For tax payments, “nudges” have helped municipalities increase revenues and decrease collection-related costs. For energy consumption, “nudges” have helped homeowners save money and utilities preserve capacity.
But in health care, the technique has been slower to catch on.
First described by the pioneering economists Richard Thaler and Cass Sunstein (who is also a legal scholar), a “nudge” is a way of framing a set of choices to essentially steer people toward a particular option without shutting out other options.
Dr. Scott Halpern, a critical care physician at University of Pennsylvania who studies the ethics and effectiveness of nudges in health care, believes the technique can play a greater role in improving the patient experience. This is especially true, he said, for those living with serious illnesses, and who often struggle to make sound decisions at times of great emotional and physical complexity.
Halpern, who is founding director of Penn’s Palliative and Advanced Illness Research Center, spoke with STAT recently by phone, from his office in Philadelphia. This conversation has been condensed and edited.
Can you give us an example of how a nudge might function in your work?
As med students, we are all taught it is important to have conversations about whether patients wanted a DNR (do not resuscitate) order. We’re told that the way to do that is to be neutral – to say something like, “In this situation, your loved one’s heart might stop. If so, would he want us to do chest compressions?” But that places an incredible burden on family members to feel like they have to know exactly what their loved one would want in this specific situation — something they rarely know with confidence. And in fact this isn’t all that neutral anyway — to say no to chest compressions requires giving up something, which is always hard to do.
That strikes me as problematic in cases where chest compressions would almost certainly do more harm than good. So as I developed more experience, I became comfortable saying, “In this situation your loved one’s heart may stop. If it did, we would not routinely do chest compressions, because they would be unlikely to work. Does this seem reasonable?” This way, I’ve set a default option, but I’ve not removed any options. I’ve now used this language several hundred times with the families of patients who were most certainly going to die, and only once has a family chosen CPR. Indeed, several families have thanked me for helping them understand what the norms are.
How commonly are these approaches used?
These ideas are still rarely considered in end-of-life setting, and yet that may be the space where they’re most powerful.
Why would that be?
Because most people only make end-of-life decisions once, and they don’t get feedback about what the alternatives might’ve felt like. We all may have deep-seated preferences about whether we prefer vanilla or chocolate ice cream, because we’ve made that choice hundreds of times and know what each tastes like. But it’s reasonable to posit that patients and family members and even clinicians don’t have deep-seated preferences about end-of-life choices because there’s no way they can be equipped with the same lived experiences. And choices about which we don’t have deep underlying preferences are exactly the ones on which nudges are likely to exert their greatest effects.
Where else in the end-of-life context might be fertile ground for nudges?
Clinicians frequently offer seriously ill patients the option of completing an advance directive, to help establish their goals of care. Most patients end up not doing so, because inertia gets in the way. But framing can help a lot.
The normal way of motivating patients to complete advance directives is by extoling the virtue of being able to control your future care. But by instead helping patients see that by completing an advance directive they’ll reduce decision-making burdens for their loved ones, many more patients will end up doing it because that’s of such great importance to people.
This approach obviously puts more of an onus on clinicians to develop a new skill set, at a time when a lot of them already feel like they’re barely treading water.
It presents a huge responsibility for clinicians, because they’re now in the position of heavily influencing the choices their patients and family members will make. But clinicians already have that responsibility, whether or not they choose to recognize it or not, because there’s always an option that will be listed first, or that exists as the default. So the task for the conscientious clinician isn’t to avoid influencing choices, but rather to avoid restricting choices. And better to influence choice mindfully in a way that likely promotes good outcomes for your patients than to continue doing so haphazardly.
At the same time, some clinicians worry about this possibly representing a return to the old paternalistic approach of medicine.
Right. Clinicians appropriately wonder if something unethical is going on here. If nudges influence choice, how can we justify it? Traditionally, nudges have been justified when they help promote the things people actually want deep down. But as we’ve discussed, in the end-of-life space, it’s hard for patients to know what exact types of medical care will best help them achieve their goals. In such cases, clinicians should rely on a standard that they have historically relied on anyway: the “best interests” standard, where, absent compelling evidence about what a patient would truly want, we should act in a way that we believe — or know, based on evidence — would promote their best interests.