NEWS - Doctors, Too, Ask: Is This Drug Right?

[Guest guest]

The New York Times

MEDICINE'S DATA GAP

Doctors, Too, Ask: Is This Drug Right?

By BARRY MEIER

Published: December 30, 2004

Across the country, doctors are struggling to decide which pain

relievers to prescribe now that they know that popular drugs like Vioxx

and Celebrex pose potentially serious heart risks. " We are desperately

in need of information, " said Brenner, an internist in New

Haven. Yet for at least two years, doctors at the Mayo Clinic, the

federal Veterans Affairs Department and the Kaiser Permanente health

plan have been sharply limiting their use of Vioxx and Celebrex.

That is because those three institutions, after undertaking separate

reviews of test data available on various painkillers, reached the same

conclusion: For most patients, Vioxx, Celebrex and a related drug,

Bextra, did not work any better than older pain relievers or provide any

safety benefits beyond them.

A growing number of health care organizations have in recent years taken

rigorous steps to close one of medicine's biggest information gaps. They

are scrutinizing findings about all drugs, new and old, available to

treat a particular health condition to determine which work best at the

lowest cost.

Such evidence-based reviews, as they are known, are an effort to

separate scientific wheat from chaff by examining not just the quantity,

but also the quality of clinical trials and studies on a given drug.

They also seek to determine how a drug's risks and benefits stack up

against competing treatments. These reviews have been applied to every

major category of medication, from blood pressure treatment to

antidepressants.

Practitioners like Dr. Brenner do not typically consult such reviews

when deciding which drugs to prescribe, but instead rely on their own

experience. But that may change as government agencies and academic

centers disseminate drug review research more broadly.

The drive to base medical practice on such reviews is not new, but it is

taking on more urgency as health care costs rise and many newer drugs

prove to be only marginally better, if that, than older ones.

L. Matteson, a rheumatologist with the Mayo Clinic, said such

reviews also help to counterbalance the forces that influence doctors in

writing prescriptions, like drug company pitches, medical specialists

championing certain medicines and patients eager to get the latest drug

advertised on television.

" The pressures to prescribe are enormous, " Dr. Matteson said. " You

constantly have people at your door. "

One of the leading institutions doing drug class reviews is the

Evidence-based Practice Center at Oregon Health and Science University

in Portland. In recent years, some states have started to use reviews

produced by the center to draw up lists for preferred drugs that their

Medicaid programs will cover.

Today, about 12 states, including Washington, Oregon and Missouri, use

the center's reports to differing degrees.

Earlier this month, the federal government announced that it was

planning to spend $15 million in coming years to pay for evidence-based

reviews that will compare the effectiveness of various procedures and

drugs used to treat 10 of the most common health conditions, including

stroke, arthritis, pneumonia, diabetes and ulcers. The studies will be

made available to doctors and the public as well as to government and

private health plans.

Physicians like Dr. Brenner may welcome the help in deciding which

painkillers are appropriate for certain patients. In recent weeks,

studies have linked Vioxx, Bextra and Celebrex - all of which are in the

class of drugs known as COX-2 inhibitors - to increased heart risks

under certain conditions. In late September, Vioxx was withdrawn from

the market by its manufacturer, Merck. Pfizer, the maker of Bextra and

Celebrex, continues to sell both drugs but has limited marketing.

Another recent study has suggested that an older pain reliever,

naproxen, which is sold as Aleve, might also increase heart attacks,

though several experts said the Aleve data was less troubling because

the numbers appeared too small to be statistically significant. Drug

companies say they support the idea of evidence-based medicine. But they

also contend that the Oregon center's approach is more about cutting

costs than about science.

" What Oregon is doing is hiding a cost-cutting agenda that they are

marketing widely under the rubric of evidence-based medicine, " said Mark

Horn, the medical director for the government relations group at Pfizer.

Mark Helfand, the director of the Oregon center, dismissed that

suggestion.

And Dr. Matteson said he had heard similar complaints about the Mayo

Clinic's work from drug company representatives. " We've had plenty of

criticism from the representatives of different companies that market

these drugs to us that we are just looking at the bottom line and not to

the patient's best interest, " he said.

Advocates of evidence-based reviews say they can help make sense of the

incomplete and conflicting state of knowledge about different drugs used

for the same problem. Newer drugs, for example, are typically tested

more extensively than older ones, and not all drugs in the same class

are tested against each other. Moreover, the quality of clinical drug

trials run on each drug and their relevance to medical practice can also

vary sharply.

Those advocates have also long called upon pharmaceutical companies to

disclose all clinical trials run on a drug so that all evidence about

the drug is available to the public. Legislation to require companies to

register their drug trials was recently introduced in Congress.

To do an analysis, researchers try to pull together published and

unpublished clinical trials and studies about all drugs in a treatment

class by doing extensive literature reviews and asking pharmaceutical

companies for data. They analyze the studies to determine their

scientific rigor, eliminating those that they believe do not make the

cut. Even clinical trials, which are considered the most thorough

because they test medications under controlled conditions, can be flawed

or misleading because of their design.

" The biggest contribution that we make is in laying out the evidence, "

Dr. Helfand said.

Because responses to pain relievers varies among individuals, a drug

that works for one patient might not work for another. But after

reviewing all relevant data, the Oregon group and others doing similar

studies decided that Vioxx, Celebrex and Bextra did not relieve

arthritis-related pain any better than other drugs they had been tested

against, like ibuprofen, which is also a nonsteroidal anti-inflammatory

drug, or Nsaid.

In terms of safety, Vioxx showed the strongest evidence of reducing the

incidence of stomach bleeding associated with older Nsaids. Stomach

bleeding is a problem that is generally limited to older patients or

those with a history of gastrointestinal problems.

So even as the prescribing of COX-2 drugs by doctors in general was

increasing, the use of these drugs by doctors working for organizations

doing evidence-based studies was falling.

Doctors working for the Veterans Affairs Department have been curtailing

their use of COX-2 drugs since late 2001. The Mayo Clinic decided two

years ago to cut down its use of the medications by 50 percent.

Those groups and Kaiser Permanente, which also sharply limited its COX-2

prescriptions, saved money. Their actions may have also saved lives. A

study conducted by an F.D.A. researcher that was released in September

found that Kaiser patients relied on high doses of Vioxx, which can

increase the risk of heart attacks, less than half as often as the

general population.

In undertaking its drug review, Veterans Affairs also re-examined the

value of a lesser-known pain reliever, etodolac, and began using it

more. The drug was first sold in 1991 as Lodine by a company that is now

part of GlaxoKline. Like many older medications, however, it was

not extensively tested.

A study based on a review of V.A. patient records that was published

last month in a medical journal, Gastroenterology, found that the rate

of stomach bleeding caused by etodolac was substantially lower than that

caused by naproxen and comparable to that of Vioxx.

Byron Cryer, a researcher at the University of Texas Southwestern

Medical School in Dallas who led the study, said his group was reviewing

those same patient records in effort to determine whether etodolac

increased heart risks. Dr. Cryer said that analysis should be completed

in about a month.

Campen, a medical director at Kaiser, said that his organization

had not yet changed its prescribing guidelines for Celebrex, which was

reported to pose heart risks during a trial of the drug as a cancer

treatment.

The Mayo system, which operates hospitals and clinics in several states,

meanwhile, has decided to reduce its use of the COX-2 drug even further.

Previously, any patient over 60 qualified for Celebrex, Dr. Matteson

said. But earlier this month, a group of Mayo Clinic doctors,

epidemiologists and pharmacists, after reviewing old and new test data,

decided to limit the drug's use to patients at risk of stomach bleeding

or to others with highly specific conditions.

http://www.nytimes.com/2004/12/30/business/30doctors.html

I'll tell you where to go!

Mayo Clinic in Rochester

http://www.mayoclinic.org/rochester

s Hopkins Medicine

http://www.hopkinsmedicine.org