Expanding Technology for Health Systems to Monitor Pandemic Outbreaks

Chris Hobson, MD, chief medical officer, Orion Health, provides tips and highlights a newly developed platform for health care executives aimed at improving practices and ensuring the safety of both staff and patients during a pandemic.

Podcast Transcript:

Welcome back to "Pop Health Perspectives," a podcast hosted by the Population Health Learning Network where we combine expert commentary and exclusive insight into key issues in population health management and more. Today, we are joined by Dr. Chris Hobson.

My name is Chris Hobson. I'm the Chief Medical Officer with Orion Health. My clinical background is 15 years as a primary care physician. Within Orion Health, I try to guide the company from a clinical perspective, and also particularly have interest in the areas of population health and some of those aspects.

I think the first thing is that pandemics will happen. We like to, once the pandemic's over, we forget about it. We've had SARS. We've had H1N1. There was MERS. There was Ebola. The first thing is, I think we do need to be prepared.

Health system executives need to have this on their radar, even when things are normal. Now, when things are going badly, as they seem to be at the moment, the next thing as well as preparedness is to think ahead and realize that especially hospitals will potentially face a huge overload in terms of the workload that's going to come upon them.

This is the problem of exponential growth, which we're seeing at the moment. Exponential growth has the power to swamp the hospital in particular. We need to think ahead on an ordinary day, and in the midst of a pandemic, we need to think ahead really carefully for the excess volume and see what we can do to keep that volume down.

A lot of the pandemic work is focused on the hospital. We want to only have patients who need to be in the hospital showing up at the front door if we can do that. We need to keep that load manageable.

There's two reasons for that. One is we want to keep the hospital itself from being overloaded. Related to that is we need to keep the staff safe. We want to minimize the effort that they have to go to in sorting out people who are sick from people who are not.

We want to make it easy for them to care for those patients who are sick and need hospital care, and we want to make it easy for the system to care for everybody else at home, away from the chance of spreading the virus.

Incidentally, we've realized that we've been saying for a long time that hospitals should focus on what hospitals are good at, and we should move as much of the other care to the community. That's why we've developed some software already.

We were already trying to think, "How do we keep the load off hospitals so it's the appropriate load and move the rest of the care to the community?" As you know, or probably know, people know that Orion Health is a health information exchange and population health software vendor.

Also, now, in the US, we actually run several health information exchanges. As I was saying, we already have the ability to track disease, to know how much from a population health perspective, what diseases are present in the population and how well we are doing in terms of quality measures, how well we're doing in terms of managing it.

We really had that kind of broad population platform, if you will, combined with some analytics to tell how well the care is going. Also, we've been very interested more recently in looking at ways where we can monitor people at home and in the community, because that's the right place, giving them the right care in the right place.

When we hit the new COVID-19, the SARS-CoV-2 came along, we thought about this, and we thought, "Well, actually, all we need to do to tweak what we're already doing is to come up with programs so  we can manage those patients in the community using our remote monitoring software.”

When patients are in hospital, we can provide complete information about them to the providers in the hospital. The additional piece that we did put together is an initial website where patients can go, patients who are concerned -- or who maybe really do have need to be concerned -- can go and check their symptoms.

There's a front end, so to speak, front-facing website where patients can go check their symptoms and get advice. Now, a number of other organizations -- the National Health Service in the UK, for instance, and various health systems -- have put up their own symptom checkers.

That's fine, but this is available to any health system that would like it. One use case, for instance, is as people turn up to a long-term care home, they can check on the website whether they have symptoms and signs.

If they don't pass, then they can't go into the long-term care home. The website's new, and it links to then assessing the patient symptoms and severity to then, as appropriate, sending them to the hospital or putting them for remote monitoring in the home and in the community.

The other key piece, which again was really just a configuration of what we already had, but let’s reflect that back. As I'm in the community, I can fill out questions on a daily or twice-daily basis. I can also add in, for instance, my temperature, pulse, respiratory rate, what we call the oximetry, my partial pressure for oxygen.

These can be detected with devices in the home and even the latest smartphones. It makes it more and more possible to collect data from people in the home and then to provide a dashboard for the providers.

As a provider, if I'm not busy treating patients face-to-face, I can look at my dashboard and see how all of my patients are doing who are at home, so I can virtually care for them. We're also running algorithms, if you will, or calculations in the background to highlight those patients who most need care, or who certainly look the most suspicious for deterioration.

We provide those as worklists for the providers. I can see a worklist of all of my patients, sorting by most severe at the top, with all of their indicators, so I can see how people are doing.

That's the technology piece that we've, it's an adjustment or a configuration of what we're already working on and some new components added to make a full-service piece, if you will.

From the terms of how will it address the pandemic, it will certainly give us...off of the website and off of the remote monitoring, we get a situation room. Now, at least we know what we're dealing with.

In terms of improving patient care in a pandemic situation, if you don't sort out patients who are unwell and need hospitalization from patients who could be managed at home, and then patients are just, don't have the disease but are very anxious...

If you don't sort those out into different groups and then handle them appropriately, they will rapidly flood the hospital, and nobody's patient care is going to be any good. It improves patient care by giving the right care in the right place.

In terms of slowing the spread of viruses, the key piece here to appreciate, of course, is what you're going to be talking about, which is social distancing.

Rather than having 50 patients in the emergency room in a tight space, coughing on each other as they worry about when they're going to see a provider, if they can stay at home in the community and keep their distance from their neighbors and not get into crowded, confined spaces, we all have our fingers crossed that that will slow the spread of the virus.

All the best advice we have from the epidemiologists is “maintain that social distance, and you'll slow the spread of the virus.” What we're doing is ensuring those patients who are living on their own at home that they're properly cared for, with the right care for them.

For me, the interest has been infectious disease and pandemics. From an interest perspective, the reality of what we're all going through, we need to acknowledge that and how we all cope with being cooped up in single rooms and so on, and not having social contact.

Once we acknowledge that, the really interesting piece to me is we've really brought the characteristics of infectious disease into our population health. The major takeaways from the system, really, all of the things that we've been talking about with population health apply here, but with variations.

We absolutely need to know who is sick and who is not. We need to know who those patients are, so we have the right patient. We obviously need to give them the right care in the right place.

By moving that care to the community, it's not a new thing. The other interesting thing is that, with the COVID-19, is that patients who have hypertension, cardiovascular disease, and so on, other chronic what we call comorbidities, those patients are much more likely to need hospitalization and much more likely to be unwell.

From our perspective, we have already developed programs for caring for cardiovascular disease and COPD in the community already. It's a great system. It builds on what we've already developed.

It will benefit healthcare systems in multiple ways. It's strengthened our understanding of population health, and it's expanded it into infectious disease and pandemic situations, which again, I think this will really benefit organizations over time, because they'll have the knowledge, and they can start to do more advanced planning.

Unfortunately, I don't want to be negative, but if you look at the Spanish flu in 1918, it came back. It hit with a wave in 1918, but then with the summer, it actually got better. Then in 1919, the epidemic came back.

Either this time around, the COVID might get better, but we need to be aware it might come back, or for sure, there will be further nasty viruses like this that we can expect to come down the track in the next year, or three, or five.

They will come regularly, and health systems, by being prepared and knowing what to do and how to do it, that will benefit them enormously, I think. Much better than being reactive a few days before an epidemic hits their front door.

The current what we call use case, the current case of an epidemic of a disease that's growing exponentially, where we have very limited treatment other than what we call supportive care or respirators, but no direct care for the actual condition, no antibiotics or antiviral agents, that's the worst case or the best case for this type of monitoring system.

That's not to say that, if you have a lesser epidemic, it still makes sense. Let's imagine the H1N1, which was about 2010 or 2011, if I remember correctly. There were a lot of cases of that, but it never got to this extreme. A lot of people were unwell. I remember my son-in-law was extremely sick.

Even in a lesser epidemic, it makes sense to be able to keep infectious patients out of the hospital. I think once things have settled down, we could take a good look and say, "Well, patients with other infectious disease, are we bringing them into the hospital unnecessarily, and could we move those to the community as a normal, standard operating procedure?"

That's all part of the right care in the right place. One interesting area that we've looked at more recently is applying. As well as analyzing a population, providing quality measures, and so on, we've looked at doing predictive modeling.

A good example of that was with the high-risk maternity. We recently did a project just a couple of months ago where we were able to look at the data in the health information exchange and fairly accurately, with fairly good accuracy, predict women who were pregnant, predicting those who were going to be at high risk.

Now, and there's quite a bit of interest in that for all sorts of reasons that we could talk about. With respect to COVID-19, it's almost impossible. At the moment, we don't know enough about the biology of the disease.

Just recently, in the last few days, several of our clients have started to log the disease in the health information exchange. For instance, there was the ICD-10 code for this disease, it came out two weeks ago, you see?

What we're hoping is, as we will inevitably gather a huge amount of data, both from the remote monitoring and just the normal care process will be in the health information exchange.

We certainly hope that maybe in three to six months, and we're starting to think about it now, but we'll have the data that we can predict more accurately who is a high-risk patient for this disease. Maybe that will help us if we do get a second wave, for instance.

We keep monitoring both how our clients are coping and looking for ways we can help them. We keep monitoring the medical literature, the journals, the World Health Organization, and the CDC. We keep incorporating things as we learn them.

For instance, it was easy to incorporate the multiple chronic disease into RPM. Over time, we keep incorporating the new knowledge into this overall package. From a digital health perspective, we feel like they're making headway, which is really exciting.