Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

May 12, 2009
By: Hilzoy

Comparative Effectiveness Research

Reading this post by Merrill Goozner (it's very good) reminded me that I meant to write about the articles on comparative effectiveness research in the recent New England Journal of Medicine. One, by Jerry Avorn, concerns the backlash against CER:

"The contested provisions were designed to support studies comparing the efficacy and safety (and, by extension, the cost-effectiveness) of alternative ways of addressing common clinical problems. Interventions to be evaluated will include pharmaceuticals, devices, procedures, and diagnostic approaches, such as imaging studies. This research will fill important information gaps facing clinicians, patients, and payers concerning what works best. Currently, the Food and Drug Administration (FDA) often approves new medications on the basis of modest-sized studies involving patients with relatively few coexisting conditions who are followed for brief periods. Sometimes the only efficacy requirement is a demonstration that a new product works better than placebo in improving a surrogate outcome measure, such as a laboratory-test result, rather than achievement of an actual clinical benefit. The bar is set even lower for medical devices such as pacemakers and implantable defibrillators, which may only have to be shown to be similar to previously approved products or simply not to be dangerous. For new surgical procedures or imaging studies, there may be almost no evidentiary bar at all.

Vigorous marketing of the costliest new approaches fills this informational vacuum, encouraging the widespread use of goods or services that may be no better, less safe, or more costly than usual care -- or all of the above. Of course, many new interventions clearly are better in one or more of these domains, but we have no systematic way of collecting or disseminating such information. It is these lacunae that the funding for comparative-effectiveness studies was designed to help fill. At 1/20 of 1% of our $2 trillion annual health care expenditure, the CER funding amounts to a fraction of what any corporation would spend to find out whether it was getting its money's worth from its purchases. It represents one of the best investments we can make to edge the health care system away from the fiscal catastrophe it faces, since such studies will help to reduce spending on poorer clinical decisions and to spare resources for expenditures that will help patients most (and most affordably). This research is a public good, like highways and clean air. The private sector is no more likely to identify badly mispriced or potentially toxic treatments than it was to spot badly mispriced or potentially toxic products of the banking industry. (...)

In calmer times, fiscal conservatives might have been expected to support a plan to generate information about treatment benefits, risks, and costs so that physicians, consumers, and payers could use this knowledge in making purchasing decisions. But these are not normal times. On January 23, Representative Tom Price (R-GA), a physician, sent out an "alert" through the Republican Study Committee, falsely warning that the CER legislation would create "a permanent government rationing board prescribing care instead of doctors and patients." The true intent of the CER provision, Price warned, was "to enable the government to ration care" (emphases in original). "Every policy and standard will be decided by this board and would be the law of the land for every doctor, drug company, hospital, and health insurance plan."

Parallel arguments appeared in a letter sent January 26 to several influential members of Congress, cosigned by more than 60 advocacy groups, and again in a January 29 editorial in the Wall Street Journal. In an op-ed by columnist George Will that appeared in the Washington Post the same day, CER had morphed from a form of research into an imaginary new federal body with broad powers. Will named the agency "the CER" and claimed that with such a system, "Congress could restrict the tax exclusion for private health insurance to 'insurance that complies with the Board's recommendation.' The CER," he went on, "which would dramatically advance government control -- and rationing -- of health care, should be thoroughly debated, not stealthily created in the name of 'stimulus.'" In fact, unaffordability rations care far more than comparative studies ever could."

You'd think that doing research to figure out which treatments are most effective would be an obviously good thing. But no: it is, apparently, the first step on the road to socialized medicine. A lot of the attacks rely on this "first step" argument. For instance, the Heritage Foundation wrote that "The type of information collected by CER could eventually be used inappropriately if a "Federal Health Board" was created to decide which types of treatment would be available to whom and when."

It could be used to do bad things! At least, if a board that doesn't exist were created and told to use this information! Pass me my smelling salts. I await with eager anticipation the Heritage Foundation's realization that this very same logic could be used to ban guns: after all, they too can be used to do very bad things, and (unlike comparative effectiveness research) actually are so used. Do you think consistency will oblige Heritage to come out in favor of a ban on all guns? Me neither.

But the Heritage Foundation is a marvel of sanity and good sense compared to John Griffing in the American Thinker, who describes the language providing for CER as "a line that would sentence millions of people to death", and adds, by way of explanation: "If you are picturing Germany circa 1930, you're right on. With the passing of this bill, government, not doctors, will decide who receives care and who doesn't, in essence, who lives and who dies." Deacon for Life, for his part, calls it "Mengele-esque". The idea that Hitler and Mengele's great sin was conducting research into the comparative effectiveness of various medical treatments is, shall we say, peculiar.

More seriously, there is something about the arguments against CER that I have never understood. The opponents of CER claim that it will inevitably be used to make decisions about care. Insurers will not want to pay for care that is not effective, and so people will be deprived of the care they need. But notice what "deprived of care" means here. No one is seriously proposing to make it illegal to purchase whatever medical care you want on your own.

This means that even if your insurance company decides that it will not pay for some treatment that has been shown to be ineffective, you will, under any proposal being seriously considered, still be able to get that care; you just won't be able to get someone else to pay for it. If not having someone else pay for your medical care counts as being "deprived of care", then 46 million people are being deprived of care even as we speak -- and that's just the uninsured; it doesn't include people who have insurance that doesn't cover the treatments they need. And yet, strange to say, the opponents of CER generally do not see this as a problem.

Moreover, once you notice that what the opponents of CER describe as "being deprived of care" just consists in someone's deciding not to pay for some treatment, the idea that decisions about who gets what treatment are currently made by your physician is true only if you pay for your care out of your own pocket. If, like most of us, you rely on medical insurance, then someone other than your doctor is already making decisions about your care. All CER would do is allow this person to do so on the basis of actual knowledge about what works and what doesn't.

Hilzoy 12:31 AM Permalink | Trackbacks | Comments (24)

Bookmark and Share

I might worry if insurances start saying their cost of one treatment means they're willing to let the patient that treatment doesn't work for suffer...

...But I think not knowing is worse than that worry. It's not like the insurances don't already choose the medicine maker instead of basing it upon what works for the patient.

It's really arguing that we shouldn't know what's good for us because they already choose based upon their pocket rather than what works.

Posted by: Crissa on May 12, 2009 at 1:21 AM | PERMALINK

Don't understand why the Democrats don't counter the anti-CER crap by asking opponents why they think it is a good idea NOT to know what works best.

Posted by: dcsjusie- on May 12, 2009 at 3:28 AM | PERMALINK

"46 million people are being deprived of care even as we speak -- and that's just the uninsured"

This number is fake. It has been repeatedly debunked. For example, here: http://www.nytimes.com/2007/11/04/business/04view.html?ex=1351828800&en=7ebf86b6773f35bd&ei=5090&partner=rssuserland&emc=rss)

Hilzoy knows enough about this topic to know that this number is a lie, yet here it is again.

Posted by: a on May 12, 2009 at 3:45 AM | PERMALINK

It is important that CER isn't used to promote more costly therapy that is marginally better.

I remember about 5 years ago there was a massive testing program to compare various drugs for depression treatment. Some people complained that the purpose of the study was to show that the newest and most expensive drugs were better than those that had gone generic and should become the 'offical preferred treatment'.

As I recall the study wound up for the most part not finding significant differences.

But what if it had? Say drug A is 10% better than drug B. Would that have been enough to say that 'good medicine' uses A which costs $600 a month as opposed to B which costs $10?

Posted by: MonkeyBoy on May 12, 2009 at 4:55 AM | PERMALINK

The type of information collected by CER could eventually be used inappropriately if a "Federal Health Board" was created to decide which types of treatment would be available to whom and when." - Heritage Foundation

I have to agree with this. The FDA during the Bush administration oretty much proves that point, doesn't it?

Posted by: Danp on May 12, 2009 at 4:58 AM | PERMALINK

"Germany circa 1930"...?

Chancellor Bruening was rationing health care?

Posted by: Severian on May 12, 2009 at 7:48 AM | PERMALINK

a's post at 3:45am is the lie. The reference is to a puff piece written by N. Gregory Mankiw and published in the New York times in 2007. Mankiw was previously an adviser to President Bush and was at the time advising Mitt Romney in Romney's campaign for the Republican presidential nomination.

The short blurb published by the NY Times was nothing more than a piece of campaign propaganda by another anti-health care conservative catering to the base. There is nothing in the article that supports the source of the numbers Mankiw threw around. The most likely source was Mankiw's anal orifice.

a is practicing what the Greeks recognized to be sophistry. Throw out an assertion that can be quickly stated, and takes time and effort to rebut. The rebuttal of such garbage is of sufficient difficulty that most do not do so, so the assertion remains in the public mind. But it is in this case the utter lie.

Posted by: Rick B on May 12, 2009 at 8:11 AM | PERMALINK

Yesterday the Commonwealth Fund published a new study which demonstrates that 70% of all women in the US forgo needed (essential, primary and preventive) healthcare due to cost or accessibility.

Yet, primary and preventive care isn't even on the table for healthcare reform discussions. The only seats at the table are occupied by the big money Republican funders: pharma, hospitals, physicians and insurers. Patients and nurses (3 million and the providers of over 95% of ALL healthcare services - The Commonwealth Fund)aren't represented in any meaningful way. The definitions of even what reform is are ALL Republican talking points and for-profit healthcare marketing terms.

Meanwhile, people are suffering and dying every day - to the tune of thousands - of entirely preventable problems.

This time, the bus isn't running over people - it's careening right off the cliff, and the US will not be able to recover from the lethal sham and scam that is being perpetrated on its citizenry.

Posted by: Annie on May 12, 2009 at 8:35 AM | PERMALINK

The rebuttal of such garbage is of sufficient difficulty that most do not do so, so the assertion remains in the public mind.

Thanks for the rebuttal, Rick B

As for CER, how brain dead does one have to be to argue knowledge is a bad thing. I guess that's why there's a GOP. Who else would stoop so low to help the millionaire crooks of the health care industry? Besides, we already have insurance companies deciding what is usual and customary medical practice for us. Now it will just be backed up by science.

Posted by: about time on May 12, 2009 at 9:22 AM | PERMALINK

Hilzoy knows enough about this topic to know that this number is a lie, yet here it is again

Wow, an article from Greg Mankiw, Harvard economist who's noted for:

1) Being Chairman of the Council of Economists under Bush as it contracted

2) Saying how fantastic the economy was under Bush

3) Not foreseeing a recession and indeed DENYING it when it started

4) Arguing in favor of sending American jobs overseas as "better" for the American economy

5) Arguing that we don't need to worry about the loss of our manufacturing base because when you make a hamburger at a fast food restaurant, you are really "manufacturing" it.

A worse spokesperson you could not have found.

But getting to the meat of his criticism, the 47 million number is arbitrary. It may include illegals and people who could qualify for Medicare but haven't signed up. The reality is, from a standpoint of measuring Americans with no or insufficient coverage, 47 million is far too low, probably by a half or a third. Over 60% of bankruptcies arise from medical bills, even by those with health insurance -- it's just that their coverage is insufficient to protect them from being financially destroyed for life by ending up in the hospital and need to be treated. On the flip side hospitals are laboring under the costs of the uninsured hitting their emergency rooms, or Canadian emergency rooms in the case of those living close to the border.

The reality is that tens of millions -- maybe not 47 million depending on who you measure -- have no insurance at all and can't get it, and that probably close to one in four have insurance that is insufficient to address their needs and/or protect them from financial ruin if they are injured or ill. And as health care costs spiral upward and coverage and employment rapidly diminish, those figures are getting more dismal.


Posted by: trex on May 12, 2009 at 9:26 AM | PERMALINK

[Ask] opponents why they think it is a good idea NOT to know what works best.

My thoughts too. I can't come up with a justification other than that ignorance makes it easy for someone to profit by selling something that would otherwise be recognized as ineffective or unneeded, or simply a much more expensive alternative to another equally-effective choice.

Posted by: Fleas direct the era on May 12, 2009 at 10:16 AM | PERMALINK

From the Mankiew piece:

"The 47 million also includes many who could buy insurance but haven’t. The Census Bureau reports that 18 million of the uninsured have annual household income of more than $50,000, which puts them in the top half of the income distribution. About a quarter of the uninsured have been offered employer-provided insurance but declined coverage."

Noting that 18 million uninsured people have above-average income does not demonstrate they could have bought insurance. Indeed, it suggests that insurance is too expensive, at least in some cases, for those with means. A similar argument applies to those who refused employer-provided insurance (partially subsidized would be a more apt descrition).

Posted by: dob on May 12, 2009 at 10:45 AM | PERMALINK

@ a: Have you noticed that 'a' and 'asshole' both start with 'a'? There could be a reason. May I suggest that you navel gaze on the possibilities? And, please, take your b.s. elsewhere. You can eat it all you want, but don't expect those of us who live in reality to join you. Enjoy your meal!

Posted by: Get Real on May 12, 2009 at 11:09 AM | PERMALINK

Hilzoy, you might find it encouraging that the theme of this year's HMO Research Network annual meeting was comparative effectiveness research, with the feds present in large numbers telling us how much money they had to help with this research:

Posted by: snow on May 12, 2009 at 11:23 AM | PERMALINK

>>> Vigorous marketing of the costliest new approaches...

I know I always feel better when I recall that the primary source for continuing medical education for my doc is the pharmaceutical manufacturer, their reps and marketers. If you haven't read this, you should google "pharmaceutical companies spend more on marketing than they spend on research and development."

Posted by: Jeany on May 12, 2009 at 11:27 AM | PERMALINK

I have allergic asthma. I have been in several studies of new asthma medications. They were carefully monitored and I felt reasonably confident that I would not be harmed. And in any case I was hopefully helping advance treatments that would eventually help me as an asthma patient.

Since I have allergies I also volunteered for allergy studies, but was always turned down because I had asthma.
What that means, of course, is that when new allergy treatments become available, and since most people with asthma have allergies, that a sizable percentage of the patients using that treatment have never been tested to see how that treatment interacts with their asthma or their asthma medication.

This true of most new meds and it ends up killing people. Just another example of the drug industry using patients as guinea pigs. Maiming and killing patients is more cost effective, to them, than good research.

Posted by: Marnie on May 12, 2009 at 11:37 AM | PERMALINK

@ Marnie
Two problems I find with your note:
This statement "Maiming and killing patients is more cost effective, to them, than good research." is embarrassingly overblown.

Second, while most people with asthma have allergies, the converse is not true - most people with allergies do not have asthma. And the thinking is, with a fair amount of evidence behind it, that allergy sufferers who have asthma need a different type of treatment.

Now, I expect accusations of being a pharma shill to rain down on my head, and I'm probably not even going to read them through, let alone bother to answer. So don't bother.

Posted by: Snow on May 12, 2009 at 12:39 PM | PERMALINK

Snow, one of the reasons CER is so important is that it would look at how medications work in cases where patients have multiple conditions. Many/most studies look at 'otherwise healthy' patients (so in the case of an allergy treatment they will not look at patients with asthma, heart conditions, ...). This isn't because they think other medications will be made for people with multiple conditions, but because it makes the results cleaner (they don't have to worry about all the possible interactions) and cheaper (if the subjects are otherwise healthy then fewer will drop out and less groups will be needed).

Posted by: JohnL on May 12, 2009 at 1:02 PM | PERMALINK

I feel like the conservative argument would be stronger if they reframed it. The theoretical potential risk of CER isn't rationing care in general (that already happens), but insurance companies refusing to cover treatements that *are* effective, but only for a small minority of patients.

But that just means we need to advance personalized medicine to where we can handle such things.

Posted by: Jon on May 12, 2009 at 1:36 PM | PERMALINK

I'm in the unenviable position of having a complex constellation of problems that requires a trial-and-error approach to treating. It makes me feel like a dupe sometimes to pop a pill thinking, "Hey, whatever works works." Comparitive effectiveness is all well and good when it's safe for you to take any drug on the list. But when you can't safely take the drugs specifically approved for symptom x, you're in off-label-land. Your doctor thinks that Expensive Drug A might work, but it certainly won't hurt you. Drugs B, C and D would work very well, but will harm you in other ways.

Posted by: Sara Anderson on May 12, 2009 at 1:44 PM | PERMALINK

Insurers will not want to pay for care that is not effective, and so people will be deprived of the care they need.

If the care that your insurer is refusing to pay for is ineffective for your condition (or no more effective than a cheaper alternative care option), isn't it by definition *not* care you need? The true danger is not that people will be deprived of the care they need, but that they will be deprived of the care they don't need. Man the barricades!

Posted by: chris on May 12, 2009 at 3:08 PM | PERMALINK

My point, chris, was that they were effectively doing this already. For instance, the maker of of Premarin got deals with many insurance companies so they would only cover their version of estrogen - an extract from horses - instead of lab created (and available as non-name brand) Estradiol.

My allergy medicine, for instance, fexofenadine isn't covered, because they'd rather I take an over the counter med, which doesn't work for me, hence having a prescription.

While I personally know about this one, this happens in thousands of medications and treatments, every day, without an independent study to determine if the choices are made by science at all. I think we ought to have the independent study.

Posted by: Crissa on May 12, 2009 at 4:18 PM | PERMALINK

Overlooked by everyone is that CER will just perpetuate the medical, allopathic model that got us into this health care abyss,

Note that the NEJM article says that CER will be used for "pharmaceuticals, devices, procedures, and diagnostic approaches".
Nothing about diet, exercise, supplements,mind body therapies, hydrotherapies and so forth. Only actions that generate big profits at high costs and that are controlled by hospitals, equipment manufacturers, conventional M.D.'s and Big Pharma will be evaluated. And you expect significantly better outcomes at lower costs from this approach?

How could you all be so blind to this fundamental scam?

Posted by: Herb on May 13, 2009 at 8:41 AM | PERMALINK

I'm somewhere between intrigued and amused at how Big Pharma has tied itself in knots on this one. On the one hand, they're on the record as supporting CER when it comes to, say, comparing drug therapy to certain surgical interventions. Some manufacturers would even stand to benefit from research that would show the cost effectiveness of routinely overprescribing certain very safe drugs (e.g. statins and SSRI's). On the other hand, cost effectiveness research comparing new, patented drugs to older, generic ones threatens the foundation of their business model.

My prediction: Big Pharma will come out in favor of a "compromise" that promotes CER, but only in ways that permit drug companies to control the research agenda.

Posted by: pain perdu on May 13, 2009 at 8:53 AM | PERMALINK



Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM

buy from Amazon and
support the Monthly