Milton Packer on the terrifying power of self-deception In some circles, it is now acceptable for one to dismiss truths in favor of falsehoods that confirm and reinforce a preconceived worldview. Some are inclined to embrace "alternative facts" based on the premise that they can reject traditional authoritative sources as long as a "democratic" process deems them unreliable. Proponents of "alternative facts" argue that, if enough people believe a lie, it deserves a place among confirmed facts in information "marketplace." This is precisely the logic by which evolution deniers insist that "creation science" should (or must) be taught as a reasonable alternative. Like it or not, "alternative facts" drive much of the discourse these days, particularly in politics. We might like to think that the allure of "alternative facts" has not encroached the science of medicine. But sadly, that is not true. Take, for example, the presentation of the results of the CABANA trial at the annual meeting of the Heart Rhythm Society in Boston last week. Here is the background. Atrial fibrillation is an exceptionally common form of abnormal electrical activity affecting the minor chambers of the heart. It accompanies many cardiac conditions and may be intermittent or continuous. In some patients, episodes of atrial fibrillation are quite symptomatic and distressing, but, for most patients, they are not. The consequences of the irregular rhythm can be greatly minimized or even prevented with available (and inexpensive) drugs, as long as the heart does not beat rapidly and sludging of the blood is inhibited with anticoagulants. In the past, physicians tried to prove that elimination of atrial fibrillation with drugs was a good thing, but such trials showed that it was generally appropriate to just allow the rhythm to continue and block its adverse consequences. Accordingly, in clinical practice, most patients with atrial fibrillation do not receive treatments to abolish the rhythm disturbance. However, in recent years, catheter ablation (an invasive procedure that relies on a controlled localized burn of small pieces of the heart) has been used to interrupt the electrical circuits that cause atrial fibrillation. The procedure frequently abolishes the rhythm at first, but the rhythm often recurs. Ablation carries important and serious risks, and it is expensive. What is its financial impact? Catheter ablation is performed by specialized cardiologists, known as electrophysiologists. For most, performing catheter ablations has become a major source of revenue; it is currently estimated to be a $4.5-billion industry. Undoubtedly, if catheter ablations were found to be unhelpful, the annual salaries of the electrophysiologists would plummet. And the revenues of healthcare systems, cardiology group practices and device manufacturers would be adversely affected. From a societal perspective, healthcare expenditures would be reduced -- but, at this point in history, can we honestly claim that we really care about that? So, catheter ablations have become very popular. Those who perform them on a regular basis believe that they are dramatically effective. However, there is a dark secret here. If you look for reliable clinical trial evidence that catheter ablations actually prevent the serious consequences of atrial fibrillation, you won't find any. A $4.5-billion industry has been built on the promise of relieving symptoms in the small minority of patients who find them to be intolerable. To justify the extraordinary increase in the number of ablation procedures, everyone agreed that we needed hard evidence from rigorous clinical trials. Many weeks ago, I reported on the results of one such trial, known as CASTLE-AF, which claimed that catheter ablation reduced the morbidity and mortality of patients with heart failure. The trial was very small and had numerous disqualifying methodological and analytical flaws. Without a doubt, we needed much larger and far more definitive trials. Fortunately, the largest trial ever carried out on the usefulness of ablation was nearing completion. The trial, known as CABANA, enrolled more than 2,200 patients with atrial fibrillation at increased cardiovascular risk. Half of the patients were randomized to receive ablation treatment; the other half to treatment with existing drugs. The leaders of the trial had a specifically defined goal: to determine if there was any benefit of catheter ablation to prevent the occurrence of death, stroke, bleeding or cardiac arrest. They selected this endpoint because it included all the serious clinical events that they thought were important in patients with atrial fibrillation. Throughout the planning and execution of the trial, the investigators stated that they were going to use an analytical procedure known as the intention-to-treat analysis. This approach is exceptionally well-established as the least biased and most appropriate. The intention-to-treat approach is not controversial. It has been the bedrock of clinical trials for 30 years, and it is the undisputed gold standard of how the data from a trial like CABANA should be analyzed and interpreted. And in clinical trials, there are no silver or bronze awards for using invalid methods. Clinical trials have meaning only when they are done right. The intention-to-treat analysis works because it has rules. Its major rule is that all patients who are randomized in the trial must be analyzed according to their assigned group regardless of what happens during the course of follow-up. If you violate the rule, you are going to get invalid or unreliable results. The leaders of the CABANA trial knew this, and they designed the trial properly. They declared from the onset that achieving success on the primary endpoint using an intention-to-treat analysis was the unequivocal requirement for making a claim that ablation was useful. They also knew doing the trial would be challenging. From the start, they anticipated that some patients who were assigned to ablation would not receive it, and some who were assigned to drug therapy might receive ablation. In fact, in their protocol, they specifically stated that 25%-30% of patients assigned to drug therapy would be expected to receive ablation during the trial. That key assumption was built into the design of the study, and they planned to do the intention-to-treat analysis with that knowledge in mind. The investigators received a great deal of public money (from the NIH) based on the premise that they would do a good trial and interpret the results properly. Alas, when the trial was finished and the investigators looked at the data, they were profoundly disappointed. Based on the intention-to-treat analysis, the results did NOT show the expected benefit of catheter ablation. The implications of these results were devastating. Suddenly, a $4.5-billion industry appeared to be at risk. Perhaps more importantly, any dreams that the revenues of electrophysiologists would suddenly soar were dashed. So what did the investigators do? Did they stand in front of their fellow physicians and declare that the trial was a disappointment? Nope. Such a presentation would have been received with horror. So instead, they created "alternative facts." Creating "alternative facts" in a clinical trial is easy to do. You just need to violate the intention-to-treat principle. If you depart from the rules, you are at risk of looking at the data in many biased ways, all of which are known to produce unreliable results. The CABANA investigators presented several "alternative" analyses, which totally undermined the integrity of the trial's randomization process. And surprise, surprise! The "alternative facts" led to the conclusion that catheter ablation worked. Here is an example: In one "alternative analysis," if a patient responded favorably to drugs for many years but received ablation towards the end of the trial, the entire period of benefit on drug therapy was ignored. The other "alternative analyses" that were presented were just as biased. Now for the worst part: The researchers proposed that the "alternative analyses" were the ones to be believed! What reason did the investigators give for showing these "alternative facts"? They said that the alternative analyses were justified because 25%-30% of the medical-therapy group crossed over to ablation. But the investigators knew from the start that that would happen. And knowing that, they stated categorically and properly that success of the trial would be determined by the intention-to-treat analysis. Did they present the results of the intention-to-treat analysis? Yes, they did, for a brief moment, but they rapidly discounted it. Instead, their presentation emphasized the "alternative" analyses and conveyed the message that the trial was a success. Many physicians were persuaded. According to an article in TCTMD, an electrophysiologist said that she was "convinced" by the alternative analyses. Another electrophysiologist agreed: "the message should not be ... that it was a negative trial... . the takeaway on the totality of evidence that we have is that ablation works." And John Day, MD, a former president of the Heart Rhythm Society, claimed in his blog that half of the patients never received their assigned treatment. This was a total fabrication. He opined that the intention-to-treat analysis was inappropriate. And he criticized the media for calling the CABANA trial a negative study. Sounds familiar? Criticizing news sources with a long history of journalistic integrity of promulgating "fake news"? Gee, where have we heard this before? One fact is not in dispute: Self-deception can be amazingly powerful when you have strong preconceptions and financial motivations, and are presented with "alternative facts." Why did the electrophysiologists reject the intention-to-treat approach? The deluded response from some: only "purists" use this approach. It is the approach the "establishment" wants us to use, but we are free-thinkers! We can believe what we want to believe. Down with the intention-to-treat analysis! Let us celebrate our victory! Ignorance is power! At that very moment, the world of electrophysiology changed. But not for the better. Did all electrophysiologists share in this self-deception? No. But the Heart Rhythm Society never gave them a forum to speak. In clinical trials, we have a term that we use to describe an intention-to-treat analysis. That word is "integrity." We also have another term that we use to describe "alternative analyses." That word is "opportunism." As Upton Sinclair famously wrote: "It is difficult to get a man to understand something, when his salary depends on his not understanding it." For the record, I want to make sure that my message is clear. Many drugs used for atrial fibrillation are suboptimal, and some are awful. These drugs need to be taken daily for years, and they can have important side effects. So if I personally had symptomatically troubling atrial fibrillation, I would probably decide to undergo catheter ablation if initial drug therapy was unsatisfactory. However, the point of this post is not whether patients should or should not undergo ablation. Its entire point is to ask whether it is proper for electrophysiologists to distort and deny the facts to support a preconceived notion. Why not be honest? They could simply say that patients are likely to prefer ablation, even though there is no evidence that ablation prevents any of the serious consequences of atrial fibrillation. Distorting and denying the truth is not a morally defensible position. "Alternative facts" should be rejected, whether they exist in politics or medicine. Do you agree? I want to hear from you. Source
