One-on-One with Virginia Commonwealth University Health System CIO Rich Pollack, Part I
Located in Richmond, Va., Virginia Commonwealth University Health System (VCUHS) is a $1.4 billion organization that includes the VCU Medical Center, a top-ranked hospital by U.S. News & World Report. MCV Hospitals is the teaching hospital component of the VCU Medical Center, which also includes outpatient clinics and a 600-physician faculty group practice. The 779-bed VCU Medical Center is a regional referral center for the state, and is the area’s only Level I Trauma Center. Recently, HCI Associate Editor Kate Huvane Gamble spoke with VCUHS CIO Rich Pollack about a number of issues, including the health system’s major wireless transformation, the challenges involved in balancing different clinician preferences, and the decision to offer EMR access via smartphones.
KG: How long have you been CIO at VCU Health System?
RP: About three and half years.
KG: So the wireless transformation started pretty much as soon as you got there.
RP: Yeah, I got here in December of 2005, and at that time, the extent of wireless was pretty limited. We had a pilot area on the inpatient side, maybe about 2,000-2,500 square feet that was lit up with some Cisco-based 802.11 point solutions for data. That was more of a pilot for the technical guys to gain some understanding and knowledge around what it would take. We had a lot of desire for wireless from our customers and from the clinicians and nurses who wanted greater mobility, but that’s all we had that that time. We had an older version of Ascom phones — the portable nurse phones — that was in place in a number of the inpatient units, and then of course we had point solutions for telemetry of patients in specific areas. There was also use of pagers throughout the organizations and cell phones in some areas, but just the individuals’ own cell phones.
So what I started in 2006 was a push toward an aggressive deployment of wireless infrastructure, both on the inpatient and outpatient sides. On the inpatient side, we have about a 780-bed acute hospital.
We were in an early stage of our EMR roll-out; we were less than halfway when I got here. But it was pretty obvious that in order to affect completion of the electronic medical record — in other words, completely roll it out with capability and functionality in the patient space — we would need to provide at least the facility for wireless use. It kind of went hand-in-hand. So in having that understanding and also that feedback from our clinician partners, I quickly determined that I had to get the wireless out in front of this EMR roll-out. So that’s why fairly shortly after I got here, in early 2006, I gave the directive to my technical staff and my CTO to aggressively pursue rolling out wireless in the hospital and the clinics, primarily focused on the 802.11 data piece.
KG: I imagine you had to lay quite a bit of groundwork in terms of the infrastructure.
RP: We had to invest in both the infrastructure and the equipment. It required upgrading of our backend wired switches in many cases. Because at some point in time, any wireless infrastructure eventually connects t o the wired infrastructure. So if you deploy pretty extensive wireless infrastructure, you find that you are sometimes putting a significant load on your backend wired infrastructure and you need to upgrade that. So we went for a significant upgrade of network switches, the network backbone, etc., to accommodate this ever-increasing deployment of wireless across the enterprise, across both the inpatient and outpatient space.
Around the same time, a couple other things were also going on concurrently. Looking back, we had so many balls in the air at the same time. We were building a critical care tower of about 250 beds that would replace 250 beds in our older building. It had all private rooms and was designed as a state-of-the-art facility. So that was just getting underway at the time that I got there. They basically just had the hole in the ground. I drove toward a complete state-of-the-art wireless infrastructure for this new building. There’s always the advantage when you have a new construction project where you can start clean.
So in the new building, unlike our existing facilities, I wanted to go with a distributed antennae system — we selected InnerWireless — so that we would have guaranteed wireless coverage, wall to wall and floor to floor, and floor to ceiling, wall to wall, for all of the various wireless technologies; not just 802.11, but everything. So we put in a distributed antenna system, we put in a newer Ascom phone system, and we put in connectivity for telemetry. We have wireless connectivity between the nurse call system and the wireless phones, we have five bars of cell phone coverage for the major carriers, we’ve got pagers covered, and everything else. So in that new building, we really are very state-of-the-art, cutting-edge. And we still continued to roll out in the existing building.
The other thing that was occurring in parallel with that rollout was on the physician side, we decided to pursue a PDA-based mobile solution from PatientKeeper that provided both a charge capture capability and clinical results review from our Cerner EMR. That started out as a pilot in 2007, and grew to what is now an expansive rollout to 200-300 physicians. And it runs on smartphones — it ran on Treo and we’re pouring it over now to Blackberry. But obviously in order for that to be successful, we had to ensure pretty solid cellular coverage within our facilities as well. All of this is, besides getting all the infrastructure in place to accommodate this, the support and the tuning and the management of it is pretty daunting because we’re in a difficult environment for any kind of wireless.
KG: How so?
RP: We have a lot of older buildings with various types of construction, and some of them date back to the 1940s. We have every generation of building you can imagine. And the construction materials of course vary and have different attenuation effects on the wireless signal. In addition to that, because it’s a dense downturn area, our wireless signals, or what I call “bubbles,” are constantly bumping into wireless bubbles from the university that we share the campus with and from the state government buildings for the city of Richmond, which are right across the street. So we’ve had a really challenging couple of years of not just getting the infrastructure in to provide ubiquitous wireless coverage, but also insuring that it’s configured in such a way that it can play with these other systems in place.
So three years later, we have probably over 1 million square feet of wireless and that includes the 802.11. And we have extensive telemetry, we have smartphone use by the physicians, we have that application over the cellular network and we’ve had to have the carriers put in additional towers to give us coverage on that. So we’ve done quite a bit; we’ve spent quite a bit of money and did a lot of work over the last three years.
KG: That’s a lot to tackle when taking over a new position. I’m sure that couldn’t have been easy.
RP: I don’t know that I was really aware that we were going to get that heavily into it at the time I took the job. It was just one of situations where one thing led to another. Like okay, I see where we are with the EMR, the organization wants to kick-start that and drive that to completion and is willing to provide the support. And then I’m thinking, okay do I have the underlying infrastructure to support rolling this out and expecting the nurses and clinicians to make greater use of it. And the answer came back pretty quickly — no, I don’t have enough devices. I don’t have enough mobility, I don’t have enough coverage. So that was the incentive for me to aggressively push forward on extensive wireless deployment.
Now the new building was great because we were able to do things the right way with optimal technology like a DAS system from InnerWireless. In the older buildings, we didn’t have that option. The retrofitting of that would’ve been way too difficult. So we’ve had to figure out how to make the best possible compromise and solution we could with conventional point technology.
