condition and variation in regions on chromosomes two and 15.

The region they cited on chromosome six contains the variant rs3923809, as well as several other polymorphisms in the BTDB9 gene.

On chromosome two, variations in the homeobox gene MEIS1 were associated with RLS, while on chromosome 15, the variation was found in the genes encoding mitogen-activated protein kinase MAP2K5 and the transcription factor LBXCOR1.

As in the other study, the researchers began with a discovery case-control cohort. There were 401 patients with familial RLS and 1,644 healthy controls.

Analysis of their genetics showed 28 single nucleotide polymorphisms in six regions that were possibly linked to the condition, Dr. Meitinger and colleagues said.

The researchers then tried to replicate the findings in two separate cohorts — one in Germany with 901 patients with either familial or sporadic RLS and 891 controls and the other in Canada with 255 cases and 287 controls.

Analysis of the first group reduced the significant regions to the final three, which was confirmed in the second group, the researchers said.

“A major proportion of the risk for RLS is explained by variants in the loci identified,” the researchers said.

The Icelandic finding is “exciting and important,” according to sleep researcher John Winkelman, M.D., Ph.D., of Brigham and Women’s Hospital and Harvard Medical School, while the Nature Genetics report “makes this finding even more secure.”

Writing in an accompanying editorial, Dr. Winkelman noted that the Icelanders appear to have found a gene for periodic limb movements in sleep, rather than RLS itself.

But because the German researchers did not make that distinction “it is unclear whether the other sequence variants they found are truly for RLS, for periodic limb movements in sleep, or for some other RLS marker,” Dr. Winkelman said.

But he added that the finding “offers hope” that better understanding of the syndrome’s pathophysiology will lead to better treatment.