When DrRich left his medical practice nearly a decade ago, he spent much of the next few years as a consultant to certain companies that make implantable defibrillators.
Most of his work was in research and development, and had next to nothing to do with defibrillators themselves, or any aspect of treating cardiac arrhythmias. Rather, DrRich was interested in developing physiologic sensors that could be deployed in implantable devices, and the algorithms that could use these sensors to predict and detect various developing medical conditions (so as to enable early intervention, and potentially prevent said medical conditions from becoming manifest). DrRich considered this work a) interesting, and b) representative of a business model that could potentially flourish within a healthcare system whose chief concern is reducing costs.
And whenever the captains of industry who signed his checks would ask him something about implantable defibrillators, usually seeking his opinion on a proposed subtle variation in some unbelievably complicated programming feature, DrRich’s reply was likely to be something like this: “Sir (or Madam) – I will be happy to study the question you pose to me, as I am working on your dime. But it greatly saddens me to see all this time, energy and talent wasted on adding yet more irrelevant features to a mature technology, in pursuit of a business model that is fundamentally broken.”
To which they would smile indulgently, hand DrRich the document describing the proposed changes, and schedule a meeting to discuss them.
The indulgent smile was in recognition of the fact that DrRich never made a secret of his disdain for the business model embraced by implantable defibrillator companies. The fact that these captains of industry put up with DrRich’s disapproval was a clear indicator that they considered it to be “quaint,” and apparently not worth taking seriously, and that the value DrRich provided in other arenas at least counterbalanced the annoyance of having him criticize their core business every chance he got.
DrRich’s disdain for the implantable defibrillator business model was based on two factors.
First, their business model relies on the artificially high prices the system will pay for their devices. DrRich has discussed this before. While these high prices are not directly the fault of the companies themselves (rather, they are fundamentally the fault of Medicare processes that distort and destroy natural market forces), these companies have now come to rely entirely on this artificial price structure, and have established all their business practices around this high-margin enterprise. The problem is that this high-price model absolutely precludes any reasonable penetration of this life-saving technology into the vast population of patients who might benefit from it. Also, because the price structure is not only artificial but arbitrary, a few simple changes in Medicare processes could abruptly destroy their business overnight.
Second, nobody is really interested in preventing sudden death. It’s difficult to sell any product when there’s no demand for that product, and there is no demand for sudden death prevention. In contrast, most other medical problems have a built-in constituency: breast cancer has a constituency; cerebral palsy has a constituency; even erectile dysfunction has a constituency. But sudden death does not. Even though more than 1,000 Americans each day drop dead from cardiac arrhythmias, very few of them are aware they are at risk (until, possibly, that last 5 seconds, which is not enough time to organize a PAC). Sudden death has no constituency.
Furthermore, most people who die suddenly from cardiac arrhythmias are collecting Social Security, and also have underlying cardiac diseases that require ongoing and often expensive treatment. The longer these people live, the more healthcare and Social Security dollars they consume, so their sudden deaths are a boon to our federal budget. Not only is sudden death itself free (that is, it does not require an expensive end-of-life hospitalization), but also it puts a stop to chronic withdrawals of money from the government coffers. Even if implantable defibrillators were given away for free, their use would probably cost society a lot of money. Therefore, not only is there no constituency for it, but preventing sudden death is also bad public policy.
Because the prices paid for implantable defibrillators are artificially and arbitrarily high, and because nobody is really interested in preventing sudden death anyway – not the patients themselves, and certainly not society – the business model is broken. And it’s only a matter of time until it all collapses.
DrRich has proposed an answer to this problem several times to implantable defibrillator companies. But because DrRich’s answer would require a fundamental restructuring of the core business (and because DrRich is so obviously and so quaintly “out there”), he has been ignored.
But now, it would appear, the chickens are coming home to roost.
On September 1, the MADIT-CRT trial was published in the New England Journal of Medicine. This study evaluated the effect of cardiac resynchronization therapy (CRT) in certain patients with relatively “mild” heart failure, and concluded that CRT significantly reduced the episodes of heart failure in these patients.
CRT is a special pacing technique in which the muscle contraction of the left and right ventricles of the heart are coordinated (or synchronized) by a pacemaker that stimulates both ventricles at the same time. Fundamentally, a CRT pacemaker differs from a typical pacemaker in that it has an extra lead attached to the left ventricle (whereas typical pacemakers accomplish ventricular pacing only from the right ventricle). Many patients with heart failure develop a “dyssynchrony” in cardiac contraction, whereby parts of the heart contract at slightly different times. This dyssynchrony reduces the efficiency of the heart, thus making symptoms of heart failure worse, and probably causing the heart failure to progress more rapidly. By pacing both ventricles simultaneously, CRT can re-synchronize heart muscle contraction.
Previous studies have shown that in patients with fairly severe heart failure and signs of dyssynchrony (assessed by measuring the duration of the QRS complex on the electrocardiogram), CRT can significantly reduce symptoms of heart failure. MADIT-CRT was undertaken to assess whether CRT could also help patients with much milder (i.e., earlier) forms of heart failure who have signs of dyssynchrony. Because these patients have relatively few (or no) symptoms of heart failure on a typical day, in order to assess whether CRT was beneficial the investigators used the composite endpoint of death or a non-fatal heart failure “event,” i.e., an episode of heart failure producing significant symptoms and requiring hospital admission or intravenous drug therapy. The study results showed that after about two and a half years, the incidence of heart failure events was significantly reduced with CRT therapy by 41%. In addition, cardiac function itself (i.e., ejection fraction and ventricular volumes) improved in patients receiving CRT, at least suggesting that the worsening of heart failure was slowed with this therapy.
At first glance, even though it was sponsored by Boston Scientific (a maker of implantable defibrillators and CRT devices), MADIT-CRT appears to have been a carefully conducted and rigorously controlled clinical trial, whose results should be taken seriously. While undoubtedly more information will be provided in the next few years as additional data come to light about the patients in the study, we already know that in appropriately selected patients with relatively mild heart failure, CRT therapy can make a significant impact on clinical outcomes. Furthermore, by reducing the requirement for hospitalization, and by potentially slowing the progression of heart failure, CRT at least has the potential of saving money over the long term in patients like the ones enrolled in this clinical trial.
But there’s a problem.
The problem surfaced immediately. Our friend Dr. Wes points it out. In the editorial in the same issue of the New England Journal that reported on MADIT-CRT, Mariell Jessup, MD, from the University of Pennsylvania, while acknowledging that CRT was effective in these patients, warns against adopting expensive new technologies such as this without careful consideration. She asks, “Is this money that can be spent more wisely?” In the September 1 issue of the Wall Street Journal, Douglas P. Zipes, MD (a well-known electrophysiologist, and for decades a strong champion of the implantable defibrillator) is similarly unenthusiastic. He says, “This study may widen the net to capture more patients for (heart resynchronizers). But whether that is appropriate should be questioned.”
Dr. Wes, himself a prominent electrophysiologist as well as a famous blogger, observes, “….the expense of this incredible technology might now be biting us on the hind side,” and notes that the device industry has a serious image problem.
In other words, even the clinical experts we would expect to urge adoption of CRT – a technology that has proved beneficial to patients with heart failure – are now quite circumspect regarding the prospect of expanding its usage. Makers of CRT devices have got to be concerned over this reaction.
To which, DrRich has three observations.
1) You will hear critics of the MADIT-CRT trial complain that this technology is too expensive because “12 patients would have to be treated to prevent one heart failure event” (Jessup). Such analyses, while based on the published data, are premature. MADIT-CRT is yet another clinical trial whose endpoint was determined by a statistical technique known as “group sequential design.” As DrRich understands this technique, it is a way of determining the precise moment when a clinical trial reaches the desired level of statistical significance (either positive or negative), at which point the trial is stopped. Trials are conducted using this technique so as to reach a significant endpoint at the earliest possible time, in order to reduce the extreme cost of conducting a lengthy clinical trial, and get the results out to the public as quickly as possible so you can sell more stuff sooner.
While the group sequential design is statistically legitimate, it does have its drawbacks – especially in a study like MADIT-CRT. The benefits of CRT therapy in the patients enrolled in this study (now that we know there are indeed benefits) will tend to accumulate the longer you observe the patients. That is, if CRT therapy actually slows the progression of heart failure, the differences between patients who receive CRT and those who do not will likely become greater as time goes by. So it is quite likely (though not provable, at least not now) that the actual number of patients you need to treat to prevent a heart failure event would be calculated as substantially fewer than 12, if the trial had been allowed to go on longer, that is, if it had been given a more traditional, arbitrary, time-based end-point (such as 3 or 4 years of follow-up). A more traditional trial design also would likely have improved the cost-effectiveness calculus in favor of CRT therapy.
So, in a possibly short-sighted effort to finish the trial more quickly, the sponsors left themselves wide open to criticism for offering a therapy that, while effective, is “too expensive.”
2) Critics of MADIT-CRT have already focused on the expense of the technology, and are asking whether improving the outcome of patients with heart failure justifies the cost that would be required to do so. This question meets DrRich’s definition of healthcare rationing – intentionally withholding at least some useful medical services from at least some of the people who would benefit from them. And while rationing is an economic imperative and is therefore unavoidable, and while, for this very reason, DrRich has urged a straightforward discussion of healthcare rationing, doing it in this way is not quite what DrRich had in mind. A legitimate discussion of rationing does not entail arbitrarily selecting some specific subset of patients (e.g., heart failure patients) for whom some new therapy offers hope (e.g., CRT), and based on intuition, prejudice, or thinking you can get away with it, decreeing that THESE are the patients, among the entire universe of patients, against whom it is legitimate to ration healthcare.
CRT might well exceed a reasonable cost-effectiveness threshold in a formal system of rationing. But to find out, we would need to decide what that cost-effectiveness threshold is, and to subject all medical services, equally, to the same test. Only in this way can we legitimately decide that CRT is “too expensive.” Arbitrarily discriminating against heart failure patients – or any other group of patients – because one does not like the companies that develop treatments for their conditions (or for any other reason) is fundamentally unethical.
3) While MADIT-CRT is all about the benefits of CRT – that is, the only difference between the two randomized groups in the study is that one group received CRT and the other did not – the unspoken issue when critics complain about the costs of the technology actually has little to do with CRT, and everything to do with implantable defibrillators.
CRT devices fundamentally are just fancy pacemakers. That is, it is a pacemaker with an extra lead (to pace the left ventricle). You can sell one at a profit for a couple of thousand dollars more than you can sell a typical pacemaker – and pacemakers go for something like $3000 – $6000.
But MADIT-CRT did not use a fancy pacemaker. It used a fancy implantable defibrillator – that is, an implantable defibrillator that incorporates CRT pacing instead of standard pacing. So the cost of this CRT device is probably $25,000 – $35,000.
Why were implantable defibrillators used in this study? It’s because patients enrolled in this study already had an indication for an implantable defibrillator. That is, the FDA, CMS, and various clinical guidelines already say that many patients with a history of heart failure and a low ejection fraction – patients like those enrolled in MADIT-CRT – should be considered for the implantable defibrillator.
For this reason, to have conducted this trial with a CRT pacemaker instead of a CRT defibrillator would have been unethical, because “everyone” agrees that these patients should be offered defibrillators.
In the real world, however, relatively few patients like those enrolled in MADIT-CRT actually receive implantable defibrillators. This is because in the real world, nobody cares about preventing sudden death, and in general doctors only do so if somebody with authority makes them, or respected peers shame them into doing it. And you can be sure that nobody in authority is interested in getting doctors to follow this particular clinical guideline. (DrRich would roll over and die if he ever saw a Pay for Performance checklist that includes the requirement to offer implantable defibrillators to indicated patients.)
So a cynic might say that Boston Scientific, when it sponsored MADIT-CRT, did so largely in order to entice (or shame) doctors into finally offering their heart failure patients an implantable defibrillator. For it would be exceedingly difficult to offer one of these patients an implantable CRT device, and still neglect to mention to them that, perhaps, they ought to have one with a defibrillator attached (since the implantation surgery would be nearly identical, and they have a formal indication for a defibrillator).
Whether the sponsor was being cynical or merely ethical, the implantable defibrillator industry is probably going to be very disappointed with the reaction of the medical establishment to the MADIT-CRT trial. Even their natural allies – physicians caring for heart failure patients – are backing away, palms outward and heads a-shaking, in fear of embracing a new indication for CRT therapy that will end up driving up the cost of caring for their already-expensive patients, thus labeling themselves as among of the worst of those money-grubbing specialists the President likes to talk about, and possibly attracting extra-special federal scrutiny. But to be clear, the actual expense they are backpedaling away from is the expense of the already-indicated defibrillator which (to date) they are successfully avoiding, and not of CRT.
And as for those heart failure patients, 50% will still die within five years of their diagnosis, and up to 50% of those deaths will continue to be sudden.
And for that, society can breathe a sigh of relief.
*This blog post was originally published at The Covert Rationing Blog*