Giving a patient medications in the ER, having them pop positive on a test, and then withholding further medications because…
HIStalk Interviews Lissy Hu, MD, CEO, CarePort Health
Lissy Hu MD, MBA is co-founder and CEO of CarePort Health of Boston, MA.
Tell me about yourself and the company.
I’m a physician by background. I started the company to better bridge hospitals and the care partners that they work with, such as nursing homes, home health agencies, hospice, all the post-acute settings that patients will need after their hospital stay. We are in just over 1,000 hospitals, 180,000 post-acute care providers, and 43 states. It has been a pretty amazing journey.
How is the pandemic changing the relationship between hospitals and skilled nursing facilities?
People are realizing more than ever that nursing homes are critical part of the care continuum. These long-term care facilities, where we house our elderly and our vulnerable populations, are incredibly susceptible to COVID. A huge crisis is going on in nursing homes across the country. As a result, they’ve stopped admitting patients. They are scrambling for PPE just like everybody else and for staff to care for their existing residents.
But in American society and healthcare, these nursing homes are also short-term rehab centers, where they take patients from the hospital. That helps to keep the whole healthcare system flowing, so that that you don’t have bottlenecks on the hospital end. They don’t have any places to safely discharge their patients for rehabilitative care, especially for COVID patients, where they are in the hospital for a long time.
Consider a 50-something patient who has never had any rehab needs. Once they’ve been in the hospital for a couple of weeks and on a vent, they’re deconditioned. They are going to need rehabilitative care. The pandemic has made it clear just how interdependent hospitals and post acute-care providers are.
A lot of the hospitals we work with are strengthening their partnerships and their connectivity. With some of the software that we’ve built between themselves and post-acute, they have been able to leverage some of those existing relationships in this time of crisis. It has been heartening to see hospitals continue to value their post acute-care providers.
On the other hand, being connected to these post-acute care providers, we see their EHR data. We see the spikes and deaths. Each of those data points is someone’s grandmother, someone’s parent. It’s just hugely, hugely sad.
Will the level of information exchange between hospitals and post-acute care providers change with this new level of dependence?
I’ve been on the phone with state and federal government officials talking about the pandemic response need for more support for post-acute care providers and more tools that help them. It’s not exactly in the area that I work in, in terms of our software that is connecting hospitals and post-acute care providers. But in those conversations, it is surprising that they are recognizing, for the first time, that these nursing homes play these dual roles. A lot of people think of nursing homes as these residential facilities. Awareness is building that they are also an outlet, a step-down unit if you will, for hospitals.
Prior to the pandemic, people were thinking about how to better work with their post-acute care providers on the hospital end. Because of things like bundles and ACOs, hospitals needed to think about the patient, not just in terms of their particular hospital stay, but their recovery period. I think we’re going to continue to see more of that with the interoperability rule.
One thing that a lot of people didn’t expect with the CMS interoperability role was mandating electronic notifications. Not only to the physician — CMS included skilled nursing, home health, and other post-acute care providers. That’s recognition that these post-acute care providers play an important role in the care continuum.
Here’s one example. When a skilled nursing facility transfers a patient into a hospital, they don’t know what happens to that patient. They’re calling the patient or calling the hospital to find out whether that patient is going to come back. Should they hold the bed, or should they not? When we built our software to be able to better communicate between the hospital and the post-acute provider, our infrastructure allows them to get notified about what actually happens to that patient. Are they just there for observation, or will they be admitted? That allows the skilled nursing facility to prepare.
That became even more important with COVID. The skilled nursing facility would send the patient back for testing into the ED. Maybe they would get tested, stay there for a couple of days, and then get sent back with one negative test. But with one negative test, because of how vulnerable that patient population is, the skilled nursing facility is still going to put that patient in an isolation room and use PPE. You need to know about that second negative test, which is when you can start to put the patient back into the larger residential community and start to conservative PPE.
We made some small modifications in our platform that transmits those lab results back to the skilled nursing facility. These skilled nursing facilities get confirmation that the patient is negative and can be moved out of an isolation room. Even small improvements in connectivity can have a big impact in terms of the skilled nursing facility and their ability to care for these patients, while also protecting all the other residents. I expect to see much more of that coming on to the other end of the curve with the CMS interoperability rule and in some of the requirements on the notification side. Not just the PCP side, but on the post-acute side as well.
Sometimes the biggest interoperability challenge involves integrating the received information into workflows. How do you see that working with ADT notifications?
It’s funny that even though I’m a physician who built a technology company, I always think that technology is probably just 50% of the answer, if not less. It may be a little bit heretical to say that as a CEO of a tech company. But the other big component realistically is thinking through the workflows. If people send notifications in a way that requires someone to log into a portal and view an event, you’re taking the nursing home out of their own workflow. That presents a huge barrier to adoption in terms of making the information usable and actionable for that skilled nursing facility.
We have 180,000 post-acute care providers on our platform, so we think that we’re in the right position to surface these notifications into the workflows of skilled nursing facilities and other post-acute providers . They are in our systems every day. We see them doing really practical stuff with this information, like deciding whether to hold the bed of a patient who has gone to the ED while waiting on confirmation that they will be admitted. The hospital benefits as well, since when they send a referral to that the skilled nursing facility for a different patient, the skilled nursing facility has a bed available because they aren’t holding one unnecessarily.
They are going to be able to use this information in practical ways. But it’s important that the information is delivered into their workflow rather than every hospital adding another place that the skilled nursing facility needs to log into and look at. It’s hard to do the right thing in using that information if you put barriers in place.
How has the company’s focus changed with the pandemic and what new requirements to you expect from customers?
We’re seeing more focus on electronic communication. For example, we have a product called CarePort Guide that helps patients and families make decisions about post-acute care. It has things like the quality scores, pictures, and virtual tours. We built that tool because even pre-pandemic, it was the patient’s adult children who were making decisions about where the patient would go. We’ve seen a huge spike in use of that platform because now hospitals don’t allow visitors. We’re seeing more usage because of the need to do virtual tours since nursing homes have also locked down.
Our tools allow the hospital and post-acute care providers to communicate. Instead of somebody at the admissions office leaving a phone message for a hospital nurse case manager, they can communicate bi-directionally since both of them are on the platform. There’s just a lot less friction. We’ve seen the number of electronic communications spike because the nursing home staff are no longer able to go into the hospital to screen these patients or to talk with them in person prior to receiving that transfer.
We’re going see, beyond just telehealth, more and more electronic delivery of care in a lot of other areas. Even in areas that people wouldn’t typically think about, such as the communication among the hospital, the post-acute care provider, and the patient who is making these decisions.
Since our platform connects hospitals and post-acute care providers, we are tracking patients from the time they enter the ED all the way through their recovery period. A lot of our customers are asking us to track COVID patients to understand how to prepare post surge. What will their recovery needs be? People didn’t really know. We are starting to aggregate data across all of the 1,000 hospitals that we work with and all of their EHRs — Epic, Cerner, Meditech, Allscripts, and all the electronic systems used on the post-acute care side. We are tracking something like 22,000 patients from the minute that they enter into the ED through their inpatient course, through their ICU course, through their post-acute course. We’re starting to see trends that are helpful for our customers as they are managing these patients across the continuum.
Do you have any final thoughts?
As we move into this new normal, we are seeing the interdependence between hospitals and post-acute care providers. Although the interoperability rule has been delayed for good reason, people will start to see this rule as being really important coming out of the pandemic, or going into the second wave of the pandemic. There’s a real need and opportunity to be able to share patient information in real time so that we can monitor and track these patients and communicate better with one another. That need is being crystallized in the heightened reality of COVID.
The interoperability that Dr. Hu discusses is so challenged that you ‘might’ say it does not work. At least it has been demonstrated not to work to the point that most facilities would rather use a fax than rely on anything greater than an ADT, certainly a discharge summary, transfer of care, or other C-CDA constructs are not interoperable. Few vendors are participating in cross solution interoperability testing and engineering (including CarePort) so the fidelity of semantic clinical exchange remains abysmal.
What fidelity to you expect in the exchange of semantic clinical data? If you are thinking in the terms of 9s (90%, 99%, 99.9%) then you would be sadly disappointed. Demonstrations routinely show fidelity in the 70 percent range per data item, if there are hundreds of data items per document then it becomes almost impossible to have an error free document. On top of the low fidelity rate for the data itself there is also the problem of information selection — which is also challenged in many EHRs — all, or nothing. Think about a long term patient, how do you get to the information you need when you get 10 — 100 pages of “information”?
Now, take that to the analytics that Dr. Hu is talking about in this article — if you have garbage data, you will have garbage analytics — so I sincerely doubt she is getting the type of data and information, even in the aggregate, that would allow for quality analytics.
Claiming that the bi-directional communication problem is solved is also just a bit disingenuous — it is still more often than not, fax and phone calls. Until 360X and more robust FHIR solutions are implemented this will be a huge problem.
Happy talk is fine, that is what CEOs do, but instead of taking the opportunity to ask hard questions of the industry that Dr. Hu is ‘connecting’ it was all “Mission Accomplished” and golden retriever puppies.
Based on your input, my guess is that you’ve at least occasionally brushed up against the Direct Project and Carequality/eHealth Exchange.
Do you have thoughts on either those pull or push-domain exchange models as they exist today? Am interested in your evaluation of both transport and content, and any vendor-specific notes would be gravy.
Yes, brushed up against both. Direct Trust is doing good work but it is mostly around how to get the next generation out and building standards for the solutions to use. Who uses, how they interpreted, and what they exchange via those standards will dictate how effective their efforts will be.
Carequality is an interesting concept but if you don’t solve the underlying interoperability failures then you are back to the same problem. I know several EHRs are trying to get together and do formative testing between themselves but it is slow going and has been recently hobbled. We know with certainty that certification testing is not sufficient to solve this problem
Here is a challenge for you. Can you exchange your top 100 problems, allergies, medications, procedures, labs, results at 100% accuracy with the top three ambulatory, acute, and SNF solutions? Can you then create a longitudinal record for the top 10 most common conditions with and without co-morbidities — and exchange that with 100% accuracy? Lastly, bring those records back — watch the reflection, refraction, distortion, and echoing that occurs when you try to incorporate the data and information back from a second (or third) solution.
Push or pull — I think you need both. A disease registry is sent a CCDA to initiate a case and the registry reaches back to the source via FHIR to get their periodic updates. CDS and FHIR would be brilliant — perhaps a CDS that is focused on emergent diseases — that can be updated quickly to allow for emergent knowledge. All too relevant today, but it is also last year, the year before, and the year before that (e.g. vaping, zika, opioid, and ebola) so if we don’t invest in that class of technology we will always be handicapped.
The food and drug industries have solved the data problems — if healthcare also solved these problems we could get from data to information, to knowledge, to wisdom — but it starts with the data.
Pardon the passion 🙂
Brody