A Conversation With Dr Deepak L. Bhatt

Public Reporting as Public Shaming, and Potential Solutions

Interviewer: Robert A. Harrington, MD; Interviewee: Deepak L. Bhatt, MD, MPH


January 29, 2018

Robert A. Harrington, MD: Hi. This is Bob Harrington from Stanford University, here on theheart.org | Medscape Cardiology. For the past couple of months, I've been struck by conversations I've had with colleagues in the interventional cardiology community about the issue of public reporting of procedural outcomes data. This has certainly been a topic of great interest to those of us practicing in procedural fields, and much of the discussion goes back beyond a decade. It's been interesting that when I talk to my friends outside interventional cardiology, they just accept that public reporting is a good thing to be doing and that cardiology should continue to progress in this field.

The same opinions are not necessarily held by many people in the interventional cardiology community. Not that they want to go back to an era of opaqueness in reporting, but rather they want to think about some of the issues in public reporting. So, I was delighted to see a viewpoint paper from my good friend and colleague, Deepak Bhatt, that appeared in JAMA, called "Taking the 'Public' Out of Public Reporting of Percutaneous Coronary Intervention."[1]

Today I'm thrilled to be joined by Deepak Bhatt, who is the executive director of interventional cardiovascular programs at Brigham and Women's Hospital and a professor at Harvard Medical School. Deepak, thanks for joining me on Medscape Cardiology.

Deepak L. Bhatt, MD, MPH: I'm so excited to be on the Bob Harrington Show. It's my first time, actually.

Dr Harrington: That seems extraordinary, for all of the conversations you and I have had over many years now.

Dr Bhatt: It's probably a good thing that many of those were not on tape.

Dr Harrington: There is probably some truth to that.

What Is the Evidence Behind Public Reporting?

Dr Harrington: Deepak, you heard my introduction about the notion of public reporting. I've been used to debate and dialogue about this in the interventional community, but I've been less accustomed that this is just accepted in many ways without question outside the interventional community. I was really delighted to see your piece in JAMA and thought it was particularly well-written. Let's go through your discussion, starting with evidence behind public reporting.

Dr Bhatt: As an evidence-based sort of guy, I'm glad that you led with that. What is the actual evidence that public reporting of percutaneous coronary intervention (PCI) outcomes is improving anything? In fact, the evidence is really weak to nil.

The preponderance of data really supports the fact that public reporting has not improved outcomes. Instead, it appears to have led to such things as risk aversion. A few good surveys of physicians[2,3] in states with public reporting show that 80%-90% of interventional cardiologists say they personally have avoided particular cases because of fears of public reporting. That 80%-90% is based on anonymous surveys of physicians. The rate is probably 99.99%. I think that cognitive dissonance accounts for the 10%-20% who say they haven't—they cannot bring themselves to admit that they are participating in a flawed system.

The vast majority of interventionists will say that this risk aversion is occurring in those circumstances where public reporting is happening. And the other thing that has happened is, for lack of better terms, upcoding or gamesmanship, where patients are coded as being sicker than they are.[4] A patient whose blood pressure is 89/59 mm Hg and in shock is perhaps excluded from public reporting in states that exclude that type of sicker patient.

Not only has there not been any evidence, or at least any good evidence, that it has improved outcomes, there is actually evidence that it's causing harm. It's the law of unintended consequences.

Transparency in Reporting

Dr Harrington: Let's go back to the notion of public reporting and transparency. I've known you for a long time, and certainly in your clinical trials work and in your clinical research work more broadly, you are someone who believes in transparency. I'm guessing you want to see transparency in the reporting of procedural outcomes, but you are just not convinced, on the basis of the evidence, that what we are currently doing is the way to continue.

Public reporting of decision outcomes and procedural outcomes, in particular, is not [perceived as] public reporting but rather public shaming.

Dr Bhatt: That sums it up well. It's hard to argue against transparency. That would be like arguing against apple pie. You cannot really say, "Transparency is bad. Let's keep it secret." Even if you believe that, you cannot say that sort of thing anymore. And I do not believe that. Transparency as a general principle is good. But like everything, common sense needs to be applied.

Public reporting of decision outcomes, and procedural outcomes in particular, is not public reporting but rather public shaming. At least that is how it's perceived by physicians. It completely ignores human behavior and basic psychological and social principles to think that physicians would not react in that way. Nobody wants to be named in the city newspaper as having a really high mortality rate. That sort of public shaming is unlikely to lead to quality improvement and is particularly unlikely to lead to improvement in actual outcomes. It's more likely to lead to behaviors that keep from being named in the city newspaper, and those behaviors could be destructive. They could be bad for medicine and patients.

As we do with the drugs, devices, and strategies where we insist in cardiology and increasingly in other fields of medicine that there will be evidence to support things, we should do the same with policy interventions. If there is evidence that something is not working, maybe we should reassess whether it's a good policy and maybe we should get rid of it.

Implementing Policies Without Supporting Evidence

Dr Harrington: That last statement on needing more evidence-based policy resonates with me. A number of things have been put upon medicine. For example, the policy on trainee work hours was well-meaning and well-intentioned, but it was not well-studied before implementation. You are saying somewhat of the same thing. Where do we get the evidence? How do we get the evidence? And as we get evidence, how do we react to it? Is our society, including the cardiovascular society, willing to change our opinions on the basis of accumulating evidence about the policy?

Before implementing a policy that...makes for a good sound bite...why not actually study it first and see whether it works?

Dr Bhatt: Absolutely. The road to hell is paved with good intentions. Of course, everyone in the policy world has good intentions, but those who are not physicians—or even physicians not practicing in the trenches—oftentimes do not understand the subtleties behind policy decisions.

Work hours, which you mentioned, is another good example of something never tested that has data sort of arguing against the value of their restriction. Pay-for-performance is another example of something widely implemented without being studied and has mostly negative data as far as whether it works.[5,6] Public reporting of PCI outcomes may be generalized to other [kinds of] public reporting, because public reporting is spreading like wildfire throughout medicine and surgery.

Before implementing a policy that sounds good or makes for a good sound bite on television or in the newspaper, why not actually study it first and see whether it works? Let's approach cases where that has not been done and after the fact were shown to be ineffective or to have legitimate questions as scientists, and look at the data and react in rational ways—not just emotional ways.

'Yelpification' of Society

Dr Harrington: Well said, Deepak. Do patients actually use these data? We are now in a "Yelpification" society. We are getting our Press Ganey patient satisfaction scores. We have star reports on the Web about our practices. Isn't this just a reflection of a societal cry for more of these data that previously had been opaque to the to the patient/consumer?

Dr Bhatt: You are really going to get me going. First of all, I think the different report cards, score cards, and data are contributing to physician burnout. Obviously, that is a problem for physicians, but it's also a problem for hospitals and healthcare systems. It's ultimately a problem for patients. You do not want to be cared for by people who are basically burned out, frustrated, and cannot take it anymore. That is just not a good or sustainable system.

Physician burnout is more than just a personal issue for the physicians involved. It is a larger societal issue. All that reporting is really killing physicians. But to answer your question of whether patients actually want the data, some segments of society do want more data and just want everything listed on the Web. But I think the average person does not use these data and would not necessarily know how to interpret them out of context—particularly a public reporting of primary PCI outcomes. To me, that is almost worthless. You are having an acute myocardial infarction (AMI). What are you going to do? Are you going to get on Google and look up how your hospital is ranked, or the mortality rate of your particular operator whom you do not know? It's just not practical to do those things.

The smartest thing to do that will most improve that person's outcome is to call 911. They should state that they are having whatever chest pain syndrome they are having and get treated in an expedited fashion. That is going to improve their outcomes as opposed to whether you or I are the one doing the primary PCI.

It's a matter of focusing on what is important. Sure, some folks will always want more information. It might matter more for an elective procedure or outcomes that are more patient-reported. But for something time-sensitive, where you need to go to your closest hospital quickly—such as primary PCI—I do not see the value.

Disease-Based Outcomes Reporting

Dr Harrington: Let's dive into that a little bit more. We talked about risk aversion and upcoding, and I'm certainly familiar with your examples: the risk aversion of very sick individuals, and the upcoding of borderline people so that they can either be avoided or risk-adjusted.

You articulated quite nicely in your viewpoint the notion of changing from procedural-based to disease-based outcomes reporting. It gets at this consumer perspective issue. You might not want to look up who your acute angioplasty doctor is going to be, but don't you want places to weed out poor quality or operators who have had bad outcomes?

Dr Bhatt: Great question. I'm a big believer in collecting, examining, and acting on data. That can be done, but public reporting does not have to be part of it. For example, places within the Veterans Affairs (VA) healthcare system report outcomes. I've been on both the delivering and receiving end. A committee of cardiologists from other VAs—not my own—have reviewed my cases and provided feedback, which can be constructive or negative. I've also done that to physicians at other centers. I have found this to be extremely useful. It is educational and something that helps patients, but not in a way that's punitive to the physician and leads to risk aversion or gaming the system.

[D]isease-based reporting could get rid of some risk aversion and shunting of patients to pathways that are not in their best interest....

One way is reporting but not making the reporting public by keeping it within the healthcare system, where there are people who can interpret the data and act on them in a meaningful way. Some may find that entire rationale paternalistic and will say, "If doctors can see the data, then everybody should see the data." I do not happen to agree with that viewpoint. But for those folks who hold that viewpoint, I still think there are ways of public reporting that can be more useful.

You touched upon one. Instead of reporting, in this case, primary PCI-related outcomes, why not report ST-segment myocardial infarction (STEMI) outcomes, or acute coronary syndrome outcomes for a hospital or, even better, an integrated healthcare system? With that, there is less of an incentive to shunt that patient in [cardiogenic] shock and about to die from primary PCI to medical therapy, because they will definitely die. So, it would still "count" as a mortality there. Nothing would be gained by risk aversion and shunting patients from PCI to placebo or medical therapy in the case of shock and STEMI. That type of disease-based reporting could get rid of some risk aversion and shunting of patients to pathways that are not in their best interest, but done to make the numbers look better. So, that is one potential solution.

As well, I mentioned looking at the healthcare system and not just at the hospital, because there is always the possibility of hospitals taking healthy patients but sending sick ones to the referral center downtown. Sometimes sending the complex patient elsewhere would be appropriate, of course, but sometimes it might be done just to avoid the patient who is going to die. Having the integrated healthcare systems metrics reported publicly would probably do more in terms of actually improving care than when reported for single hospitals, and certainly for single operators and single procedures.

Sentiment on Public Reporting

Dr Harrington: You brought up a lot of issues for listeners to consider. For our last topic, you have heard from not just angioplasty docs but from people in other fields, particularly surgeons. Do you want to give the listener a sense of what you have heard?

The feedback I've gotten from the physician and surgeon community has been incredibly unified in their objection to this.

Dr Bhatt: Absolutely. First of all, you mentioned that there are times when interventionalists do not agree on things, but the feedback I've received about this article from interventional cardiologists was uniformly positive. I'm sure there are some who disliked it, but they certainly did not contact me or reach out to me. Beyond interventional cardiologists, cardiologists in general (at least the ones that contacted me) seemed to get it, and we are in widespread agreement.

Moving more broadly to the physician community, I was surprised at how many surgeons reached out to me and said, "You really got it right," and that they feel persecuted by these public reporting programs that are popping up. California is starting to publicly report individual surgeons' outcomes across a variety of different domains within surgery. I get that maybe there are patients who want that information. Maybe it's different if you are talking about elective plastic surgery versus a primary PCI. But again, it would depend what outcomes you are looking at. If you are looking at mortality, the rates are going to be so low that I'm not sure there is immense value in that sort of reporting. If you are looking at more subjective outcomes, I do not know that those are recorded.

The feedback I've gotten from the physician and surgeon community has been incredibly unified in their objection to this. Having said that, I'm sure there are some physicians who are policy wonks out there. I'm sure there are members of the public, the media, and so forth who would hold a very opposite view to the one that I've espoused. But I think what I'm reflecting in this conversation with you is what the majority of physicians in the trenches feel.

Dr Harrington: I'd love to have a broader discussion with one of our policy colleagues, and perhaps I could convince you to come back for a three-way conversation. Thank you for your enormously valuable contribution to the community.

My guest today to talk about public reporting of procedural-based outcomes has been my friend and colleague Deepak Bhatt, the executive director of interventional cardiovascular programs at Brigham and Women's Hospital and a professor of medicine at Harvard Medical School. Deepak, as always, thank you for your insights. Thank you for your commitment to use evidence to answer important questions in the clinical domain and, in this case, the policy debate.

Dr Bhatt: Thanks very much, Bob. I really enjoyed being on your show.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: