Benchmarking Health Systems Across Multiple Attributes, in Multiple Markets | A Conversation with the Executive Vice President & Chief Medical Officer of Banner|Aetna

The following is an excerpt from a conversation on benchmarking across multiple attributes to inform a local strategy with Trilliant Health's SVP of Market Strategy and Chief Research Officer, Sanjula Jain, Ph.D., and Robert Groves, MD Executive Vice President & Chief Medical Officer of Banner|Aetna.

Highlights

Key Questions

Sanjula Jain Ph.D.: Healthcare is local, every organization is different, every market is different, the payer mix is different, and the patient mix is different. As an industry, why is it that we benchmark against “ranked” lists instead of our true peers?

Robert Groves, MD: This is the fit test; and for a lot of metrics it is a challenge. People that train only in one program for their entire pre-career training don't get how different healthcare can be in different markets. It's true that we need some significant standardization, across markets for certain things; human beings respond to medications, similarly, no matter where they are. There are some fundamental issues that need to be standardized across healthcare. When you get down to delivery, when you get to the processes that will effectively deliver that care, then it becomes intensely local: culture, notions about healthcare, certain communities are very open to taking a doctor's advice, and certain other ones are not. You have to have different strategies in those different places. That becomes the art of healthcare. What I've learned from you Sanjula and from Trilliant’s comparisons is that there are markets that are very similar. I've noticed in my travels across the US and the different places of practice, that there are areas where there's a lot of similarity and philosophy, and then there are areas where it's very different. Understanding that can help me when I want to compare a strategy that I read about or when I have information on what's in New York and I want to know how that applies to Phoenix, Arizona. There are so many things that go into that, that are local, including psychographics, which people tend to shy away from for reasons that I don't yet understand. Understanding what those differences are and how my population compares to another is critically important if I want to evaluate a strategy for improving the care that we deliver to populations.

Sanjula Jain Ph.D.: When Banner|Aetna is thinking about their strategies and learning from others and thinking innovatively, what are the standards, whether other organizations or best practices? How do you compare performance whether it's internally or externally?

Robert Groves, MD: It's a rough guess, it's judgment, basically, what we have today. The more information that I have about reality, where patient meets provider, whether that's a physician, a nurse practitioner, a physical therapist, whoever that is, the more that I understand about the realities in that location, the more effective I'm going to be in delivering care. This is another facet of evidence-based strategy. I can have a strategy that's great for New York; I don't know how it's going to be in Phoenix. What might happen before I had the option of comparing similarities, is I would say 'Okay, I liked that strategy I think that I'm going to have to tweak it because of my population'. Then you have a process of trial and error and iteration to try and get it right. Sometimes you can figure out how to make it work. Sometimes it just doesn't seem to apply. What this does is allow me to avoid, is making those necessary errors, it improves the efficiency of my strategy.

Sanjula Jain Ph.D.: What are some challenges that you face right now in pursuing evidence-based strategies, and managing these complex patient populations?

Robert Groves, MD: Let me give a concrete example because it's one that's fresh. There is a company on the West Coast, it's a not-for-profit, that I've been evaluating. They have figured out how to create community groups so that when a physician sees a patient, they can literally write a behavioral prescription. They refer the patient to these community groups which do the basics. We all know it's the behaviors that impact most health outcomes. Yes, there are diseases we need to treat, but it's things like exercise, diet, sleep, and connection, meaning, purpose, that really have an impact on how long and how well patients live. This group has come up with a strategy that's working really well in Northern California. It was designed initially for the Medicaid population in California. The question becomes, would that work here? Can that work here? How do I translate it to have the best chance of success because it is possible to bring a program in that would work and do it so badly that you lose the opportunity? Understanding the nuance of how that methodology translates to this market, could be important for the survival of a very effective program. How do I compare those two populations and this model? That’s where the SimilarityIndex™ comes in, I can look at what those differences actually are.

Sanjula Jain Ph.D.: What are some of the assumptions that you have had about markets that you've operated in or even facilities or physicians or patients that you think have started to change?

Robert Groves, MD: The most basic assumption of all is: if it works in this Medicaid population, it will work in that one, and when you look at the significant differences between populations, you know that can't possibly be true. That has been a realization to me, that there are enough differences across the United States, big geography, right? We're not talking about Sweden here. This is huge geography with lots of different embedded populations, with very different cultures. It's amazing that we hang together at all, frankly. I'm at the point now after looking through the SimilarityIndex™ from a variety of parameters, where it's foolish not to investigate what those similarities are or are not because there's so much difference. We’ve had notions about what will work in Medicaid versus what will work in Medicare versus what will work in the commercial market. What we haven't had is how different those payment methods can be across geographies. I've just started to understand how the demographic in one location might be very similar to mine, but the psychographic profile is very different, or the culture is different to a large degree and the things that they value, etc. I feel like we're just scratching the surface, it's a new tool, and you're not sure all the ways it can be used. I think it's going to take a lot of us to figure out where the real value is long-term in making decisions about healthcare strategy, and what solutions we bring in, what things do we develop? Do we need yet another brick-and-mortar hospital? What does that mean for this population who's going to embrace it? There have been things that have certainly surprised me, we brought a company in to look at a behavioral influence strategy, in other words, getting patients to engage in managing their own health. We thought it was only going to work for a younger population because it's highly based on a digital strategy. What we found is that we had 70% plus engagement at a location in the west valley of Phoenix where the average age is 80, that was a surprise. Now, there's something about that population that is likely to be different from populations where the average age is elsewhere. It's going to take time but it’s my belief that we will understand not only that these differences exist, but increasingly what that means for our strategy going forward. The only way to do that is to use the tool and start to learn those lessons.

Sanjula Jain Ph.D.:  Where do you see the opportunities today for health systems looking to compete against new entrants like Amazon or CVS in utilizing consumer data?

Robert Groves, MD: There’s no doubt that the Amazons of the world and even if you think about retail, the CVSs’ of the world have used consumer data. You go into a CVS in one part of the country, even one part of the city, and you're going to find significant variations in specialty items. Why is that so? Well, it's because they've got data, they know what's going to sell in this store in this part of town versus a CVS and another part of town, they understand their market to that degree. Amazon has had the luxury of parsing data in a virtual world. One of the things that I've tried to convince my colleagues of is, look if we don't begin to understand our consumers in a way that Amazon does, but specific to healthcare, we run the risk of losing this battle for the delivery of healthcare to Amazon. Our advantage really is that we are local, we're in the market. We have the opportunity to engage and connect with people that don't exist virtually. Can you connect with people virtually? Absolutely, particularly a certain psychographic, but for the vast majority of people, when it gets down to healthcare, it is intensely local, intensely personal, and it's based on connection. It's not as if we can ignore A/B/X testing data, patient as consumers, we have to embrace all of that, because it has become part of the expectation based on other experiences. We have a significant advantage in that we live in these communities, we are able to get face-to-face, we're able to do the immeasurable of connecting with patients, understanding who they are, and where they come from. If we combine the marketing, and retail strategies of an Amazon or you know, a Nordstroms, for that matter, if we combine that with our ability to understand and connect with people in our communities locally, that's the Holy Grail. That combination is what wins the battle because then you are meeting patients where they are because you understand them, you understand the culture in which they're embedded, you're in that community, and you can meet with them face-to-face, establish the connection. That is a fundamental part of influencing behavior over time, people do what their physician says because they trust them. They trust them because they've spent enough time with them to develop that trust. When you lose that, then you get a lot of people visiting Dr. Google and doing whatever the heck seems like the biggest or latest fad, instead of having a nuanced conversation with a knowledgeable professional who has their best interests at heart.

Sanjula Jain Ph.D.:  I appreciate how you expose just how local healthcare is. I know that the concept of similarity has really started to change your view of the markets and comparisons and benchmarking. We're about to release our next version of the SimilarityIndex™, which will be focused on hospitals, so now organizations can directly compare specific hospitals on any number of metrics from quality, which is something that we're all very familiar with, to new features like market share and outpatient presence and services. At this point in time, what health systems or hospitals would you consider Banner Health's peers from any of those dimensions?

Robert Groves, MD:  I’m more critical of my home hospital, probably than I should be. I want to be better, and when I think about Banner Health, I think about Intermountain, I think about Geisinger, I think about Providence. Those are the three that I want to emulate. That's different than the Mayo Cleveland Clinic model, but that's not who Banner is. I think Banner is much more like or at least aspires to be like, those other systems that I described. Why is that? Well, I think Intermountain has a long history and a culture of really demonstrating the use of data to improve care, Geisinger has done it with a broad population of rural members which is enviable, and we have that same issue, particularly in Northern Colorado, but also in Arizona. Arizona is, you know, vastly rural if you get outside of Phoenix and Tucson. Providence I think of as a very forward-thinking system that is interested in innovation, and how we can change things for the better. So those are the associations that I make, rightly or wrongly, those are who I compare us against. 

Learn more about the SimilarityIndex™ and explore our publicly available tool to learn how your market ranks.