CIOs focus on proving IT's usefulness to drive adoption

After spending billions of dollars on new technology solutions, healthcare leaders must figure out ways to ensure a solid return on investment. That's not always easy, especially when some clinicians continue to resist using digital tools to their fullest potential, or when the technology itself creates barriers.

Pamela Arora and Ken Lee offer real-world examples to show how technology can improve care and lower costs. Managing Editor Matthew Weinstock recently spoke with Arora, from Dallas-based Children's Health, and Centura Health's Lee about their strategies as health IT leaders. The following are edited transcripts.

MH: How has the shift to offer more care away from the hospital impacted your IT planning?

Arora: We've developed a robust telemedicine program and an infrastructure around it. We want to deliver care to the patients in schools, at home, and in rural areas across the state and through our tele-NICU and tele-ER programs.

We're in over 100 schools today. We also have telemedicine going into the social work aspects. So not only from a school nurse standpoint but we are also meeting the emotional needs of that child.

It's really key to our strategy and we believe that we need to help children in wellness as well as when they're sick within the walls of our hospital.

MH: When you say that you're in 100 schools, is that clinicians based in the school, telemedicine or a combination of the two?

Arora: It's telemedicine into the schools. I think that's very key because we're basically leveraging the infrastructure that's within our community to be able to reach those children so that they can leverage our specialty skill sets when they're needed.

MH: What are the conversations like with the rest of the C-suite and board about reaching deeper into the community?

Arora: They've been incredibly supportive. It's mainstream technology, so we're very confident in its reliability.

There are other aspects of the technology when we connect with small physician practices—one- and two-doc practices—where in some cases the security hasn't been at the level that we'd like it to be. In those instances, we have partnered with our small physician practices to work with them to find security solutions that … can scale to their needs at a price point that suits them.

There's a program called CyberAid. We've partnered with a vendor that, for the price of a cup of coffee a day, was able to provide monitoring and services to address the malware that was coming into these small doc practices. In some cases, we were hosting some of those physician practices on our EHR and it created a better environment, not only for those physician practices but across the whole continuum of care, as far as that data flow.

Those are areas where we try to get creative.

MH: Part of this is also interoperability. Talk a little bit about data exchange with those pediatric providers.

Arora: We believe in the importance of delivering information and we sign up to any method of sharing that we can. For example, we share information around disability determination with the Social Security Administration and, while that's a small subset of our patient population, that actually helps them get reimbursed much more quickly and it saves us administration costs.

That's one example, but we plug into a number of different exchanges and we leverage the interoperability capabilities of our EHR. We find that it's very capable of sharing records. But even when you're doing that, there are still challenges with interoperability between different systems.

I'm going to give you a couple of examples. So a simple one with babies and naming conventions. If you have information like the (parent's) email address and the phone number, your ability to match a baby, to make sure that you're certain that you're matching the record appropriately, is increased by 10%. But many times, the standards don't require that you have an email address or a phone number.

From a lab standpoint, there is currently no standard way to differentiate the type of lab result. So for example, there could be a radiology result or a microbiology result. Both of them are sent in a (consolidated-clinical document architecture) document. If the C-CDA specifications were to differentiate between the various types of results, vendors, let's say Epic, Cerner or Allscripts, could better integrate outside data into the patient's chart, giving an overall better experience to the clinicians and the patients. Without the specification, it's difficult to know what type of results are being sent … and clinicians are forced to look at all results together without categorizing them, sorting them or filtering them, and when they intermingle them, they can get confused.

MH: You come to healthcare with an outsider's perspective. Are there investments you are making to break the mold of what healthcare has traditionally done?

Lee: No. 1 I would say is around Epic. One of our physicians asked, “Are we practicing Epic, or are we practicing medicine?”

We've gotten into some detailed analytics around the performance of our physicians and their use of Epic and we've been partnering with Epic around how we measure that. It includes things like leveraging data science to derive an Epic score for every physician. The data includes things like Epic use after hours, percentage of orders using defaults; things that indicate how Epic is being used.

We've taken that information and have started a program where we're providing an Epic coach to every single one of our physicians in an ongoing relationship to increase the physician's Epic score.

We're really early in the process, but the feedback has been very positive. We're starting to receive emails thanking our coaches for their tutorial sessions. We're also partnering with Epic on Lean processes to improve workflows, so the combination of improved proficiency and workflows should help our physicians around driving toward the Triple Aim goals.

MH: Do you share that Epic score, similar to the way you would share clinical quality scores?

Lee: We're starting to trend the data. My goal is to use the trending, the summary-level information, to understand how we're performing across different specialties. It's also to look at our performance across our coaches and help inform them on how to improve the population of doctors that they are responsible for. We're in our infancy, but I would expect that we're going to realize some really great results. In fact, I was just reading an article about another health system that had done this and they experienced, I think, a 47% increase in proficiency across their users.

MH: How do you see this interacting with your goals for the Triple Aim?

Lee: The more time we can give back to our users, the more time that they're going to be able to focus on the patient experience and impact better outcomes that are going to impact the cost of healthcare. On the population health piece … we've been working with Epic on improving capabilities around care management, around the risk profiling; taking the information both from payers as well as Epic, and then being able to risk-profile the populations.

MH: Are there specific segments of your patient base that, as you look at analytics more broadly, you want to roll some of your analytical work out to and build a broader population health environment?

Lee: We have a partnership with UnitedHealthcare on a program called Colorado Doctors Plan, which is still a fee-for-service model, that's taking a look at more proactive management of the member population to reduce the cost of care as well as increase the whole customer experience. It's a narrow-network product that we partner on.

As a part of that, we've taken algorithms—both data we have from United as well as some of the machine-learning practices that we've pulled from other healthcare systems—to be able to come up with a risk rating for that population. The investments that we're doing on machine learning and AI right now are pretty tactical. As a part of our Centura Health 2025 strategy, we've got a plan to increase the maturity and size of our market, and one of the things we're working on is a prioritization process of how we're, as an IT organization, staying aligned with our business goals and focusing on current-year priorities aligned to that 2025 strategy, and then making sure we're aligned on what we're not working on.