EP103: Dr. Mitesh Patel on the intersection of healthcare and behavioral economics

Dr. Mitesh Patel is a physician and behavioral scientist! He helped found the Penn Medicine Nudge Unit, a first for healthcare. We talk about behavioral nudges, both for healthcare and home life, and the small things that we can each do to nudge behavior. Dr. Mitesh Patel MD MBA is a physician executive and behavioral scientist. He is currently Chief Clinical Transformation Officer at Ascension , the largest non-profit healthcare system in the US. He works on strategic vision and implementation of programs to transform clinical care using behavior nudges, EHR Apps, machine learning, artificial intelligence, remote patient monitoring, and virtual care. Previously, he was the Founding Director of the Penn Medicine Nudge Unit, the world’s first behavioral design team embedded within health system. 

Transcript

Copy Transcript Copied

Christine Ko: [00:00:00] Welcome back to SEE HEAR FEEL. Today I have the honor of being with Dr. Mitesh Patel who has an MD, MBA, and he's a physician executive and behavioral scientist. (So of course I love that.) He is currently chief clinical transformation officer at Ascension, a large nonprofit healthcare system in the US. He works on strategic vision and implementation of programs to transform clinical care using behavioral nudges, electronic health record apps, machine learning, artificial intelligence, remote patient monitoring, and virtual care. Previously, he was the founding director of the Penn Medicine Nudge Unit, the world's first behavioral design team embedded within a health system. So welcome to Mitesh.

Mitesh Patel: Thank you for having me here.

Christine Ko: Could you first share a personal anecdote?

Mitesh Patel: Sure, outside of doing all the work and the behavioral science and working in the health system, I'm a big sports fan. When I was in high school, I used to be a soccer player, and then I got recruited to be the [00:01:00] kicker on the football team, and my claim to fame is that before I went to college, I could kick a 50 yard field goal back in the day. 

Christine Ko: Wow, that's impressive. 

Mitesh Patel: I don't know if I could still do that anymore, but I still remain competitive and I'm a big kind of football fan and sports enthusiast. 

Christine Ko: Sports can have a lot of analogies to teach maybe us in healthcare how to do things a little better, but I'll have you talk about that. With your experience as a soccer player, football player, physician, behavioral scientist, can you talk about metacognition and behavioral nudges?

Mitesh Patel: I think generally, as I understand metacognition, the idea of being able to recognize yourself and be more aware of what you're doing and your behaviors. I think a lot about behavioral nudges and in two senses, one is this idea of choice architecture or the environment that you're in. There've been a lot of things around, like, where do you place food in a cafeteria that will influence what you end up picking or choosing? But you can also think about in your daily life, or as it relates to health care, what's the order of the list in the electronic health record? Is it [00:02:00] alphabetical? And if it is alphabetical, is the best drug really start with an A? Or is it more likely to start with a W? And should you change the list if it starts lower on the list? The other aspect, I think, of behavioral nudges is in the way we frame information. There's lots of information, and this can apply to healthcare or your daily life, and whether you're a clinician and you're in the electronic health portal or you're a patient and you're just trying to understand what it means when you get a new diagnosis or you're being asked to get some sort of cancer screening....

The way we frame information, how we deliver it, when we deliver it is really important. And I think we spend so much time just getting the information out there. Making sure that everybody has all the choices available to them and spend a little bit more time thinking about how the design of those choices or that information can influence our behavior.

Christine Ko: Yes. I think part of the problem, at least in health care, for doctors and therefore patients, is there's so little time in health care. To really communicate effectively and think about how to [00:03:00] get a message across in the best way is difficult, I think.

Mitesh Patel: Yeah, I think a lot of health care interactions often seemed rushed. And there's actually this concept of decision fatigue that when I was at Penn, we studied. We looked at vaccination, cancer screening by the time of day that you saw your doctor. And what we would find is that at 9 a. m., cancer screening/ vaccination rates start at a certain point, but then slowly throughout the morning, as you get tired of giving the spiel to patients or patients get tired of making decisions, that rate goes down. And we actually followed patients a year out and found that if you are less likely to get offered cancer screening or have it ordered during the visit, you're less likely to have it a year later. But what we saw was something interesting that after lunch there was a slight bump up, meaning that, clinicians who are like, it's 11 45, they're running behind schedule, they have four patients to see between noon, they might try to go faster.

And then you get a little bit of a break and then at one o'clock you might start again. And for patients the same way you might get a bite to eat, you might get recharged and refreshed. What we find is the [00:04:00] bump up doesn't get you quite back to where you were at eight or nine in the morning, and it continues to decline.

So patients who saw their doctors at four or 5 p. m. in the afternoon had the lowest rates of preventative care. And actually there's been new studies come out showing they have higher rates of opioids and inappropriate antibiotics and things of that nature. Really relate this to decision fatigue, both clinicians and patients, and then rushed visits, which in health care is a huge challenge because we're always trying to get more in with less time.

Christine Ko: Yeah. So with your study, did you, or maybe afterward with a different study, could you ever find a way around that?

Mitesh Patel: That's an interesting question. Well, we did a couple of things. We introduced a nudge, that's what my team would do. And so what we did is we tried to introduce a nudge in the electronic health record that would prompt a clinician to ask the patient about cancer screening or vaccination. We heard from clinicians as they liked the prompt, but it's just another thing they have to do in all of their time with all the noise. So instead, what we did is we prompt the medical assistant. The medical assistant in the practice was [00:05:00] the person that got the patient from the waiting room, took their weight and their height and their blood pressure.

And what we did is we had the practices train the medical assistant to say, Hey, you're due for breast cancer screening, a colorectal cancer screening, or a flu shot. We're going to go ahead and pend an order for you. You can talk to your doctor if you have questions, or if you're just ready to get it, you can just let them know and they'll go ahead and sign it.

And so what that did is it removed the noise from the clinician, had made the default so that the patient already had it ordered, and prompted the conversation because you either have to sign that order or you have to remove it. And we found that boosted vaccination cancer screening by sometimes up to 20%.

Now, going back to this thought about time of day and decision fatigue, while it boosted vaccination by 20%, we found there was still a steady decline during the course of the day. So it didn't actually impact decision fatigue, but in the group that got the nudge at 4pm, they were doing just as well as 8 a m. for the group that didn't get the nudge. And so it was as if we almost reset the baseline, [00:06:00] but we hadn't completely solved the problem. So there's more work to be done there. 

Christine Ko: Yeah, that's nice. I had a bunch of medical appointments this past year. And it goes along with that because what happened is I would go to appointment and then they would say, okay you need to go see like cardiology or something. And they would go ahead and make that appointment for me. It wasn't like here's the number, call cardiology. And so then I could, if it wasn't the right time or something, I could always call and change it.

But I realized there was this sort of psychological thing for me where it emphasized to me, this is important enough to my doctor, the healthcare system to go ahead and schedule this appointment for me. If the time really wasn't convenient, I would still just call to change it, but I realized that probably I would be less compliant almost with that order or recommendation to go see a cardiologist if the appointment hadn't just been made for me.

Mitesh Patel: Yeah, I think part of the challenge of helping people to do what they want to do anyways is just removing all the friction and the [00:07:00] work. And so making the appointment for you is the easiest thing.

Being able to schedule an appointment online with a couple of clicks is maybe the next level. Having to call a phone number or figure out where the phone number is, wait on hold for 10 or 15 minutes, then go back and forth with: they give you one appointment, you look at your calendar and you're like, it doesn't work, they give you another appointment... it's just a really challenging process. I think there's a lot of opportunities in health care. We call this removing sludge. Sludge is a term coined by Cass Sunstein around all of the friction and the burdens that come with just doing things, getting coverage, certain types of insurance coverage can require lots of paperwork. And so on and so forth, and they found that removing that or automating some of that work can really get enrollment up significantly. There's a huge opportunity there. 

Christine Ko: Yeah, it's so interesting that you just mentioned Cass. I just spoke with him. 

Oh, yeah, he's terrific. 

He's amazing. 

Mitesh Patel: Of course, Cass and Richard Thaler wrote the book Nudge, which I read when I was in business school and really got me interested. That was at a time when I was in [00:08:00] training and started to notice. Hey, the iPhone has come out. Patients are really using this. The electronic health record was just being adopted by doctors. A lot of emphasis was on using the technology, but no real emphasis on the design or what behaviors this would lead. And I was reading the book Nudge at the time and started to connect the dots and said, Hey, there's a real opportunity here. And then found some mentors at Penn, Drs. Volpp and Asch, who were already working on some of the incentive work and was able to collaborate with them and really start to kickstart this field in nudges and healthcare.

Christine Ko: Nice. So is that the Penn Medicine Nudge unit still ongoing? 

Mitesh Patel: Yeah, the Penn Medicine Nudge Unit is still going. Dr. Kit Delgado leads the work now. And it's exciting to see some of that initiative. While I was there, we helped other health systems across the U. S. and in other continents start nudge units within their organizations themselves. About two and a half years ago, I moved to Ascension, a non profit health system in the country, to really take some of those insights from the research and apply them at real world scale. 

Christine Ko: That's cool. Do you use behavioral nudges in your own life?

Mitesh Patel: I think you have to practice what [00:09:00] you preach. Otherwise, you don't really know the true impact of it. I try to do a couple of things. I try to set goals. I think carefully about the defaults. What makes it easier or less easier for me to do. I try to set certain reminders and calendar things in so that I'm forced to make decisions about things that I otherwise might put off. I could certainly do a better job.

I have two boys, five and eight years old.

Christine Ko: Congratulations.

Mitesh Patel: A lot of energy in the household. One interesting thing that I heard was, kids have a lot of leverage because they can take all the time in the world to negotiate with you. But even if you frame things as simple as, would you like a vegetable with your pizza or pasta? Instead of saying, would you?, then there's, which vegetable would you like? And saying, would you like some carrots or broccoli? I've tested that out and found if you can frame the wording the right way, they're more likely to say, Oh, I'll take the broccoli today. Or I'll take the carrots. As opposed to prompting it. So that's moving from like active choice to like almost enabling. We call it enhanced active choice, almost enabling people to pick as opposed to saying [00:10:00] just, No 

Christine Ko: You've addressed a little bit already some of the behavioral nudges in healthcare. You talked about the vaccination and cancer screenings and training the medical assistant to already just pend the order. Is there any thing that an individual physician can do?

Mitesh Patel: Yeah, I think what's unique about healthcare is it's one of the areas where it really needs collaboration between clinicians and patients. And human behavior is the final common pathway, whether it's a vaccination to help prevent you from getting sick, a medication that might reduce your risk of heart attack or stroke, or some really targeted therapy or treatment for a chronic condition. Two things have to happen. A clinician has to recognize that you can benefit from this treatment, and then bring it up and discuss it with you, and then the patient has to understand what it means to them, what it costs, and how it fits within their life, and then actually accept it and use it as directed.

If neither of those things happen, a clinician doesn't recognize and offer it or a patient doesn't accept or use it, it's not going to lead to any benefit [00:11:00] for patients. And so I think clinicians have an important opportunity here in thinking about, first getting a sense of like, where are they starting out?

A lot of clinicians are surprised to learn their own behaviors. I remember when I was in training, there was a lot of emphasis on prescribing more generic medications at my health system because we were doing a poor job compared to others. And, I thought generics are exactly the same as brand names but more affordable to patients and they tend to adhere to them longer.

I looked at my own data, and I found that more than half the time I was prescribing brand name drugs. You just don't think about it because it's just easier to type in "Lipitor" than atorvastatin. Now there's ways we can change that. And I started working with the health system to think about, how do we set the default so that generics pop up first, no matter what you do? And we showed that we could improve generic prescribing. So that's one way of getting a sense of, what's your baseline behavior? Are there things that you can do about it? 

And then I think thinking carefully about how you communicate to patients. I think as you were mentioning earlier, a lot of these visits are very rushed. We tend to get in a certain habit that may have formed when we started training and, and you may want to think about what's [00:12:00] the best way to discuss vaccination with a patient that may be resistant to it.

When do you, how do you bring up, cancer screening and the risks and benefits in a way that makes sense to a patient but stays true to what they are? There's a lot of opportunities for clinicians to engage patients in their care. 

Christine Ko: Do you have any final thoughts?

Mitesh Patel: As I mentioned, there's a lot of opportunities in terms of ways to apply nudges in health care, but the question I always get is, I'm interested in starting a nudge unit, or I want to apply some of the work where I am at my health system, but I don't have funding to hire a behavioral scientist, or I'm not plugged in with the people who set our electronic health record settings, even though I see something. And I always tell people, the ideas end up snowballing. Look for an opportunity within your department or your division. Speak with the clinician informaticists who often are plugged in with the IT group as opposed to going directly to IT. There's often things that you think are really challenging to change in the electronic health record that are actually much simpler. There are many things that are complicated, but there is opportunity there. Or you could pilot something without [00:13:00] changing things in the electronic health record, giving people feedback, but focus on finding one idea. Demonstrate that there's potential there. And that often, if you can have that initial proof of concept, it often gets people who you work with to implement that on your team and on board and opens the door to a lot more. I think that's one way to get started in whatever situation that you're in. 

Christine Ko: That's cool. Thank you so much for your time. I appreciate you doing this with me. 

Mitesh Patel: Yeah. Thank you for having me on. Appreciate it.