HIStalk Connect Interviews Terry Edwards, President & CEO at PerfectServe

Terry Edwards is the president and CEO at PerfectServe of Knoxville, TN.


Tell me a little bit about yourself and PerfectServe.

PerfectServe has built one of the most advanced clinical communication and care team collaboration platforms available in the industry. We’re focused on enabling communication-driven workflows that bring together care team members.

Our core differentiating capability is the capability we call Dynamic Intelligent Routing. It allows us to receive our communication, whether it be a secure text message, a phone call, or data out of a server system, like an EMR. Based on the variables at play, such as time of day and clinical situation, we’re able to automatically identify and then route that interaction to the right care team member at that given point in time.

We’re deployed in some 140 to 150 hospitals and some 20,000 practices and acute care facilities. We’ve been in the space quite a while and are well established.

Describe what’s happening with the Joint Commission’s position on the texting of orders.

It’s interesting and it’s in flux. Joint Commission was pretty specific about wanting to avoid this. They prohibited it and wrote rules around it a couple of years ago.

The problem was that those rules were inconsistent with the way people are communicating today. Their concerns are valid. A couple of months ago, Joint Commission attempted to write requirements that addressed many of those concerns. For example, being able to ensure that whatever system is being used for the texting of orders that there’s a way to authenticate the user, making sure the person initiating or receiving the communication is who they’re supposed to be.

They were good guidelines. I wasn’t sure if they were comprehensive enough, because we know that there’s variability in the communication process. It’s not always as simple as, "I think I’m going to write this order and get it to you." You might be the right person right now, but are you actively on shift at the time that this order should be received, or is it somebody different? Those are things that still need to be addressed.

Anyway, they’ve pulled back, and it appears that someone or a group of people at CMS felt that they needed to weigh in. So now we wait. We’re expecting them to issue revised guidance late summer or early fall. I expect that they will come up with a solution, or revised guidelines and recommendations because texting is just a part of how people communicate today. We have had effective ways of dealing with verbal orders for years. We just need to come up with effective ways of dealing with texting orders. Just need the right guardrails.

What is the optimal workflow for receiving and processing a text order in an EHR?

Let’s say we have a situation with a nurse who notices a change in status. That nurse thinks that the physician needs to take action in terms of an order. Ideally, the communication system would have some kind of user interface in the EMR. We haven’t seen that with any of them to date that are comprehensive enough to address the workflows. Upon selecting that physician from the EMR, the system would automatically identify and connect that nurse to the right physician, because it may not be the person who’s listed as the attending, or it may not be the person who would have even seen that patient earlier that day. There can be variability based on call schedules, hospital shifts, and things like that. The nurse would then type up the request and put the appropriate information in the text, maybe even ideally pulling some information out of the EMR screen, and embedding it accordingly and sending it.

Then it would be sent to the receiving physician in a secure way. He or she would receive notification. There would be fail-safe processes built into the notification to make sure that the doctor retrieves it in a timely manner, and then upon reviewing it, the doctor would be able to send a reply back and ensure, that the reply goes back, not only into the record, but also notifies the appropriate nurse, who can then act on the order.

What are some of the common communication bottlenecks that cause delays or negatively impact patient outcomes?

A Harris poll that our organization commissioned last year studied the role of communication under population health business models. There were 1,000 healthcare professionals that participated and 97 percent agreed that effective communication with the broader care team was critical to ensuring population health or ACO-type care delivery models. But there was also strong agreement, with 96 percent reporting, that inefficiencies are creating barriers to success.

What we found is that more than half the time, the initiator didn’t know who the correct care team member was that they needed to contact. That’s because of the variability in the workflows to determine who should be responding to a particular situation at a given moment in time. A lot of variables can be at play there.

About 70 percent of those who responded spoke about disparate IT technology. For example, I might be a nurse and I’m using a nursing communications system, but I’ve got to talk to doctors. I have to use just regular voice communications systems to talk to them and they’re not integrated.

We’ve got organizations that are using two different texting platforms, for example. The technologies are fragmented, which impedes communication.

The EMR was only used 12 percent of the time when communicating with providers inside of one’s organization and eight percent of the time when communicating with providers outside. It really hasn’t addressed many of the communication challenges that we thought that it would.

The impact is meaningful. Two-thirds of the providers who we surveyed indicated that they had experienced delays in care and delays in transitions that were due to breakdowns in communications

What operational value can be derived from performing analytics on the information that flows through your system?

It’s two-fold. The analytics allow us to see if we’re hitting certain utilization targets. For example, we have deployed in enough hospitals that we know for a 300-bed hospital how many communication events should be being processed through our platform. We can fairly quickly monitor if the system is being utilized at the level that it should be. That helps with overall performance improvement. We’re also able to also monitor our compliance against HIPAA.

There’s communication that’s occurring all throughout an organization, but you can’t really measure it because it’s very fragmented. As you begin to move traffic onto a common platform, you can then look at not only the utilization of the platform itself, but then also the modalities that are being used. You can look at, for example, the percentage of communications that are handled in real time versus a secure text message, those that are handled as a secure text versus a page or an SMS. It’s important to be able to address all these modalities because that’s absolutely key to getting the adoption that’s needed.

If you look at applying something like PerfectServe to a code STEMI process or a code stroke, we can monitor response times, for example, but also monitor the adoption of best practices contact methods as well. Then use that data to coach physicians, nurses, and other providers as time goes on. It’s continuous process improvement.

How do you manage and present secure messaging to optimize physician workflow?

That’s where the Dynamic Intelligent Routing comes into play. I mention this because three years ago or so, the surgeon consumer adoption of smart phones and text messaging went into healthcare. Then in 2013, the HIPAA Omnibus final rule went into effect, which really put more teeth into the HIPAA law, especially around breaches. That created a lot of concern, leading CIOs to think that they had to buy a secure messaging solution.

Overnight, there was a ton of venture capital money that went into the market. We’ve seen some 100-plus secure messaging startups, but what we haven’t seen since that time is high adoption of secure messaging amongst physicians outside of those small groups.

The reason for that is that secure messaging, when it’s purchased and deployed by itself, requires two things. It requires the initiator to always know who it is they need to contact by name. Second, it requires the recipient to be accessible and available at all times.

There’s so much communication directed at physicians and other people based on clinical situations. I might think that I need to contact Angela about a routine consult because she’s a cardiologist, but Angela’s workflow is that she only handles the urgent consults when she is at work during the weekdays. Her physician’s assistant Joan handles the routine.

There’s the routing of the secure message to the right person based on the workflow. Then in terms of physician accessibility, the kinds of communication that physicians are dealing in the delivery of care oftentimes require action. For patient’s safety and medico-legal liability issues, they don’t want to receive things that they can’t act on. The point is that secure messaging by itself is inconsistent with the different workflows that are common in care delivery. That’s why there’s a need for this Dynamic Intelligent Routing that PerfectServe provides.

Do you have any final thoughts?

Most everybody in the industry, and certainly the people within our category, are pretty excited about the fact that the Joint Commission has revisited this topic around the texting of orders. Our platform is unique to address this in the safest way. We look forward to the final guidance after they consult with CMS. We’re awaiting it and we know that many of our clients are as well. We think they’ll make the right decision and recommendation.

↑ Back to top

Founding Sponsors

Platinum Sponsors