Last week the newswires hummed with reports that doctors from all over America are grossly over-utilizing implantable cardioverter-defibrillators (ICDs), much to the detriment of patients themselves (whose persons are being physically violated by avaricious and/or ignorant physicians), and to the hard-pressed Medicare budget (ICDs being so incredibly expensive).
These reports were based on a study appearing in the Journal of the American Medical Association, which analyzed data from the National Cardiovascular Data Registry (NCDR) in an attempt to determine the proportion of ICD implants in the US which constitute deviations from government guidelines. CNN put it like this: “Of more than 100,000 people who received ICDs, almost 23% did not need them according to evidence-based guidelines.” As the lead investigator of the JAMA study told CNN, “It’s a lot of people who are getting defibrillators who may not need them.”
Specifically the new study shows that a full 22.5% of patients receiving ICDs in the US from 2006 to 2009 received them outside of guidelines sanctioned by CMS. Furthermore, patients receiving “non-evidence-based” ICDs had a significantly higher rate of in-hospital mortality (0.57% vs 0.18%, p<.001), and of post-procedure complications ((3.23% vs 2.41%; p<0.001). Notably, ICDs implanted by non-electrophysiologists were significantly more likely to be non-evidence-based than ICDs implanted by electrophysiologists (24.8% vs. 20.8%).
The lead author stressed this latter point to theHeart.org: “Electrophysiologists — who do these procedures day in and day out and are more likely to be more familiar with the guidelines and the evidence that supports ICD use — were significantly less likely to use these non-evidence-based devices.” And an accompanying editorial in JAMA, also written by a couple of electrophysiologists, says that the results of this study indicate that the “intensive training” which electrophysiologists undergo “may improve both the preoperative evaluation of patients as well as the operative and immediate postoperative care of patients undergoing ICD implantation.”
So this study purports to tell us several things: A) Doctors who implant ICDs are surprisingly poor at following clear-cut, evidence-based guidelines; B) As a result, patients are receiving unnecessary medical devices, and suffering unnecessary harm; and C) At least one mitigation for this problem would be to make sure all ICD implantations are conducted by electrophysiologists. Further, ominously implied in some of the news stories regarding this study is the notion that, perhaps, so clear-cut an abuse ought to be looked into by federal prosecutors, similar to cases we have heard of lately involving the abuse of coronary artery stenting.
To all this, DrRich has a few observations:
1) Guidelines are No Longer Guidelines. “Guidelines” implies, literally, a guide, a signpost, a general set of factors that one ought to take into account when making specific decisions regarding specific individual patients. Guidelines are a strong set of recommendations which (all other things being equal) one ought to follow in the majority of cases, and when one chooses not to follow them, one ought to have a good reason for making that choice.
When the use of clinical guidelines is considered in view of this now-quaint notion, one does not expect 100% compliance. After all, patients being patients, they bring to the table lots and lots of special considerations one ought to take into account when deciding how to apply guidelines. Depending on the level of evidence upon which a certain set of guidelines were established, and considering the array of variations on the mean which patients still insist on bringing to a doctor’s notice, the optimal applicability of a given set of guidelines to a given population of patients ought to look something like a bell-shaped-curve. It is not immediately obvious, for instance, that a rate of compliance with a set of guidelines of 77.5% is simply too low. Indeed, a rate of compliance with your typical clinical guidelines well north of that number might imply, when one fully considers the matter, an abrogation of the physician’s duty to make informed clinical decisions based on ALL available evidence, including those introduced by an individual patient’s specific circumstances.
As a matter of fact, the very guidelines regarding ICDs which doctors are now accused of abusing admit that “the ultimate judgment regarding care of a particular patient must be made by the physician and the patient in light of all of the circumstances presented by that patient.”
In this light, a very striking feature of this new report is its baseline assertion that the strict following of guidelines is “evidence-based” practice, while any deviation is “non-evidence-based;” that is, by implication at least, it is good medicine vs. bad medicine. And so, “only” 77.5% of ICD implanters are practicing good medicine, and that is clearly a major concern – one for which urgent solutions should be sought.
It is one thing for the government to insist that doctors follow their guidelines to the letter, or face fines or worse; it is another for physicians themselves to internalize the same paradigm. Where does that leave patients who are relying on their doctors to use their clinical judgment for their own, individual benefit?
Anyway, guidelines are no longer guidelines; they are directives. Even the doctors agree with this.
2) Most of the Patients Who Received “Non-Evidence-Based ICDs” In This Study Actually Were Indicated For ICDs. News reports of this study, and public pronouncements from the authors themselves, imply that patients in this study who received ICDs outside of the guidelines were getting devices that were unnecessary; that their ICDs, which are deemed “non-evidence-based ICDs,” should never have been implanted. This is a misapprehension.
In the large majority of cases, the deviation from the guidelines was simply in the timing of ICD implantation. Patients received their indicated ICDs earlier than the guidelines specify. CMS guidelines say that patients who are indicated for ICDs should not receive them for 40 days after a heart attack, or three months after the diagnosis of heart failure. Most deviations occurred when patients who were supposed to get ICDs got them during the 40-day (or three-month) window.
So the doctors who violated the guidelines were deciding that, for one reason or another, their patients who needed ICDs would be better off receiving their potentially life-saving devices now rather than two or three months from now.
DrRich will leave aside for now the relatively weak evidence upon which CMS based its recommendation to delay ICD implantation following a heart attack or heart failure diagnosis, and simply assert that it is probably the least evidence-based portion of the ICD guidelines, and in fact, the language in the guideline’s supporting documentation, provided by CMS itself, admits to a certain amount of aribitrariness here. (Perhaps DrRich will discuss in detail in a future post the very “interesting” process CMS followed in establishing these ICD guidelines in the first place.)
But even if you buy the notion that the delay prescribed by the guidelines is fully legitimate, there are still many good reasons one might choose not to wait. Perhaps the patient also needs a pacemaker, and implanting a pacemaker now, and subsequently removing it and replacing it with an ICD (which also functions as a pacemaker) in less than 40 days makes little sense. Perhaps the patient will soon be losing her health insurance (not an uncommon situation these days). Perhaps there are features suggesting that the heart failure is particularly unlikely to improve during the next 3 months. Perhaps there are features that imply that a patient has a particularly high risk of sudden death in the near term.
Whatever. The point is that this study does not show that 22.5% of ICD implants are unnecessary. It shows that sometimes ICDs which everybody agrees are indicated are being implanted a few weeks earlier than the Central Authority would like. The NCDR database the authors used to determine guideline compliance did not allow them to assess the legitimacy of the doctors’ decisions to implant them earlier than CMS prescribes.
Back in the 1990s, when the enlightened idea of “medicine by guidelines” was first being promulgated, it was taken as a basic tenet that, after sufficient training and education had been accomplished regarding a set of guidelines, if deviations from the guidelines still exceeded expectations, then it is necessary to consider that there may be something amiss about the guidelines themselves, and the rationale behind the guidelines ought to be formally revisited. But that was back when guidelines were still guidelines, and not directives.
3) The Important Outcome Is Conspicuously Absent In This Study. One can surmise that the main reason doctors implanted ICDs earlier than the guidelines recommend, 22.5% of the time, is that they thought their patient might experience sudden death during the waiting period. That is, they wanted to protect their patient from sudden death now, instead of two or three months from now. Maybe they were just being obstinate, or stupid, but that was their rationale.
This being the case, the critical information we would want to know is whether the early implantation of ICDs might have led to an overall difference in survival. But alas, that critical information is also not available in the NCDR database. So we know (because the authors were quick to point out) that patients who received “non-evidence-based” ICDs had a worse in-hospital mortality (a difference of roughly 0.4%), and a worse post-procedure complication rate. But what was the difference in survival at, say, one year? Did the early implantation of ICDs increase overall mortality (which is the impression the authors and the newswires leave us with), or did it reduce overall mortality by offering extended protection from sudden death? An overall reduction in mortality was, after all, what the physicians intended when they selected a subset of patients they thought would benefit from not waiting for their ICDs. And it is entirely possible that their decisions did just that.
It seems to DrRich that we might want to know this information, before we castigate too severely (or submit for prosecution) the physicians who judged that “early” ICD implantation would be the best approach in a certain proportion of their ICD-indicated patients.
4) Electrophysiologists Can Be As Self-serving As Anyone Else. This last observation saddens DrRich, himself an exceedingly humble and self-effacing electrophysiologist, as his many thousands of great admirers will attest.
The authors of this study – and the editorialists who wrote in the same issue of JAMA – are all among DrRich’s brethren electrophysiologists. All of them seem to conclude from their analysis that ICDs ought to be implanted by electrophysiologists pretty much exclusively, since we EPs are demonstrated to be (thanks to this study) more likely to follow the guidelines, presumably because we are more “familiar with the guidelines and the evidence that supports ICD use,” by virtue of our “extensive training,” our vast experience, &c.
But once again, the majority of guideline “deviations” which were seen in this study were in the timing of ICD implantation, and not in the fact of ICD implantation. In effect, therefore, the authors are arguing that electrophysiologists are simply better at counting to 40 than those other kinds of doctors. DrRich does not find this a compelling argument for instigating an amendment to the guidelines aimed primarily at protecting the electrophysiologists’ turf.
Furthermore, DrRich suspects that the better compliance with the guidelines evidenced by electrophysiologists has less to do with their superior guideline-following prowess, and more with the fact that there tends to be a built-in delay when EPs implant ICDs. Patients with fresh heart attacks and recent heart failure diagnoses are under the care of non-electrophysiologists (many of whom can implant ICDs themselves, whenever they think they ought to), while patients seen by electrophysiologists generally have to first be referred – a process that introduces a fortuitous delay, and thus, of better guideline “compliance.”
Indeed, when one considers this built-in advantage enjoyed by EPs, one must wonder at the fact that, even for patients implanted by electrophysiologists, nearly 21% still received “non-evidence-based” (i.e., “early”) ICDs. This value, statistically-speaking, may indeed be significantly less than the overall value of 22.5%. But practically speaking it is pretty much the same rate of non-compliance. Which leaves one wondering: Why are electrophysiologists – who suggest that they alone ought to be doing these procedures – themselves so lousy at following the central directives?
Perhaps they, too, need remedial counting lessons. Or perhaps they, with their superior intellect and experience and so forth, actually agree with their non-EP colleagues that delaying ICD implantation in all patients with recent heart attacks and heart failure diagnoses may sometimes (roughly a fifth of the time) be counterproductive.
But no matter. Guidelines are guidelines, which is to say, they’re directives. Following them to the letter is good. “Interpreting” them is bad. It is now apparent that even sophisticated physicians, who should know better, have completely bought in to this new paradigm on guidelines favored by our Central Authorities, and appear less concerned about the implications of this paradigm on their patients and on the practice of medicine, than about how to turn it to their own, narrow advantage.
DrRich hates to be so darned prescient. On January 19, it was revealed that the Department of Justice has launched an investigation regarding ICD implants, as related to “proper guidelines for clinical decision making.” DrRich tells electrophysiologists, and other ICD implanters, what to make of this rather scary prospect, here.