Do the Right Thing But circumstances are making it much more difficult | CardioSource WorldNews Interventions

Cover Story | "Quality health care means doing the right thing, at the right time, in the right way, for the right person—and having the best possible results."

— Agency for Healthcare Research and Quality

In a 1986 profile, Edmund D. Pellegrino, MD (1920-2013), the "Father of Biomedical Ethics," described a doctor's moral obligation to competence: "[A doctor] places the well-being of those he presumes to help above his own personal gain. If these two considerations do not shape every medical act and every encounter with the patient, the profession becomes a lie: The physician is a fraud and his whole enterprise undiluted hypocrisy."

Dr. Pellegrino was a long-time professor at Georgetown University and former director of the Kennedy Institute of Ethics. In that same profile in The Washington Post, he noted that when he went to medical school in 1944 medical ethics was taught in about 2 hours. In 2014, that's not enough time to even begin to review all the stresses making it harder and harder today for clinicians to do the right thing for their patients.

Patient Rights—and Wrongs
Among a long list of challenges (all of which are particularly applicable to the interventionalist):

  • Patients not seeking but rather demanding a certain treatment.
  • Cost containment, meant to make medicine "lean," does seem to be making it more "mean."
  • The pace of public reporting is accelerating but, in some cases, is misleading and may be promoting more risk-averse behavior.
  • The increasing complexity of disease and treatment is making it more difficult to communicate important details to patients.
  • The growth of treatment options and the associated "technological imperative" to use them.
  • An aging population, yet we maintain an aversion to address end-of-life issues.

Then there's Karen E. Joynt, MD, MPH, from Brigham and Women's Hospital, who recently noted in an editorial commentary in JACC that cardiovascular medicine has the dubious distinction of being the clinical area where there is the most thorough documentation of racial and ethnic disparities in access to care, as well as sex-based, economic, and geographic disparities.1

Health care professionals are forced into a game of whack-a-mole with the various problems that pop up in a health care universe that's operating in a constant state of flux: a perpetually changing health care climate; an era of shrinking margins, increasing costs, and declining reimbursements; structural changes, down to an organization's DNA, as employment shifts, coalitions change, and "cultural integration" is necessary as long-time competitors suddenly find themselves partners... And all of this operating under a dome of "predatorial bureaucracy" wherein—as just one example—coding errors can threaten careers and reputations. No wonder this period has been called "The Valley of Death," with more creative destruction to come.

No one can be blamed for wanting to throw in their stethoscope; in one survey of more than 5,000 physicians, 43% say they may retire in the next 5 years (23.7% said "very likely" and 19.2% said "somewhat likely" in a survey by The Doctors Company). No surprise that at about the same time, another survey found 45.8% of physicians reporting at least one symptom of burnout.2

Instead of heading for the door, Jeffrey Cavendish, MD, co-director of the cardiac catheterization lab at Scripps Memorial Hospital, La Jolla, California, calls for greater "cardiology fortitude," a term he came up with when teaching fellows in the Navy. He describes it as a strength of mind and character that enables a cardiologist to make tough decisions with compassion and resoluteness, standing up to peer pressure, family pressures, economic pressures, and fear of getting sued. Synonyms, he said, "include backbone, constancy, fiber, grit, grittiness, guts, and courage."

Medicine in general has lost some of that, he said. "We really need to come up with better ways of making these really tough decisions."

Demanding Times
Often, those tough decisions contain an ethical component. At the recent annual meeting of the Society for Cardiovascular Angiography and Interventions (SCAI), Dr. Cavendish participated in a symposium on the Ethical Challenges in Interventional Medicine. One scenario discussed: patient autonomy gives them the right to refuse or accept treatment, but patients (or their families) do not have the right to demand treatment. "That's getting to be a gray area in medicine today where we are seeing patients or their families coming in and demanding treatment," said Dr. Cavendish, even when what they are demanding is not in their best interests. "The family wants everything done, but as physicians we have to step back—and this is where the heart team approach can help."

Referring to the four principles of medical ethics, Dr. Cavendish said physicians should act in the best interest of the patient (beneficence) and "first, do no harm" (non-maleficence), but be aware of the doctrine of "double effect," where a treatment intended for good unintentionally causes harm.

"We deal with that every day in a higher percentage than most professions," he said. A pediatrician, for example, will rarely see a deleterious effect of prescribing antibiotics for an ear infection, but many common interventional procedures have a 1% risk of major adverse effects, such as bleeding, stroke, or myocardial infarction (MI). "This doctrine helps you make difficult decisions about whether actions with double effects can be undertaken. With the aging of the population, we're going to have to have more of these discussions of what to do to treat very elderly patients."

With a growing populace of patients in their 80s, 90s, or even older—recently, Dr. Cavendish's first patient of the day was a 100-year-old male with unstable angina—distinguishing frailty from futility is increasingly important.

Frailty is a distinct syndrome, reflecting a state of decreased physiologic reserve and vulnerability to stressors. It is characterized by a vicious cycle of decreasing muscle mass, energy expenditure, and malnutrition culminating in vulnerability to adverse events. The prevalence of frailty ranges from 20-60% depending on the cardiovascular pathology, as well as on the tool and cutoff chosen to define the syndrome. Epidemiologic studies have consistently demonstrated that frailty carries a greater than 2-fold relative risk for mortality and morbidity across a spectrum of stable cardiovascular disease, acute coronary syndrome, heart failure, and surgical and transcatheter interventions.

Dr. Cavendish recommends a pair of recent papers from JACC and JACC Cardiovascular Interventions providing guidance on frailty assessment in cardiovascular care: "Frailty Assessment in the Cardiovascular Care of Older Adults"3 and "The Impact of Frailty Status on Survival after Transcatheter Aortic Valve Replacement in Older Adults with Severe Aortic Stenosis: A Single-center Experience."4

Leaner and Meaner?
Sometimes the "double effect" refers to a policy with unintended consequences, rather than a treatment. Case in point: one of the greatest demands in health care today is cost containment, said Peter L. Duffy MD, medical director of the cardiovascular service line at FirstHealth of the Carolinas (Pinehurst, North Carolina). Will "lean" lead to too many shortcuts—like fewer labs and less ancillary care—and make health care "mean"? He asked, "If you do reduce resource utilization, are you being diligent about being sure that you are not decreasing quality?"

Similarly, Dr. Duffy asked whether moves to decrease length of stay will lead to decreased patient satisfaction (not desirable in an environment where patient satisfaction is a key quality metric) and increased readmissions. In other words, while there may be some "defects and recalls," he noted, "Will we accept some increased percent of readmission if overall cost goes down?" Or, will the drive for cost control just lead to use of the cheapest valves or lowest-cost drugs that simply won't work for all patients?

"These are ethical issues we are going to have to deal with, especially when we look at costs and discuss them with hospital administrators," said Dr. Duffy.

There may be some help soon regarding cost-related issues, dilemmas, and conundrums. On August 4, 2014, the ACC released a report to mark the 30th anniversary of the first ACC/AHA clinical practice guidelines.5 In it, the authors note that nearly three-fourths of cardiologists believe that clinicians have a responsibility to help society provide optimal care for all patients and that clinical practice guidelines should incorporate available information about cost/resource utilization to educate providers, payers, and patients about the value of various strategies.

That has not been the case, with past guidelines writing committees excluding cost considerations from recommendations. However, the report states, "it is clear that with limited health care resources and rising costs, it is important and appropriate to consider cost, cost effectiveness, and value. Going forward, it is recommended that the COR (class of recommendation) and LOE (level of evidence) criteria summarizing the strength/level and quality of evidence supporting clinical benefit, based on validated evidence assessment tools, be supplemented by an assessment of value."

That may help, but it won't solve all the problems. Dr. Duffy said he is currently fighting a battle for SCAI with the New York Medicaid Program, which is asking for recovery of any payment they made after July 1, 2013, for percutaneous coronary interventions (PCIs) they deem inappropriate by their appropriate use criteria (AUC)—not even the National Cardiovascular Data Registry®'s AUC. Being forced to adhere to local AUC may not be good for your patients. "All of us can think of many examples (of patients) who clearly benefit from a procedure even if it falls under the rubric of 'rarely appropriate' terminology. You might strive to never be the outlier and plan to be in the middle," said Dr. Duffy, but that is not good for your patients.

Another unintended effect: quality metrics that drive action when a moment of inaction may be preferable. Kenneth Rosenfield, MD, a co-chair of the SCAI ethics symposium, said he is "constantly getting heat" from some of his partners at Massachusetts General Hospital who wonder why the cath lab was activated for a specific case of ST-elevation myocardial infarction (STEMI). He said, "I would rather we get activated and come in early, then if you have to stand down you stand down. The problem with that approach is when you are all called in, often at 3 in the morning, and the patient is on the table, you should still take 5 minutes and do a real good assessment of that patient and make the proper decision," regarding how—or even whether—to proceed. Dr. Rosenfield admits he has felt the crush of time and not taken an opportunity, just a few minutes, to think things through for the specific case that activated the cath lab.

Joseph D. Babb, MD, the other co-chair of the SCAI symposium, admits being a little cynical when he says the dominant factor in treating STEMIs today is not the patient on the table but rather the clock on the wall. "That drives our behaviors because we have to treat the chart; door-to-balloon has to be [the priority] otherwise we won't get a platinum-level rating for our hospital," said Dr. Babb. Yes, it's cynical, but he says there is some truth there. "And that's a misplaced priority for us." Of course, time-to-treatment matters, but we should not be playing beat the clock at all costs. "We need to be taking time to assess that patient. The difference of 5 to 10 minutes, I don't care if you don't make the door-to-balloon time; do we really think that 5 extra minutes makes a difference in mortality? Excuse me, show me the data. I think we're missing the boat, but we all get caught up in it. The frenzy is the clock on the wall."

Dr. Duffy added, "This really points out the problem we have of trying to create quality metrics and having them backfire on us to some degree. That's where we have to be really careful with this to not let the pendulum come so far that we can't be good clinicians and thought leaders anymore and good doctors to our patients because we are forced to go by a rule."

The Opaque Nature of Transparency
Another area where that seems replete with unintended consequences: public reporting. Nobody would hire a plumber rated as the worst in the area to fix a clogged pipe, so why would you want your clogged coronary fixed by a "negative outlier"? Sounds clean and simple, but that is not the case. Public reporting is complex and makes many in health care nervous. They secretly wish for it to disappear like a bad dream, but Gregory J. Dehmer, MD, a past president of SCAI, says forget about that: public reporting is here to stay.

Numerous ethical issues are related to public reporting. According to Dr. Dehmer, a professor at Texas A&M University College of Medicine, one real concern is refusing higher-risk cases to avoid being the outlier. If you don't trust the risk adjustment model—and there is no perfect risk adjustment model out there—you may not be willing to take on higher-risk cases. In a recent commentary in JACC Interventions, he wrote: "How many drivers instinctively slow down when they pass a police officer issuing a ticket to another motorist? If your hospital were zapped as an outlier in a public report, would you want to try for two in a row?"7

He was commenting on a paper by Dr. Joynt and her colleagues. Massachusetts has publicly reported risk-adjusted mortality for PCI since 2003.8 During this period, four of the 24 hospitals performing PCI in the state were identified in a public report as outliers because of a risk-adjusted mortality higher than the state average (FIGURE 2). The risk profile of PCI patients at outlier institutions was significantly lower after public identification compared with non-outlier institutions, suggesting that risk-aversive behaviors among PCI operators at outlier institutions may be an unintended consequence of public reporting in in the Bay State.

Dr. Joynt also recently analyzed NCDR data looking at PCI outcomes in New York, where mandatory public reporting of outcomes after PCI has been in effect since 1992. Dr. Joynt said in an interview that public reporting may be an immensely powerful quality improvement tool: "If you know that people are watching, you are going to pay extra close attention and make sure you do the right thing," she said. Conversely, some data suggest that public reporting makes hospitals and providers risk averse. Public reporting, she noted, might "drop the use of PCI but not necessarily for good reasons but because people are a little bit scared about the reporting."

The investigators compared New York data with that of Michigan, which has a record of continuous quality improvement through a partnership with Blue Cross/Blue Shield where hospitals get feedback on their performance but, unlike New York, the data are not publically reported.

What did this tale of two states reveal? Patients undergoing PCI in Michigan had more risk factors and seemed sicker than those undergoing PCI in New York. PCI patients in Michigan were more likely to have cardiogenic shock on presentation or presented following cardiac arrest, suggesting that sicker patients are being taken to the cath lab in Michigan. "Mortality rate was also a little bit higher in Michigan than it was in New York by about half a percent," she said. "Still very, very low rates of mortality overall: just over 1% [in Michigan] versus under 1% [in New York]."

Dr. Dehmer notes that mortality cannot be ignored, but when thrust into a situation where a high-risk PCI is the best (and sometimes only) option for a patient, the outcome may not be good despite the best efforts of the operator and the entire catheterization team. In his JACC commentary, he wrote, "Primum non nocere or 'first, do no harm' is a phrase we heard early in medical training, but we should not be paralyzed daily by fear that one bad outcome from a potentially lifesaving and appropriate procedure will harm a physician's or hospital's reputation. How is the line drawn between a procedure that is truly futile, and another that has a chance, albeit small, of saving a life?"

Putting Patients into the Ethics Equation
John Spertus MD, MPH, University of Missouri, Kansas City, and clinical director of outcomes research at Saint Luke's Mid America Heart Institute, noted, "We're talking about a lot of ethical dilemmas. I think a lot of these dissipate if we actually engage the patients; we start making decisions that are respectful of the patient's preferences and values and not just our own; it takes a lot of the load off of us."

In theory, patients facing medical decisions should be told about their options, have the opportunity to discuss the pros and cons, and have their preferences reflected in the final decision as reported in their medical record. In practice, that is often not how it works. For example, Dr. Spertus said patients are not engaged in decisions regarding stenting. In a 2012 study of nearly 500 PCI Medicare patients, only 10% said they were presented with options to seriously consider (such as drug-eluting stents [DES] vs. bare-metal stents [BMS]). Yes, most (77%) reported talking with doctors about the reasons for stents "a lot" or "some," but few (19%) reported talking about the cons. Only 16% said they were asked about their treatment preferences.9

"No matter how well we think we are explaining the alternatives to our patients," Dr. Spertus said, "they aren't getting it, they aren't hearing it, and they are not walking away from our conversations thinking they are in control."

How this manifests in clinical practice is best demonstrated with discussions relating to DES. They have been a great advance, but create the need for prolonged dual antiplatelet therapy (DAPT) which increases bleeding risk and can decrease quality of life; adding a second antiplatelet agent can be expensive, make future surgery more difficult, and put the patient at risk if DAPT is stopped prematurely. "We should admit there is some equipoise and there is some role for patient preferences and values," said Dr. Spertus.

The benefits of DES depend on the patient. Dr. Spertus and his colleagues developed and validated a model to predict the occurrence of target vessel revascularization (TVR) after PCI, allowing them to quantify the anticipated benefit of PCI with DES compared to BMS using data from more than 27,000 PCI patients. Using variables routinely collected as part of the NCDR CathPCI Registry®, they found a number of significant predictors of TVR including prior percutaneous coronary intervention, emergency or salvage PCI, prior coronary bypass surgery, peripheral vascular disease, diabetes mellitus, and angiographic characteristics. The value of DES versus BMS varied broadly, depending on patient characteristics.

In this large community-based population, the number needed to treat with DES to prevent one TVR compared with BMS ranged from six—in patients deemed high risk—to 80 for lower-risk patients based on specific clinical and angiographic factors.10

That's a huge difference, said Dr. Spertus, especially given that each one of those 80 will be on DAPT with all the hassles, inconvenience, and risk associated with such therapy to prevent a single TVR. As for himself, he said, if offered a choice between BMS and DES, "If I were told only six people needed to be treated [with DES] to prevent one from coming back, I would probably take the DES. But if I was told 80 people like me would have to get a DES and pay for dual antiplatelet therapy in order to prevent one return, I don't think I would take it."

Before PCI is selected, informed consent requires apprising the patients of alternative treatments—the risks and benefits—and engage them in shared decision making. Dr. Spertus underscored that physicians have an "ethical mandate" to improve informed consent. There are a number of prediction models, not just his, and he thinks they are ready for prime time and use in daily clinical care. It's a growing learning system that collects data through registries like the ACC's NCDR CathPCI Registry and creates prediction models currently only used for quality improvement purposes. "What we really need to be doing," he said, "is to take those models and use them at the time of medical decision making so we can improve our patient's decisions and outcomes."

Ultimate Dilemma
We've tackled a number of ethical issues so far, but a recent editorial in JACC addressed "The Ultimate Dilemma: The Disconnect Between the Guidelines, the Appropriate Use Criteria, and Reimbursement Coverage Decisions."11 It's also the ultimate headache.

Here's the gist of the problem: in 2013, the ACC and seven other organizations published AUC for implantable cardioverter-defibrillators (ICDs) and cardiac resynchronization therapy (CRT).12 These criteria were developed to critically review clinical situations that may warrant implantation of an ICD or CRT device.

The writing committee recognized that some of the scenarios deemed "appropriate" or "may be appropriate" were discordant with the clinical requirements of many payers, including the Medicare National Coverage Determination (NCD). To charge Medicare for a procedure not covered by the NCD may be construed as fraud.

As Dr. Fogel and colleagues wrote in their JACC Viewpoint article, discordance among the guidelines, the AUC, and the NCD force physicians to face the difficult dilemma of trying to do the "right thing" for their patients—especially when the "right thing" may not be covered by the payer or insurer.

In addition, the last major revision to the NCD came in 2005, meaning it simply does not address many of the scenarios for primary-prevention ICD use that the AUC authors determined to be appropriate. This may place practitioners and hospitals at risk for denial of payment or investigation for possible abuse or fraud even when the decision was clinically justified.

So what should cardiologists and electrophysiologists do? Dr. Fogel and his coauthors strongly encourage the ACC and its various partners who developed the AUC to continue to advocate for legislation that protects physicians who follow the clinical guidelines or the AUC. "Simply put," they wrote, "a physician who follows the standards of his profession in the best interest of the patient should not be subject to civil or criminal penalties."

However, physicians and their hospitals still have an obligation to understand the current guidelines, be aware of the AUC scenarios, and "practice within the scope of the ICD NCD whenever possible." Conversely, inappropriate practice outside the guidelines and AUC should not be tolerated.

In the short term, if an ICD is considered medically appropriate for an individual patient, yet falls outside of the NCD, clinicians can consider several options. Not perfect options, for sure, but options:

  • The patient may be a candidate to wear an external defibrillator vest until the requisite time period has passed.
  • Alternatively, the hospital may choose not to charge for the device or may contact the Medicare fiscal intermediary to prevent any allegation of fraudulent billing or deception.
  • When there is no good alternative, the patient can be asked to sign an Advanced Beneficiary Notice to acknowledge that the ICD implant may not be covered and he or she may be responsible for the costs of the procedure and the device.

In all cases, when physicians believe that a device is indicated in a situation not covered by the NCD, it is important to use the medical record to clearly document their thought process and rationale.

Cracking the Code
Dr. Fogel and colleagues added that physicians also can protect themselves and their health systems from legal action by becoming involved with "teaching coders about what we do, what constitutes a myocardial infarction [not all troponin elevations represent an MI], and clearly stating in the chart when events occur, such as the diagnosis of heart failure and the initiation of guideline-directed medical therapy."

Columbia University College of Physicians & Surgeons, New York, New York is a case study in how charts and coders can cause problems. The Department of Justice (DOJ) conducted a civil investigation to determine whether some institutions had submitted claims for payment for implantation of ICDs that were not medically indicated or violated the Centers for Medicare and Medicaid payment policy.

Jonathan S. Steinberg, MD, and Suneet Mittal, MD, described their experience and responses to this audit in JACC,13 particularly the complexities and nuances of practicing evidence-based medicine in the setting of heavy regulatory oversight. Both Drs. Steinberg and Mittal are based at Columbia and they direct electrophysiology (EP) programs at neighboring nonteaching facilities. As director of the EP program audited, Dr. Steinberg provided expert assistance to the hospital's administrative and legal teams.

In an interview, Dr. Steinberg said, "It turns out at our institution, when we did a detailed analysis of the patients who were identified by the DOJ, we found that the vast majority actually required or could benefit from the ICD but technically violated the NCD." There were coding errors, too, but overall clinical judgment was strong with only a small number of implants ultimately determined as not indicated (1.3% of all implants and 15% of a targeted subgroup of cases.)

While it was a painful, time-consuming lesson, in sharing their experience, Dr. Steinberg said, "There are clearly a few lessons. One is, as we have heard over and over, documentation is crucial. If your patient falls neatly into any of the [ICD] categories, things will work very smoothly. But as all of us know, [patients] don't fall neatly into one category or another. If that's the case, make sure there is vast documentation in the medical records so it is available to any auditor internal or external to the institution.

"Secondly, you have to make sure your coders are aware of some of these subtleties in coding. You cannot code an acute MI at the same time as you're coding an ICD implantation. Those will set off alarms.

"Lastly, we find there are certain circumstances that actually are just not covered by the NCD. But advances have taken place since the last NCD update, so there are many circumstances that are not accounted for by the NCD and the Department of Justice has acknowledged that there are circumstances that don't quite flow into these guidelines and they will be sensitive to that as long as you have documented it very carefully in your chart."

In other words, do the right thing—and make sure you have the right documentation to back you up.


  1. Joynt KE. J Am Coll Cardiol. 2013;62:2282-3.
  2. Shanafelt TD, Boone S, Tan L, et al. Arch Intern Med. 2012;172:1377-85.
  3. Afilalo J, Alexander KP, Mack MJ, et al. J Am Coll Cardiol. 2014;63:747-62.
  4. Green P, Woglom AE, Genereux P, et al. JACC Cardiovasc Interv. 2012;5:974-81.
  5. Jacobs AK, Anderson JL, Halperin JL, et al. J Am Coll Cardiol. 2014 August 4. [Epub ahead of print]
  6. Dehmer GJ, Drozda JP Jr, Brindis RG, et al. J Am Coll Cardiol. 2014;63:1239-45.
  7. Dehmer GJ. JACC Cardiovasc Interv. 2013;6:631-3.
  8. McCabe JM, Joynt KE, Welt FP, Resnic FS. JACC Cardiovasc Interv. 2013;6:625-30.
  9. Fowler FJ Jr, Gallagher PM, Bynum JP, et al. J Gen Intern Med. 2012;27:911-6.
  10. Yeh RW, Normand SL, Wolf RE, et al. Circulation. 2011;124:1557-64.
  11. Fogel RI, Epstein AE, Mark Estes NA, et al. J Am Coll Cardiol. 2014;63:12-4.
  12. Russo AM, Stainback RF, Bailey SR, et al. J Am Coll Cardiol. 2013;61:1318-68.
  13. Steinberg JS, Mittal S. J Am Coll Cardiol. 2012;59:1270-4.

Keywords: CardioSource WorldNews Interventions

< Back to Listings