David Bensema, MD, CIO, Baptist Health, Chapter 1

As an internal medicine physician, one of the most important lessons David Bensema learned was that if he was going truly engage with a patient, he first needed to establish trust. As CIO, he employs that same strategy to engage with physicians — something that doesn’t always come easy, even for an MD. In this interview, Bensema talks about what has surprised him most during his first year as CIO; the tough part when it comes to creating a task force of the “best and brightest”; and the strategy his team is using to roll out Epic across 7 hospitals. He also discusses why he doesn’t like the word ‘optimization,’ how he keeps the end user’s needs as top priority, and the one thing that needs to change when it comes to CIOs and project planning.

Chapter 1

• About Baptist Health
• 15-month Epic rollout by region
• “We’ll be troubleshooting and making adjustments that are required immediately.”
• Creating & populating Epic record in physician practices
• Staffing for go-live — “We need them present.”
• “I don’t know that we can ever optimize, but we can always improve.”
• Personalization labs

LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED

Podcast: Play in new window | Download (Duration: 13:31 — 12.4MB)

Subscribe: Apple Podcasts | Google Podcasts | Spotify | Android | Pandora | iHeartRadio | Stitcher | Podchaser | Podcast Index | Email | TuneIn | RSS

Bold Statements

The team’s been doing a great job, but we know that there are glitches, and so we want to be able to find that with a very supportive environment and a smaller cadre of patients and staff before we initiate the rest of the rollout.

Being a physician myself, I love seeing when physicians are able to continue in independent practices, and so as we do our Epic implementation, we’re very aware of that and how we support those independent physicians and involve those physicians in the inpatient build.

We need them not only for staffing, because obviously we don’t want to have any impairment of patient care; we want to make sure that the quality and the safety of our patients’ experience is paramount. But we also need them to learn how to use Epic in their clinical settings or in their everyday roles.

‘Optimization’ is not my favorite term because I don’t know that we can ever optimize, but we can always continuously improve. So we’ll apply continuous improvement principles and go live with the next hospitals with the improvements identified in place.

Gamble:  Hi David, thank you so much for taking some time to speak with us today.

Bensema:  My pleasure, Kate. Thank you.

Gamble:  Sure. To give our readers and listeners some information, can you just talk a little bit about Baptist Health — what you have in terms of hospitals, ambulatory care, things like that?

Bensema:  Sure. Baptist Health is a seven hospital system in Kentucky. We cover the breadth of Kentucky, from Paducah in the far west to Corbin in the far southeast. Our hospitals range in size from 100 to 450 beds, with a total of approximately 2,100 acute care beds in the system. We have currently one up-and-running transitional care unit, LTAC, and we have two long-term acute care facilities in Madisonville and Paducah that are going to be going online shortly. We also have a psychiatric unit in our Corbin hospital, and we have approximately 180 ambulatory sites of care, which includes imaging centers, ambulatory surgery centers, physical therapy offices, and also primary and specialty physician offices.

We employ 450 physicians — probably plus 10 by now because it moves so fast — and an additional 250 to 280 non-physician providers, APRNs and PAs. We have 17 currently retail sites in the Wal-Mart supercenters staffed by APRNs; they’re limited service clinics. We also have approximately six urgent care centers in the Louisville, Kentuckiana region and two additional in the Lexington/Richmond area and one in Paducah. So we cover the full spectrum.

We have a very active home health agency that is in the process of assimilating a couple of purchases to expand our breadth of support even further. We do not have skilled nursing facilities. We align with and affiliate with skilled nursing facilities throughout the state to provide that portion of the continuum of care. We deliver one in four babies in Kentucky, and make I believe one in four cancer diagnoses in Kentucky. We just finished doing our update of the master patient index for Epic; and we have 2.3 million entries in a state of 4.2 million population. So we touch a lot of lives and are very active in the state.

Gamble:  Okay. Now how long have you been CIO there?

Bensema:  I am just barely over one year in the role of CIO. April 24 was the anniversary of accepting the position.

Gamble:  And prior to that you held the CMIO role?

Bensema:  Right. I became CMIO on July 1, 2013 and held that position, and then added the CIO role to it in April of 2014, and continue to have both those roles.

Gamble:  Okay. So that plays into what I wanted to talk about with the clinical application environment. At what point would you say the Epic rollout is right now?

Bensema:  We are just beginning phase 4. If you know the Epic world, it’s the testing and implementation phase. We are still immersed in the build. Most of the ambulatory clinical content and order sets have been built. The acute phase clinical content and order sets are now being built. We will be doing testing of those over the summer, with a target date of July 27 to have completed that testing. We have corporate goals to have the ambulatory workflows and testing and applications all ready to go on August 31. That’s the end of our fiscal year, so we need have 90 percent-plus of those ready to go. We’ll actually make that quite handily at this point.

As you know, in the build process, it’s always touch and go, but we are far enough along that we are very confident we’ll make that goal. We’ll have our first clinic go-lives October 27, 2015 with two early adopter clinics to make sure that we’ve got things put together the right way and can find the glitches that will inevitably be there, though we are very hopeful and confident that those will be minimal. The team’s been doing a great job, but we do know that there are glitches, and so we want to be able to find that with a very supportive environment and a smaller cadre of patients and staff before we initiate the rest of the rollout.

Gamble:  What’s the approximately size of those clinics? They’re small?

Bensema:  The two clinics are primary care ambulatory practices. Both are family medicine, and each has five physicians and three non-physician providers, so a total of 10 physicians and six non-physician providers will go live October 27.

Gamble:  That strategy makes sense as far as starting with a clinic of that size, so then whatever does need to be adjusted, fixed, whatever, it becomes at least a little easier to move on from there.

Bensema:  Right. We’ll go with two waves in the Kentuckiana region — for us, Kentuckiana for us means Louisville and La Grange, where we have hospitals and we have physician practices in a kind of penumbra around each of those hospitals. And so there are approximately 150 providers, physicians and APRNs, in those two sites, those two markets, in the various practices, and so we’ll do two waves — a January and a February wave of go-lives — in those practices before we come live in the hospitals.

Gamble:  Now the entire rollout, how long is it in total? How many months?

Bensema:  Approximately 15 months, starting October 27 of 2015, and the last hospital go-lives will be January 17, 2017.

Gamble:  Okay. Do you imagine your strategy with the hospitals will be similar to the clinics as far as going on a one-by-one basis?

Bensema:  It’s going to be in regions, so the Kentuckiana region will first get the physician practices up and running with a focus on the ambulatory primary care practices being the earliest wave so that the Epic record begins to be created and populated. In addition to whatever we do in the both electronic and manual abstraction conversion process, we want to start having the primary care physicians build up the problem lists and the medication lists, etc., so that when the hospital goes live on March 5, 2016, there will already exist a record for a large percentage of our patients — certainly not all of our patients, because we have a number of very highly valued affiliated independent physicians. Being a physician myself, I love seeing when physicians are able to continue in independent practices, and so as we do our Epic implementation, we’re very aware of that and how we support those independent physicians and involve those independent physicians in the inpatient build.

So we’ll have completed all of our training in February of 2016 and will go live March 5, in those first two hospitals, and then the next region we’ll start with physician practices in April and May. In the Lexington and Corbin and Richmond area, which basically goes north/south on I-75 through Kentucky, we’ll have those physicians all completed by the end of May and go live in mid/late June—after Father’s Day and before the 4th of July.

You have to time the go-lives around the major family events and the holidays. We want our staff to forego their vacation time during the two weeks of the go-live, because we need them present. We need them not only for staffing, because obviously we don’t want to have any impairment of patient care; we want to make sure that the quality and the safety of our patients’ experience is paramount. But we also need them to be learning how to use Epic in their clinical settings or in their everyday roles. So we freeze vacation time during the go-lives, avoiding holidays and avoiding major family events. But we’ll do the two hospitals, Lexington and Corbin, in June, and then move on to the west region and bring the physicians up in July and August and bring up Baptist Health Paducah in September.

That leaves off two of our hospitals, Baptist Health-Richmond and Baptist Health Madisonville. Madisonville is on a separate contractual agreement with the current inpatient vendor, and so they’re timing was different than the first five hospitals. And Richmond is on a totally different electronic health record on the inpatient basis than all the other hospitals, so doing it separately made sense as well. So we’ll bring Madisonville and Richmond up in January of 2017, and those will be the last two of our seven hospitals to go live.

Gamble:  Right. What do you picture that the process will look as far as going with that first region and then applying whatever would need to be applied to the next hospitals? How do you think that will be mapped out?

Bensema:  A couple of things. One, obviously, even as you’re going live, you have the command center, and we’ll be troubleshooting things and making adjustments that are required immediately. There’s always going to be one of those where you find out we left a gap and we’ve got to close that gap. So we’ll have our teams there to work on that. But then we will take the lessons learned from those two weeks. We’ll be getting feedback from all the end users and we’ll do two things with that. One, we’ll revise as needed and able before the next one — there are two months between, but every third month cycles gives us some time for revision and improvement. We decided this past weekend that ‘optimization’ is not my favorite term because I don’t know that we can ever optimize, but we can always continuously improve. So we’ll apply continuous improvement principles and go live with the next hospitals with the improvements identified in place, and still learn more, improve again, still learn more and improve again.

And even after the last go-live, you come back for consolidation of the learning, consolidation of the adoption, making sure that you’re finding where you’ve got pain points. We’ll do that in each of the hospitals at 60 to 90 days post-go-live. We’ll have a secondary team that sweeps back and does that cleanup operation to find out where you’re struggling with adoption, where the bottlenecks are, and make sure that we enhance the use and understanding. 

We have a process called personalization labs; and it’s not unique to us, it’s an Epic thing, but it’s also used by a lot of other EHRs where once you’ve trained somebody to do the basic use of the product, you want to do a personalization lab to where they can create their own smart sets, order sets, smart text, and be able to conform the screen view in a way that works with their eye flow and their workflow. So we want to make sure that we not only do that before they go live, but then come back at 60 to 90 days and say, ‘okay, where are you struggling,’ and be able to help them personalize in a way that overcomes some of those barriers.

Gamble:  Right.

Chapter 2