Good description of the problems with Microsoft Viva. I usually just say it's not helpful, obnoxious, and angering. Your description…
Joshua “Josh” Pickus, JD is CEO of Net Health of Pittsburgh, PA.
Tell me about yourself and the company.
I am a serial CEO of technology companies. This is the second healthcare tech company that I’ve run. Net Health is a EHR and analytics company that is focused on medical specialties, such as therapy and wound care.
How do the EHR needs of skilled nursing facilities, senior living operations, and home health differ from those of hospitals?
Let me give you an example to make this real. I’ll do it in terms of physical therapy. In some respects, there are real commonalities. In all cases, you need to accurately document the care in a way that is compliant with the reimbursement codes. You need to do that whether you’re in a hospital, a skilled nursing facility, or an outpatient clinic.
But there are very important differences, and they often end up having to do with things like integrations. In a hospital context, in most of our situations, it’s critical that our systems interoperate with major hospital EHR players such as Epic, Cerner, and Meditech. Making those integrations seamless is frankly as important as the functionality that we have in our own product.
If you contrast that on the other end with, say, a outpatient clinic or facility, that’s a much less critical piece of what they do. They probably don’t have a direct interface with Epic, and that’s not that important to them. Skilled nursing facilities are somewhere in the middle. There are key integrations, principally to PointClickCare and MatrixCare, but it’s different players. Our functionality may be quite different, but the integrations to other systems are quite different among settings and are very important.
The early days of COVID forced hospitals to coordinate with post-acute facilities to free up beds, and CMS added some requirements around that coordination. Is that data sharing relationship among types of entities improving?
Yes. We are unique in that in the specialties we serve, we are really hospital-to-home. We have to think about that stuff. You are right that the pandemic brought those issues to the fore. As a result, the pace of progress has improved.
But the core issue is still simple. There isn’t a common system or even a common accepted language to go from hospital to home, to transmit core patient data seamlessly, easily, and accurately. We are focused on the FHIR standard, which is the closest thing we have right now to a standard that lets different systems at different parts in the continuum talk to each other.
A lot of our work is on improving our FHIR capabilities and making it truly seamless, so that basic information can easily pass from one setting to another in a way that the recipient and the provider of the information actually know what’s going on. You would be shocked that basic stuff — like if you’re a nursing home and you want to know something beyond the patient’s name and age, such as the existence of any behavioral health concerns — isn’t as easy as you would think given that it is a specific, easy thing that you get every time. Working through FHIR to improve that interoperability continues to be a key focus and challenge for us.
How do you expect the hospital-at-home and remote patient monitoring concepts to play out?
I would respond differently to the two things you mentioned, in terms of the timeframe. Remote patient monitoring is here. It’s real. It’s in use. It’s quite valuable. I think it will expand and pretty dramatically. Hospital-at-home is interesting, but in its infancy and less likely, in the near to immediate term, to affect the way that care is delivered.
We are more focused on remote patient monitoring and all types of remote care, even if the locus of care is still a hospital-based physician. There are a lot of things that can be done by that physician or caregiver without the patient in the room. That feels real to us, and we are introducing all kinds of capabilities to support that trend.
Will device connectivity and integration issues present challenges?
In the grand scheme of things, the technology is the least of the problems. If you break it down, think about the most basic form of telehealth, which is simply a audio and video call between a caregiver and a patient. That exists, it works pretty well today, and it turns out that it is really valuable. I live in Utah, and about half the time when I see my providers at the University of Utah, we do it virtually. To me, that’s here, that’s ready, that works.
Then you get into slightly more sophisticated stuff. Can a patient who has a wound that needs monitored get that captured by their iPhone and send the picture to the hospital that’s caring for them so they don’t have to make a two-hour journey? It turns out that’s available, too. You could go on and say, can you monitor a patient who is undergoing physical therapy and you want to accurately gauge their range of motion through sensors? That’s available today. too.
I don’t want to minimize the technology challenge. There is plenty of improvement, but it is much more behavioral change that is the obstacle to that than the actual technology.
What are skilled nursing facilities doing with analytics?
SNFs don’t have the budgets that hospitals do. As a result, spending on analytics isn’t anywhere near as large as it is in hospitals. But they are doing important things.
One that is critical to both hospitals and SNFs is readmissions of patients. If a hospital sends a patient to a SNF and then the patient bounces back, that is bad for everyone concerned, especially the patient. So, one of the things that SNFs are focused on is preventing hospital readmissions. How do you do that?
It turns out that oftentimes what causes a readmission is something as basic as a patient falling, reinjuring themselves and needing a more acute level of care. If you can monitor fall risk and accurately determine which patients are at greater risk and take steps to prevent that fall from occurring, you will reduce readmissions. And if you reduce readmissions, everybody, including the patient, is a lot happier. There are some tangible things that SNFs are doing with analytics, and many of them actually relate to the hospital that sent the patient in the first place.
Are hospitals rewarded for discharging patients to facilities that perform better, and do they provide technical or financial assistance to those organizations to improve outcomes?
There are two ways in which that is occurring, and it’s real. There’s kind of a formal and informal way.
In the formal way, you will have hospital-based ACOs, or accountable care organizations, and they will have formal arrangements with downstream providers. The payments to the provider will be dependent on specific metrics, of which hospital readmission is usually at the top. That exists, but it’s not yet terribly widespread.
The more common arrangement is that many, if not most, hospitals maintain networks of skilled nursing facilities and are deciding where to send the patient. Increasingly, they are focused on the patient experience. There are very different levels of sophistication that this is being done with.
Well-managed networks will pay attention to five or 10 metrics, ranging from readmission to customer satisfaction, about the patients who they send downstream. That will affect where the next placement goes. That incents the SNFs in a very real way to achieve against those metrics, because it will determine the patient flow. That became less powerful in some respects during the pandemic because there was such a bed shortage that it didn’t matter. But as we exit that period, that’s becoming relevant again. It does impact their top line, in terms of their census, based on whether hospitals are sending them patients.
How did your Tissue Analytics product earn FDA’s Breakthrough Device status and how are customers using it?
This is genuinely cool, and it is novel. In fact, it was novel to us, because it’s called Breakthrough Device status and we don’t make devices, we make software.
It turns out that software that makes predictions that affect outcomes in care is regulated by the FDA as a device. Breakthrough Device status means is that you have built something that is so novel and potentially so beneficial to patients that FDA is going to put you in this Breakthrough Device category. They will expedite the review that you need to get an approval to have your product sold and used for particular applications. It was a journey for us, as a software company, to enter the FDA regulatory scheme.
We are doing things like predicting the velocity at which a wound will heal, predicting amputation risk, and ultimately predicting which kind of treatments are most likely to lead to an expeditious and effective piece of care. We have demonstrated that we now have enough data that we can accurately predict that “this patient needs this many visits of this duration to achieve that result” or “this patient is at materially higher risk of amputation if action isn’t taken immediately.” It’s making a real-world difference out there. We probably have 50 or 70 customers using these modules, so it is becoming an accepted part of wound care practice.
Will the experience that the company gained from working with predictive AI and the FDA influence future product development?
Very much so. We view analytics as a key piece of where EHRs are going. If you think about it, EHRs have traditionally been systems of documentation. They exist so that you can document the care given. That’s a baseline requirement, but it’s not really where EHRs are going. EHRs are becoming systems of insight and systems of engagement, in addition to systems of documentation.
By systems of insight, I mean that if you are the system through which the clinical workflow is happening, you have a unique opportunity to collect data about what works and what doesn’t. This is the piece that people miss. You also have a unique opportunity to put insights back in the workflow to alert a clinician at the precise moment, which increases the quality of care that they can deliver.
The analytics piece for us and the predictive piece for us is very much about the next chapter in what an EHR is. It’s really about harvesting the data to yield insights that you can feed back to clinicians that enable them to deliver better and more cost effective care. It’s at the very heart of where we’re evolving our systems.
You’ve said previously that a lot of EHR frustration is due to entry of that isn’t used to change outcomes and doesn’t directly support the clinician who is expected to enter it. How will that evolve?
General purpose hospital EHRs like Epic and Cerner will also include analytics and predictive analytics as key parts of what they are doing. Those are sophisticated companies. They understand that this is the next chapter for EHRs, and they will participate in that. We view ourselves as additive to what they do, because the workflows and the data that we capture are unique to the specialties we serve. To be able to deliver accurate predictions and useful clinical insights, you need that unique workflow and unique data.
We think that what we are doing and what they are doing are complementary. We work in many hospitals with both Epic and Cerner. Virtually every installation of our Tissue Analytics product is with a system that runs Epic or Cerner, so it is important for us to be complimentary and to interoperate with them.
What changes do you expect in the company and the industry over the next few years?
If I were going to give you two words, it would be more interoperability and more analytics, especially predictive analytics. Both of those things will become so embedded in what we do that you can’t really separate that piece from us. The importance of connecting with other systems and the importance of using the data that you have to deliver insights is really the future that we see as Net Health continues to evolve.