Last fall, HCI Editor-in-Chief Mark Hagland spoke with Gregory Schroedl, M.D., chief medical officer at Northwest Hospital, a 250-bed facility in Seattle. Northwest Hospital, a community hospital, has for the past three years been a part of the University of Washington, under the umbrella of UW Medicine, along with University Medical Center, Harborview Medical Center, and Valley Medical Center (in nearby Renton).

Dr. Schroedl and his colleagues have been deeply involved in leveraging the adoption of evidence-based medicine tools to stimulate clinical transformation. Along with the leaders of fellow affiliated hospitals within UW Medicine, Schroedl and his colleagues at Northwest Hospital have been partnering with the Los Angeles-based Zynx Health to implement evidence-based order sets, with the progress of that work accelerated through Northwest Hospital’s implementation of computerized physician order entry (CPOE) last year.

Working closely alongside Dr. Schroedl in Northwest Hospital’s clinical transformation work has been Mary P. Horan, M.D., the hospital’s CMIO, who came to the hospital 14 years ago, and has been CMIO for nearly three years now. Hagland spoke with Dr. Horan recently about her multi-faceted work at the hospital and her perspectives on the work she’s been involved in there. Below are excerpts from that interview.

Tell me about your tenure at Northwest Hospital and your various roles there. How long have you been CMIO at the hospital?

Officially, it’s been two years and nine months. In terms of my being at Northwest generally, I’ve been at the hospital for 14 years and four months. Initially, I was a hospitalist and practicing pulmonary critical care provider, and got into the information systems area by being involved in a vendor selection process, and then became more involved. My path really paralleled the path of the development of the CMIO role in the industry as a whole. And I had been chief of our medical staff for a couple of years, years ago. And in terms of the issues right now and helping the physicians understand the issues as a CMIO, I transitioned from a position where I was 60/40 clinical/CMIO nine months ago; I’m now 80 percent administrative and 20 percent clinical.

Mary P. Horan, M.D.

What are you doing in your remaining 20-percent clinical time?

I’m continuing to practice in my outpatient clinic a half-day a week; and I’m job-sharing with another provider on ICU [intensive care unit] pulmonary care or consults. All of my clinical practice is pulmonary-oriented now, and has been for the last 12 years or so. And the hospitalist experience was very beneficial for me as preparation for becoming CMIO, even though I hadn’t done hospitalist work in a long time.

With regard to the successful implementation of evidence-based order sets, would you agree that the two sides have to evolve forward together, organizational culture and implementation?

I definitely agree about the culture; and if you were starting from scratch, you’d absolutely have to work in parallel. With regard to our implementation of evidence-based order sets, we were fortunate to go with the actual online order entry a little later; and the process of having consensus-based, evidence-based order sets in common use, was something that Dr. Schroedl and our quality folks had led for some time.

In other words, you had built that in the paper world already?

Yes. And it hasn’t always been smooth, but it’s worked out generally in the long run.

Dr. Schroedl, in my interview with him last fall, described the iterative process of iterative development with the Zynx people as having been overall very positive. Would you agree?

Yes, I would. I should add that we needed help, because we weren’t so far along early on. And once we agreed with them that we should build the order sets within Siemens [the hospital’s core EHR vendor], that made the process easier. There had been some understanding early on that everything would be developed in Zynx and then flowed into Siemens, but that didn’t really come into being. So we had to decide how to flow the process; and we ended up having to streamline that end of the build ourselves.

What have been the key process lessons learned so far around embedding evidence-based order sets into the CPOE system at Northwest?

I think the ability to view how your order set looks the way it does, for those who haven’t been involved in building the set, is greater in the electronic world than on paper. And the actual action of trying to translate a paper-based order set into an EMR is not one for one. So having a very good understanding of what your EMR can do, understanding that there are opportunities for embedding not only evidence but also other types of decision support into the order set, and making it comfortable for users—because we had already been through the order-building process on paper, I actually think that that was tremendously helpful for us as we went into the CPOE implementation.

What are the optimal ways to get primary care physicians and specialists to buy into the process overall?

That’s a great question, because approaches that work for some physicians might not work for others. And if you have a mixed-employment system in your provider network, the ways in which you can encourage participation for those who are employed versus affiliated physicians, are very different. So it actually becomes a marketing job, to be totally honest; you need to know who your audiences are and how to communicate with them. And there’s a certain stream of traffic that ends up compelling some latecomers forward. So knowing your audience is very important.

It feels to me as though we as a healthcare system are very early in this journey of evidence-based medicine in the electronic world?

I think that’s very well said, and speaks to some extent to regional variations in care delivery. Healthcare being regional is less and less of a valid argument, if one has the tools to understand what people are doing across the whole [U.S.] healthcare system. But in parallel to that, it’s important to be humble about the fact that there’s not always evidence around what we do, so you are reliant on consensus for best practices in many areas. And demanding that we all do things exactly the same way doesn’t work, either; so it’s important to be honest about what the drivers are.

So physician and administrative leaders have to be honest about what’s evidence-based and what’s consensus-based, then, right?

Yes, absolutely. And there are challenges in many areas—one example being diabetes control in the ICU. For a time, there was great enthusiasm about that, but in practice, it wasn’t always totally accepted. So it falls to providers and individual hospitals to continue to be diligent and to respond constantly to the truth about how these are living, breathing processes. You know, you think you’re done, and you’re so far from checking things off a list, because everything is always up for reconsideration. And as the provider community becomes more savvy, they’re actually able to help us design things better for them. So you have to be willing to be collaborative in that regard, as well.

What would your advice be for other CMIOs around all this?

I think having a leadership structure that clearly understands what the goals are is important, because there are obviously things that every organization has to do to encourage participation, to encourage people to understand that this is a common goal. And a CMIO by him/herself will not be able to move an organization forward without broad leadership. And it really is team leadership, getting your support going in all areas, and bringing in knowledge about how you can do things better, because each part of care comes out of that team-based understanding. And if you don’t have everybody buying in to support that, you won’t be successful.

Another success factor is having a great vendor partner, having an organization working with you that has a great vision. Zynx has really been ahead of many other vendors not only in terms of order sets, but also in terms of understanding of how the world is changing, and that this is an integral part of getting to the next level, whether it means becoming more cost-effective or improving care quality, and preparing for other transitions such as ICD-10 and meaningful use. And having their advice and vision and support has been really great.

What would your advice be for non-clinician CIOs?

CPOE is not a project with a start and end date. There’s always that focus on getting it up and running, but you need to ensure as you look at key resources, that some sort of ongoing support and maintenance are budgeted, because it’s going to be that way forever. And the other really huge issue is that there’s no tolerance for equipment failure or downtime. Another key is to ensure that the devices work as fast as possible, because if you’re going to tell people that they have to change the way they do things, then you have to have total reliability.

Do you have a team that reports to you?

There’s been an interesting evolution in that area: it actually took me two years to get onto the org chart. There are many things about the upcoming changes and what we’ve already done; and that requires understanding from the leadership and the CIO perspective, that you’re creating an individual who didn’t previously exist. I don’t have direct reports, but I’m relying on a couple of committees. So it’s creating useful committee structures within your medical staff and your organization. I report to Dr. Schroedl, and have a close relationship with our IT director. I do meet on a regular basis with our IT director and IT support team.