HIStalk Interviews Justin Sims, President, CareMesh

Justin Sims is president and COO of CareMesh of Reston, VA.

Tell me about yourself and the company.

The business of CareMesh is all about helping healthcare organizations break down communication barriers so they can better coordinate patient care. We do this by helping them solve a number of related problems. Firstly, finding the providers that they want to engage with, then exchanging patient information with them, and then coordinating actions and decisions, both within the organization and between organizations. I guess if you wanted to give us a label, I would call it care coordination.

As to me, I’ve been doing this with CareMesh for the past five years since we started the company, but I’ve worked on a number of similar problems in other health tech companies over the last 20 years. Before that, I did about 15 or 20 years in the telecom industry. But I can tell you for a fact this is a lot more fun.

How well do health systems communicate with their local, community-based physicians?

The truth is that the dominant form of unstructured communication in healthcare today is still stuck in 19th century technology. And I meant it when I said 19th century technology. The first fax, if I’ve got my history right, was sent by Alexander Bain in 1843. That technology is still the dominant form of communication in healthcare.

A lot of people incorrectly think that Direct Secure Messaging is the answer to that, but it isn’t, for a couple of pretty big reasons. The first is a practical one. Only about half of physicians in the community have Direct addresses and can technically use that form of communication. But the second reason is more a structural one, and that is that the standards for Direct messaging focused on exchanging CCDAs, structured patient records, and they didn’t focus on unstructured communications, which is where a massive flow of communications actually takes place. That’s why communication with community providers is still pretty challenging.

It is inevitable that healthcare is scorned for being behind other industries, but isn’t part of the reason that it is more limited by HIPAA and privacy concerns that preclude using the usual consumer-grade tools?

I think that’s right. We are not like others, for a couple of reasons. One is the privacy concerns, although I would say a fax is not exactly the most secure form of communication in the world from a HIPAA perspective. It was one of those things grandfathered. But the other thing that I think makes healthcare communications complicated is that a patient record doesn’t fit into a PowerPoint, an Excel spreadsheet, a Word document, or a PDF document, which is how other industries often exchange information and data. A patient record itself is uniquely structured. 

That’s why it’s important to give credit where due. The government’s efforts around interoperability and getting unique forms of patient record exchange established are important. But again, I come back to these unstructured communications, these things that are so important to coordinating patient care. Not everything fits into a CDA. Not everything fits into a patient record when you’re coordinating a referral or a discharge. Sure, you can send structured patient record information, but that’s only a small part of what needs to be done to properly coordinate with an outside provider. That’s where HIPAA and the lack of a common messaging standard for plain old communications make certain types of communication difficult in healthcare.

Unlike hospitals, most community-based practices don’t have big capital budgets, training and technical support teams, and a commonly used brand of EHR. Does that make communication with them harder?

Yes. It also needs to take into account the workflow within a practice, that there are as many different workflows as there are practices. Some community providers are organized centrally, with a coordinator taking care of everything coming in and protecting the doctor from information overload. Others are organized with perhaps a nurse practitioner or a clinical assistant helping the doctor. Others are organized where the doctor really wants to take care of things themselves.

One of the challenges is that hospitals often attempt to impose their solution to clinical communications on the provider in the community without actually taking account of the fact that they are organized differently. They have different information needs, and they certainly have different workflows. All of that needs to be taken into account when designing a communication strategy, for example, with a hospital and its community providers.

What does a hospital user see and do differently when their organization’s Epic or Cerner system is integrated with CareMesh via APIs?

There are a couple of key things that both the hospital and the community providers get out of that integration. From a hospital perspective, we literally take care of all communications from and between community providers. The hospital doesn’t need to worry about maintaining a directory, dealing with message failures, or having gaps in who they can communicate with.

Our delivery rate is pretty astounding. Only in one out of every 300 communications that the hospital gives us to deliver do we have to go back to the hospital and explain that we can’t deliver it for a particular reason. Very often, that’s because the practitioner has retired or moved out of state or something of that nature. The first thing that the CareMesh solution does is solve or address the problem of the administrative side of communication.

But from a community provider perspective, we don’t force any particular solution on the community provider. We give them multiple options. If they want certain communications delivered by fax or by Direct Secure Messaging, or they want to log into a portal to look at information, they can choose that. If they want information sent centrally or to a delegate of the doctor or to the doctor themselves, they can choose that. We have created a system that allows the community providers themselves to say that if you’re sending a referral, don’t send it to me, send it to this person instead, and I’d actually like it delivered by fax. But every now and then, I’d like to go to a portal and download the patient record. We give as much flexibility to the community providers as possible so that the communications are relevant to them.

Direct addresses were never as simple as they seemed given that a provider can have multiple employers, multiple roles within an employer, and may change employers where some messages should remain within the original practice. Is there a better way to manage the use of the Direct address system?

It’s certainly still a challenge. DirectTrust has done a great job at consolidating a directory of Direct addresses, but the information is only as good as the sources. There’s a chain, if you like, to get the information into the DirectTrust directory. It starts with the organization that employs the doctors and other healthcare professionals and then moves to the HISPs, the health information service providers that run the Direct messaging platform. Then it moves to the DirectTrust directory. That chain can and does break, so often Direct addresses are out of date or they’re not complete.

There are about a million doctors in the country, and we’ve been able to match Direct addresses with a little less than 50% of those. Some of those Direct addresses come from DirectTrust, some from hospital systems, and some from NPPES, which is now collecting that information. We get the data from a variety of sources. But there’s still half of the doctors in the country that don’t have a published Direct address. That represents a fairly big challenge when relying on Direct for communication purposes, in particular, CDA exchange.

One benefit of a fax number is that the receiving organization receives everything that is sent, then intelligently routes each message appropriately. Is there an electronic messaging equivalent?

There are certainly tools around Direct to support that within a number of EHRs. EHRs are able to route messages according to particular rules. One of the things that those of us that are close to healthcare communications would advocate for is Direct working a little bit more like POP email. But that’s really dependent upon the EHR vendors adding more sophistication to their messaging platforms.

In the longer term, there is a strategy that the government is advocating for to create what are called FHIR endpoints. These are essentially web addresses that would allow one healthcare organization to post information into the EHR of another healthcare organization. That could be used for structured communications, like exchanging patient records, or unstructured communications as well. So there is a long-term strategy that could over time close some of the gaps, but it really is a pretty long-term strategy. A lot has to happen for that to solve all of the problems that we’ve been talking about.

What are the challenges involved with maintaining a provider directory?

Maintaining a directory of about 5 million people is a labor of love, but it’s also a big data challenge, and it’s a constantly moving feat. Let’s even narrow that down to just a million doctors. If each of those doctors change a piece of information within their profile every two or three years, such as a change in a location or a communication endpoint, you’re dealing with hundreds of updates that need to take place every day. That’s not something that can conceivably be done manually alone. We heavily rely on big data technologies to pull together the most comprehensive and best view we possibly can of that provider, and we use all sorts of specialized techniques. There’s a concept called master data management that we use to make sure that we present the best possible information possible.

We have also realized that it’s important to design a directory for all, not just some, use cases. We have insurance customers. We have state and local health department customers. We have hospital customers. We have HIEs. We feel that ultimately a single resource that can support many, many different use cases is the only way that anyone is ever going to be able to maintain a strong directory. It’s got to be scaled. It’s got to be heavily utilized by many others. It’s through some of the methods that we talked about that you can gradually get to something that is fit for purpose. It will never be perfect. Anything in big data with 5 million records in it, 5 million providers in it, is never going to be perfect. But it should be good enough to meet the basic needs of registering patients, sending communications, and so on.

How do you see the company’s business changing over the next three or four years?

I think our business is going to evolve as the industry addresses and improves generically on some of the challenges that we’ve talked about. But the foundation of what we do is our directory business, which we brand Search. But a directory on its own is going to be quite limiting in terms of what it can do if it’s not integrated with a communications capability. As we discussed, we augment EHRs and other systems and their native communications capabilities by making it possible to communicate with anyone and everyone.

But the area that is perhaps most interesting as it expands and evolves into a new line of business for us is that once people can find each other and can communicate with them, they then need project management tools and workflow tools to help them manage the patient journey through their particular healthcare condition. It’s those project management tools that we’re providing today, we brand them Navigate, that I think stand the greatest chance of developing into scalable solutions to solve these care coordination challenges that we’ve been talking about.