At Banner Health, the Phoenix, Ariz.-based integrated health system, healthcare leaders built an in-home telehealth program targeting a subset of patients with complex chronic conditions following a successful tele-ICU program. The in-home telehealth pilot program, called the Intensive Ambulatory Care (IAC) pilot program, done in partnership with the Netherlands-based Royal Philips, focuses on the most complex and highest-cost patients —the top five percent of patients who account for 50 percent of healthcare spending. The program first launched in 2013 and aims to improve patient outcomes, care team efficiency, and prevent IAC patients from entering the acute care environment where costs are significantly higher.
Back in January, Banner Health, with operates 28 hospitals, reported ongoing successful results from the IAC in-home program. In this updated study, Philips and Banner examined 128 patients who had at least one year pre-IAC and one year post-IAC follow-up to examine the prolonged impact of the IAC program on patient outcomes. The analysis of patient results over the first full year of the program revealed that the IAC program helped to reduce hospital admissions by nearly 50 percent. Prior to enrollment in the IAC program, there were 10.9 hospitalizations per 100 patients per month; after enrollment, the acute and long-term hospitalization rate dropped to 5.5 hospitalizations per 100 patients per month.
Further, an analysis of the program found that it helped to reduce overall costs of care by 34.5 percent. This cost saving was driven primarily by a reduction in hospitalization rates and days in the hospital, as well as a reduction in professional service and outpatient costs, Banner Health officials reported. Additionally, the program reduced the number of days in hospital by 50 percent. Prior to enrollment, the average number of days in the hospital was 60 days per 100 patients per month, compared to 30 days after enrollment.
Banner Health further reported that the IAC program reduced the 30-day readmission rate by 75 percent; the 30-day readmission rate went from 20 percent prior to enrollment to 5 percent after enrollment.
To delve further into Banner Health’s telehealth initiatives, Healthcare Informatics Associate Editor Heather Landi spoke with Deborah Dahl, vice president, patient care innovation, Banner Health, during the HIMSS Conference in Orlando back in February. Dahl outlined the steps that Banner Health took to build a successful telehealth and remote patient monitoring program to improve health outcomes for complex chronic patients. Also participating in the interview was Manu Varma, business leader, hospital to home and Wellcentive, at Banner Health’s technology partner Philips. Below are excerpts of those interviews.
What are some of the initiatives at Banner Health using telehealth and RPM technology?
We started with tele-ICU, and what caught my attention with Philips, as opposed to the many other vendors that we’ve worked with, is they come with a turnkey solution. A lot of companies come and say, ‘Here is your really cool widget, here’s the instruction manual and you’re on your own, have a nice day.’ Philips’ approach with the tele-ICU is ‘Here is your technology, here are the marketing materials, because you’re disrupting how care is delivered at the bedside, and here are the key points you want nurses and doctors to know if you want to do it well.’ And, they say, ‘Here is the team you want to build and some interview criteria to help you’ as you want to make sure you have the right folks behind the camera [with the virtual care.] And, we had amazing results with [the tele-ICU program]. For full-year data, we had about 2,000 people in our over 600 ICU beds who lived who were predicted to die. There were about 4,000 fewer ICU days than were predicted, 65,000 fewer bed surge days, and that’s about $135 million cost avoidance. It’s never just the tele-ICU—it’s the bedside, the organizational structure, the clinical consensus groups, you need the whole thing to be working together. But, we think of ourselves as one of the glue pieces. So that helped us them move into the bed surge, then move into the ED and now into the complex chronic patient population home as we are now responsible financially, as well as clinically, for these folks when they are not in the hospital.
What has been the key to the success of the IAC in-home program for complex chronic patients?
With the tele-ICU program, it gives you those adverse trends so you can intervene before the adverse trends become adverse health. It provides us with tools that are daily reminders our foundational best practices. Our ICU teams, like most ICU teams, I call them adrenaline junkies, they are very busy and it’s not unusual to get an admission and a code and a discharge all at the same time. This tool says out of our 600 ICU patients, these three are missing that test so nurses go in and see what it is and turn to the tele-ICU doc who will write that order and make that happen. Just that core functions that happen every day.
With the IAC at-home program, it’s that same thing, the tools are collecting the data from the patient every day, and saying, ‘Here is a person whose weight is creeping up’ and it’s not something you would go to your primary care physician for, but the patient has congestive heart failure, so let’s take a look and have a conversation. The nurse will check it out and talk to the patient and have the conversation with the tele-physician or with the pharmacist, the PharmD is there as well, and they will discuss the patient and they might have a video chat with the patient, and make those changes immediately.
What you have found with regard to patient engagement with using the technology and participating in the program?
The average age of the patient in this program is 82, the youngest is 38 and the oldest is 101. It’s an interesting spectrum. When we started, we were quite concerned about the technology being a barrier but during one of the focus groups we were talking about virtual health and the patients had that blank look at their face. Then we said, ‘You know Skyping,’ and they said, ‘Oh yes, I Skype with my grandkids all the time.’ That language was so important to get people accepting and then after that it was a piece of cake. We have a health coach that is the only out-in-the-field person who is trained in motivational interviewing. Philips designed a great program for us around patient activation. The health coaches don’t come in to tell the patient how to do X, Y and Z, it’s about, ‘What matters to you? What’s important to you? And how can we help you meet your personal goals?’ The patient might say, ‘I want to get back to my bridge game’ or ‘I want to go out to lunch with friends’ so whatever your goal is, we’re looking to see how we can make that happen.
But in the early days, the health coach is also the technical hand holder, so of the first video visits we do with a new member is our medication reconciliation. The pharmacist is in the telehealth center and the health coach goes out to the home, and the pharmacist, through the video chat, will say, ‘Bring the shoebox of meds out to the table.” And they [the pharmacist and health coach] will go through that, literally opening the pillbox to make sure what’s in the container is actually the right med for that container. We’ve been through more than 1,000 patients and so far not one of them has their EHRs [electronic health records] matched what they have for medications. So you think, this patient has a cardiologist, a nephrologist and a neurologist and their PCPs, and the electronic medical record (EMR) systems, 27 or so different ones, don’t speak to each other. We found some amazingly scary things. They can be as simple the cardiologist prescribed a beta blocker and its generic, and the hospitalist, at discharge, prescribed another beta blocker, but it’s a different generic name, so the patient is taking both. So you have to get those things corrected. It’s a combination of technology tools.
The tools really do trigger us to take those next steps for those folks. The members of the program are given a tablet and they are asked a few questions once a week, and we’re looking for early signs of depression, changes in activity level, and changes in pain scale. So a message might pop up in the depression screening that is significantly different from the last score and our social worker and cognitive behavioral therapist will pick up the phone and call the patient to see what’s going on and what’s changed. We have had suicide ideation, elder abuse, and the expected depression, stress and anxiety, the things you expect when you are 82 years old and you have five chronic conditions.
Effective care management and care coordination are foundational to population health work, so are these tools critical for population health management?
I do think so. When you start your population health journey, you can say I’m going to do wellness and that’s a pretty long payback period. Or I’m going to focus on the more expensive group of folks and I’m going to use those savings to bring into the next layer and the next layer and the next layer. So we looked at this complex chronic patient population, and one of the other criteria is that they have to a $20,000 prior 12-month spend in order to get in to the program, that’s the threshold. If we can reduce their total cost of care by 34 percent, then you can improve their quality of life, reduce the number of hospitalizations, reduce the cost to them and to the plan, and so we’ve done that quality of life and cost reduction at the same time. Now, that allows us to move into some other space, like diabetic patients who are not yet into the complex chronic category.
What would be your recommendations for other health systems that would like to replicate this program?
Look for a partner that understands the complexity of the environment. So, for us, we didn’t know what we didn’t know. We knew what an ICU was like and how to do things in the ambulatory setting; it is probably 10 times more complex than a hospital ICU patient and I didn’t appreciate how complicated that was going to be. You’ve got your own employed physicians and they are on a single medical record, so you write an interface to that. You’ve got other physicians, even if they are in your narrow network in the ACO, and with our group, there are 27 different EMRs. You’ve got a lot of political changes, because you’re in other people’s space and, in the Banner group, there are about 600 primary care physicians that we are trying to work with. If you’re only interfaced with 200 out of the 600, that means a lot of faxing and phone calls to do that. You’ve got the technology, which I think is the easy part, but if it’s just feeding you data and it’s not looking to tell you that out of your 600 patients, here are the three people you should be intervening within the next hour because if you don’t in the next 24 hours, you may have a problem, then you really haven’t solved your opportunity. You can’t just look at raw data for 600 patients, your brain can’t see the trend in that fashion. It needs to be actionable data.
Varma: It needs to be synthesized data. I would add that I wouldn’t recommend that you try to reinvent the wheel. We do see a lot of that in the market with people trying to do it from scratch on their own. I would say think about the full population in your ACO strategy or your value-based care strategy. If you can see the full population, then you can know what your incentives are for each population cohort and pick and choose. And then use the technology, but also make sure to work on things from a process standpoint, such as having the right team. To this point, when we first tried to recruit patients, we said, ‘You really need help’ and the patients said ‘I don’t need help. I’m fine.’ We realized we had to talk to people completely different from what everybody was accustomed to. So there’s a whole level of design element to this and we’ve learned a lot of lessons from that. We were stunned by how embracing people were of technology, especially the tablet, which makes so much difference as the user experience is very intuitive for people. Equally surprising for us has been that we thought it was going to be a set of medical issues, but it’s not medical issues—medical issues are almost the lagging indicator, because the issue is the patients are taking those medications twice, or they don’t have the expertise available for them, there is nobody helping them through this, which leads to all kinds of care gaps, which ultimately then leads to serious medical admissions. I think those socio-economic and day-to-day issues are a real challenge.