Ron Crall talks about how, with a very limited budget, he successfully migrated the IT infrastructure while CIO at Quincy Medical Center in Boston. He took the hospital off legacy systems to deliver wireless thin-client solutions supporting the upgraded Meditech operating system.
SPEAKER'S BIOGRAPHY: Crall is the former CIO at St. Joseph Healthcare in Bangor, Maine, and Quincy Medical Center in Boston. He has extensive experience in both financial and technology management. His technical skills and accomplishments include design and deployment of ISO 9001-certified redundant network operation centers and their supporting co-location facilities, including their high-bandwidth internetworking. He holds a bachelor of arts degree in business administration from Armstrong University in Berkeley, CA.
Download for later:
- Internet Explorer: Right Click > Save Target As
- Firefox: Right Click > Save Link As
Read the full transcript from this podcast below:
CIO Ron Crall: Life support for ailing IT infrastructure
Kate Evans-Correia: Good morning. My name is Kate Evans-Correia, and I'm the Executive Editor of SearchCIO-Midmarket. I'm about to chat with Ron Crall, the former CIO of Quincy Medical Center in Quincy, MA, and the winner of our 2008 Midmarket IT Leadership Award about his innovative work there updating legacy systems without tapping out the hospital's capital budget.
First welcome, and congratulations on this award.
Ron Crall: Thank you, Kate.
Kate Evans-Correia: With regulations putting near strangleholds on healthcare facilities, you had quite a job ahead of you. Tell me about the project and explain the business goal.
Ron Crall: Kate, when I walked into the door at Quincy, we had really three really significant issues that were staring me in the face. We had a network that had failed about ninety days before I got there. We had physicians' practices that were being poorly served through legacy connectivity to the hospital. And we had over 200 obsolete dumb terminals that had a tremendous maintenance expense that was associated with them, and the maintenance contractor had just notified us that he was going to no longer support those terminals, that he no longer could get parts for them.
So with those issues at hand we immediately did some re-engineering of the network and managed to restore it to a very functional status and eliminated the engineering mistakes that had been made in the past.
With that out of the way, we concentrated on providing a better solution for our physician providers and started looking at the connectivity we had, which at the time I walked in was a wall full of 4800 baud modems, multiplexers and devices that, in the Meditech world, were called terminal servers, but that were just really just serial device agents.
All of that equipment was old. Much of it was failing. And it was very expensive. The lines themselves were $700 per year per provider. So we knew that we at least had some operating expenses that, if we could eliminate, we could provide at least $700 worth of new equipment with a completely neutral expense for the year.
And by finding and testing some thin client devices, we were able to use the public internet over a SSLVPN tunnel, and provide the physicians with modern equipment that exceeded the performance of the old equipment by a very wide margin and was extremely reliable. And I think we increased the physician satisfaction tremendously.
Kate Evans-Correia: Ron, excuse me, why thin clients?
Ron Crall: We had serial devices previously, and if we had the other deployment strategy supported by the vendor that was for personal computers, and deploying personal computers and printers and supporting the personal computers and printers would have cost us significantly more than the $700 that we had available to have an annual, cost-neutral solution.
And I'd had previous exposure to thin client deployment that had been extremely successful. We also had the 200 dumb terminals in-house that we had to replace, and we were hoping that we could have a single solution that would work for both of our off-site providers as well as solve our in-house deployment problems. So the thin client looked like it was going to address all of those issues with a consistent, standard solution that could be deployed internally or externally.
Kate Evans-Correia: What type of applications are you running?
Ron Crall: The hospital is Meditech shop. That's the core clinical application. So access to patient clinical data in that system would have been one of the applications we were trying to support. The other application would be access into a radiology system for radiology reports. So the most lab, radiology and some of the patient demographic and insurance information is always required by the physicians' practices, so getting them access into those systems were the primary needs.
Kate Evans-Correia: How long did the project take from when you walked in there to the final deployment?
Ron Crall: The network was stabilized, re-engineered and redeployed within the first ninety days. During that period of time we were also looking at the thin client solution. It took us about 45 days to do a proof of concept to make sure that the applications that we needed to deploy would work on the thin clients. It took us another month to order, configure and deploy the hardware. So we were doing first installs within 100 days.
Kate Evans-Correia: So tell me how things have been working now. I mean performance, cost savings... What's it like now?
Ron Crall: We did the first provider installations, and number one was reliability. The old equipment would frequently fail. If it failed at either end, due to a power failure or any interruption in service, usually both ends had to be contacted and equipment at both ends reset to re-establish connectivity, which required a coordinated effort of multiple people. It took a lot of time.
The new equipment is extremely robust, and we have had no endpoint communication failures that have required support services. So that alone made the physicians very happy. Even if it had been as slow, at least it was reliable. However, we went from a 4800 baud serial modem connection and the associated time that it took for the connection to be established, to instantly available broadband connectivity. So the performance we did not actually quantify in transactions per second improvement, but it was a significant order of magnitude of performance improvement that was immediately obvious.
Costs were neutral for the first year. So every physician that we updated would cost us nothing over the course of twelve months, and at that point there would be about a $700 per practice savings in following years. So it had a good return on investment, aside from the productivity gains by not having to have staff continually supporting failed connectivity.
Kate Evans-Correia: Tell me about the rest of your computer infrastructure there at the hospital. I know that Quincy Medical Center was one of the first to use electronic medical records, and that it was named one of the "100 Most Wired Hospitals". I'm sort of curious as to what that means, but explain to me the infrastructure. What you have set up in terms of security, particularly in terms of accessing patient records, who uses those thin clients? Is it staff? Nursing? Doctors? Just sort of give us a little overview.
Ron Crall: In 1999, Quincy was one of the "100 Most Wired Hospitals". However, they also changed the structure. It had been a publicly-owned facility up until that time. In 1999 it became a not-for-profit organization that has struggled since 1999 to have a good financial rate of return on their operations. The infrastructure really hasn't had much capital investment done since that point in time, so at this point virtually everything qualifies as legacy equipment.
The wireless security, when I walked through the door, as I said there were a variety of engineering issues with the facility, and they were very early adopters of emergency department electronic medical record, and also doing nursing notes on the floors. And most of that was supported with wireless.
However, the wireless deployment was not well-engineered and was left unsecured. So we had some significant issues with regard to patient confidentiality and security of the network. So we immediately secured the wireless access points and were working to redeploy so that we had, in the areas that we did have wireless, that we had coverage that was substantial enough, to ensure good connectivity. And then take some of the other areas that really didn't have to be wireless off of wireless and reuse the existing equipment to provide robust areas where it was needed. And going back to hardwired again, providing robust areas where the portability was not required. So it was really looking at the assets that we had available to us and redeployed them so that they fit the business needs of the organization.
Kate Evans-Correia: In deploying your thin clients, it's obviously the outcome that you had hoped for. But in building it out, what were some of the obstacles that you encountered?
Ron Crall: We had computers on wheels, COWS as they're often referred to, rolling carts that were -- We were early adopters so this was aging equipment. The laptops that were on those COWS were old and failing. Again, looking to have the broadest standardization possible and not having to reinvest in very expensive rolling carts, new ones, we investigated putting a thin client on the cart that supported 802.11 and retro-fitting the cart with a UPS. We found some thin clients that only drew about three-and-a-half Watts of power. That's about a half of a nightlight in power consumption. And then mated that with a thin-panel display. We found that we could have exactly the same physical device on the cart, a very expensive thin client instead of a relatively expensive tablet or a laptop, and service times that exceeded the original equipment when it was new. So again, this wasn't perhaps the most elegant solution that could be found, but it gave everybody exactly the same desktop, it gave them the portability that was needed, and it gave them good service times between recharges. So I think we found at least a great interim solution to help the facility find some robust solutions and improved reliability with a very constrained budget.
Kate Evans-Correia: Well, Ron, we are almost out of time, so I want to wrap it up, but I just do want to ask you one last question. What is the next hot technology you envision, particularly as it relates to hospitals? What will hospitals and healthcare facilities be using next?
Ron Crall: Having had a number of years experience with virtualized desktops and virtualized server environments, hospitals have, as they become more clinically centric, and are not just supporting the business office, which is 9:00 to 5:00, five days a week, but supporting the emergency departments, the patient carriers, which are 24/7, demanding environments, I think that the ability and the robustness of virtual servers supporting thin clients, where fault tolerance can be engineered in a cost-effective manner, where a hot spare is an automatic fail-over, again can be engineered in a more cost-effective manner. I think that that's going to be in the healthcare arena. Something that many institutions will be looking long and hard at. And I think we're going to start seeing significant deployment in that area.
Kate Evans-Correia: All right. Great, Ron. Very interesting stuff that you're talking about doing in IT. I appreciate your time. And again, congratulations on winning the Leadership award. You've done some impressive stuff here. I appreciate you being with us today.
Ron Crall: Kate, thank you for your time. Good day.