New post up at — The ethics of ICD decision-making

I’ve got a good one for you.

Who is the better doctor?

Is it the caregiver who–by whatever means–gets her patient on the best treatment, or, is it the doc who communicates the options most clearly?

I ask because the Institute of Medicine has made shared-decision making (or patient-centered care) a major focus of quality healthcare. That sounds good on paper, right? Medical intervention can be burdensome so it’s important to align it with the patient’s goals for care. And, clearly, patient-centric care exemplifies the first tenet of medical ethics–namely autonomy. Though it’s smart to cycle with a helmet or drive with a seat-belt, not everyone shares the same risk aversion.

Enter the second tier of medical ethics, beneficence, which is defined as doing good. There is scientific evidence that certain medical treatments reduce the risk of death. That’s a good thing. Like riding with a helmet or driving with a seatbelt.

As a heart rhythm doctor who implants internal cardiac defibrillators (ICD), the intersection between patient choice and medical benefit comes up often. Clinical trial data tells us that implanting an ICD in “eligible” patients decreases the risk of death. Medical guidelines have set out these benefits. Quality healthcare is that which provides the most benefit. We even have catchy phrases that proclaim such: Get with the guidelines.

Alas, neither the guideline writers, nor the referring doctor, nor the implanting doctor has to have the device implanted in their chest. We don’t have to live with the risk of infection or shocks. This is why I have always felt (even before the Institute of Medicine proclaimed the virtues of patient-centric care) that patients need to understand the trade-offs of an ICD. These shockers come with plenty of trade-offs.

The conflict that comes up in my mind is that presenting the benefits and burdens of an ICD (or any fateful medical treatment) candidly and honestly may lead some eligible patients to decline an evidence-based therapy. In other words, maximizing autonomy–by aligning care with one’s goals–might compromise beneficence. In the case of an ICD, it might mean a patient dies prematurely. That’s huge.

In preparation for my ICD talk at the Heart Rhythm Society meeting I jotted about 1300 words on the matter. The topic was complex enough for me to have asked for help. The essay includes quotes from five thoughtful medical leaders. The quotes are great.

I hope you want to read more. It’s over at The title and link are:

Autonomy vs beneficence? Shared decision-making in the patient considered for an ICD


P.S. Worry not afibbers, there is a nugget there for you.


  1. The challenge of “autonomy” vs “beneficence” is an ever present one – and one that never gets easy. We are all impressioned by our own experiences – and we each develop our own individual approach to this issue. As a non-interventionalist – mine was colored by an experience nearly 40 years ago when I was a resident – in which I “followed” my attending’s advice and convinced a patient to undergo diagnostic angiography. The patient did not want the procedure. She died on the table …. Thereafter I decided to never be “the convincer” who by persistence “made” a patient undertake an intervention that they did not want …. I also decided to do much better at really listening to what the patient wanted.

    For me – informed decision-making has made my approach to this issue much easier. The challenge resides in separating oneself from the patient. It is not about what I would do if I were the patient – or what I would recommend for my loved ones – even if I know in my heart that all of the world’s medical literature favors a particular approach that I think is best. Instead – it is about me being as certain as I can that the patient truly UNDERSTANDS what medical disorder they have – what the options for treatment are – what the potential interventions are – and what is likely to happen if they do or do not decide to undergo such interventions. It is also about me understanding that patient’s value system – and the reasons why that patient may or may not want to proceed with an intervention that may offer a better chance for increased function or survival.

    The patient doesn’t have to necessarily be sophisticated to understand the above. They may not always clearly express their intentions. My challenge is to assess IF I have succeeded in successfully conveying the above information to the patient – IF I think the patient (and if relevant, the patient’s family) have truly understood what I have conveyed – and lastly, IF the final decision opted for by the patient is one that is made with truly informed consent given all of the above. IF so – then in my opinion autonomy wins out every time over “beneficence”.

    Final Thought: How the above information is presented by clinician to the patient is everything.

    John – GREAT post as always!

  2. The team of bigshots who coerced my husband to receive an unnecessary pacemaker – and installed and programmed it wrong, causing severe complications – made it clear that their idea of shared decision making about ICDs was to send two electros to literally yell at him in two shockingly abusive sessions where he was called stupid and told that our values were wrong because he had found at least enough backbone to refuse an ICD that, we later learned, was explicitly not recommended by guidelines. (He further defied them later by actually getting better.) At one point he feebly attempted to justify his reluctance by saying “What if I might not want to be defibrillated at some point?” I vividly recall the turkey he was talking to sneering, in a tone of utter contempt, “WHY would you EVER not want to be deFIBrillated?” Trying to help out, I chimed in: “What if you had terminal cancer?” and he acknowledged with obvious disgruntlement that that might, just possibly be understandable. Imagine his rage if I had offered the example of an Alzheimer’s diagnosis. My husband at this point would be delighted to die of a clean quick SCD rather than complications from the pacemaker – but he’s determined not to die of an intervention cascade, which means if he should suffer permanent heart failure in the future, there will be no ICD, period. And that’s entirely consistent with beneficence, so long as the good is defined by the patient’s values rather than the doctor’s.

Comments are closed.