This article is brought to you by Medalogix. The article is based on a live Q&A session with Elliott Wood, President & Chief Executive Officer of Medalogix, which took place at the Hospice News ELEVATE conference in Chicago held on October 21, 2021. The discussion has been edited for clarity.

Hospice News: Welcome to our Q&A with Elliott Wood who is the CEO of Medalogix. Elliott, why don’t you tell us a little bit about yourself?

Elliott Wood: I am President and CEO of Medalogix. We are a home health and hospice predictive modeling, machine learning company. We have been around for around eight years. I joined the company when we were in an attic on Music Row in Nashville. There were just four of us. If you’ve ever seen the HBO show Silicon Valley, that was my life for several years and I’m so happy to be here talking about what we’re doing.

HN: Fantastic. During that eight-year time period, obviously, you’ve seen a lot of change in the industry and your business. What are your thoughts on that evolution and how it’s progressed?

Wood: We tell this story a lot. When we first started, we were extremely focused on the data we had. There was so much rich information, specifically in the home health EMR, and we were pulling all of that data. Our very first product predicted the patients on home health with a high likelihood to pass away over the course of the next 90 days. We realized very early on that if we weren’t going to take that capability and develop it into a product a clinician would use, we weren’t going to create a lot of value.

That capability, through a couple of failures along the way, evolved into what is now Medalogix Bridge. We are focused on, like I said, identifying patients on home health who would be potentially better served in hospice because they have a high probability of passing away. That’s where we started. Hospice has been very, very dear to us for a long time, even when we were serving home health customers. It was also a cultural endeavor. We started with a hard problem, apparently. Then, recently, we merged with a company called Muse Healthcare.

Muse is a similar company in terms of doing machine learning on the hospice side, where we’re taking all of the clinical information from the hospice EMR, developing a prediction around the likelihood a patient will pass away in the next seven days, and ensuring that patient is getting the right amount of utilization at the end of life. From an evolutionary standpoint, the theme for us has always been, “It’s not just about developing cool stuff and cool models and widgets, it’s about building stuff that clinicians will actually use.” When we merged with Muse, that was a very similar philosophy they had.

HN: Interesting. Earlier today one of the panelists made a comment that hospice is very good at looking at what happened yesterday. That really stuck with me. Tell me about how you guys are working with the data to identify patients where they’re at today, and where they’re going or should be going.

Wood: I mentioned both of the products that we have. We are both looking at patients in curative care with a high probability of passing away, then elevating that for a clinician to review. We are also doing the same thing once the patient is on hospice. One of the things we have found is that it’s a very complex problem. We’ve had hospice clinicians on our staff who did a great job identifying patients who were likely to pass away when there was about a week left.

If you think about that and that experience, that’s when you have to start the paperwork, start having very difficult conversations with the patient, the doctor, the family. By the time all of that has happened, maybe you have three days left on hospice. The same is true on the home health side. I’ve heard multiple people today talk about the hot topic of staffing shortages, where some patients on hospice have gone weeks without staff coming to see them.

In either case, what we’re really trying to do with the data is ensure the trajectory of the patient is known by the clinician so they can make the right decisions. These are decisions about both the conversations they’re having with the patient and their family if it’s about end of life, but also about how to prioritize utilization on the hospice side so that when a patient is actually passing away, you have a lot of attention and the right utilization for that patient at that point in time.

HN: Throughout the day, we’ve talked a lot about transitions from home health to hospice and the warm handoff back and forth. Your organization sees a tremendous amount of data. What are some of the statistics you’re seeing with that transitional period from home health to hospice, and what can we take away from that data?

Wood: As a baseline, we’re seeing that the end of life conversation is almost always happening too late. That’s not just a home health and hospice problem, that is a problem across the board. I have a very personal story, where my grandmother passed away this last December. She was 92. She fell at home, broke her pelvis, went to the hospital.

She was never going to walk again. She was 92. Someone should have been talking to her about end of life and instead, they sent her to an inpatient rehab facility where she got COVID. She spent 48-ish of the last 50 days of her life in a facility. We, as her family, were watching all of this happen outside of a window. It was terrible and that story is not unique to me. That is a systemic issue where these conversations are happening too late. That’s what we’re trying to prevent.

The problem we’re trying to address is identifying patients earlier in that process. I mentioned my colleague, she used to be able to identify these patients a week out. We’re trying to identify these patients as much as 90 days in advance. The way we track this with our customers is by looking at whether they’ve had patients pass away on their home health census—that should never happen, especially when you have a hospice counterpart.

It does. How long are these patients living on hospice once they get there? We track short lengths of stay and try to determine whether or not we’re impacting that. We track payments, ensuring that once a patient is on hospice, we’re identifying that patient and increasing utilization that’s happening at the end of life. All throughout that process of getting the patient to hospice and ensuring the utilization is right once they’re on hospice, we track those KPIs with our customers so they know if they’re using the product well and what types of optimization they need from an operation standpoint to be better.

There’s a litany of data about the data and how well you use it, and there’s quite a few markers of success there.

HN: Talking about markers of success and data about the data, I think many years ago, I saw a demo of the Muse platform and talked about all the data points that go into it. It’s hundreds of data points. Tell me how all of those data points are coming together to then help families

Wood: Yes. I shared my story. That’s a very specific one. On the hospice side, I mentioned it’s a very complex problem, especially with a staffing shortage. There are cases where patients go weeks without seeing a clinician. If you think about what’s happening to that patient and what’s happening to their family during that time, they are not at all prepared for what they’re about to go through.

A lot of times, the caregiver for that patient outside of the hospice agency is someone like me. They’re not a clinician, they’re there supporting their relative, their wife, their mother, their grandmother, whatever. A lot of times the patient might pass away at two o’clock in the morning when you can’t get ahold of the doctor and you don’t know what to do, so you freak out.

Call an ambulance. They come, they take your grandmother to the hospital, or to the ER, they admit her to the hospital, and then she winds up passing away in the hospital, which is exactly what she didn’t want. Again, what we are trying to do is ensure that doesn’t happen. Especially, the hospices that are here, the role that you guys play, it’s such an incredible thing for those patients and those families because you’re preparing them for what that’s going to be like when that patient transitions.

In each case, whether it’s ensuring that the patient has time on hospice or ensuring that the family is prepared for that moment, once it does happen, that’s really what we’re trying to do.

HN: In the context of hospice, what does the data scientist say about hospice and where it’s going?

Wood: Where are we going as an industry? All the conversations around value-based care and how agencies evolve into those types of payment structures are very interesting to data companies, because you need data in order to do that well. Data science, for the most part, is always to optimize towards an outcome.

In each of the cases we’ve talked about today, we’re trying to ensure that the right care is delivered to the patient irrespective of payment. I think from our perspective, we’re very excited about the role hospice can play in value-based care. Two very practical things that we see are one, with the VBID demonstrations, hospices that take the best care of their patients and that structure will win.

From a data scientist’s perspective and someone who’s really trying to drive outcomes, that’s very exciting to us. On a larger scale, we do a lot of work with risk-bearing entities. Groups that are taking risks, physician groups, conveners, etc. And so we have a lot of conversations with them, that wind up being about how to better utilize hospice because they don’t utilize hospice very well.

In a lot of cases, it’s similar to the home health and the hospice problem where they don’t really know how to identify those patients appropriately. Number two, they don’t know how to have that conversation with the patient even when they do identify them. A stat that I’ve seen recently is we spend, as a country, $180 billion dollars on patients in the last 8 months of life every year.

1% of our GDP is spent on taking care of patients in the last six months. When you think about the opportunity this industry has to shift that care to hospice or to palliative, it is just tremendous. As a data science company, we are very excited about the potential to help support that shift, and from a patient outcome perspective, think that that’s very exciting, too.

This article is sponsored by Medalogix. Medalogix is transforming home health by combining state-of-the-art data science capability with world-class clinical expertise. To learn more, visit https://medalogix.com/.