As a CIO, having all hospitals within a health system on the same EMR is the ideal scenario. But as Todd Richardson knows, it isn’t always a reality, particularly when physicians are minority owners. In this interview, Richardson, who serves as CIO of the six-hospital Deaconess Health System, gives his honest take of how having physician owners can impact IT decision making, and why organizations need to accommodate the growing need among patients to access their data. He also talks about the flaws with statewide HIEs, the challenges of managing the different needs of device users, the IT rounding program his team implemented, and the importance of viewing technology as an enabler.
- The organization’s bring-your-own-device strategy
- “We don’t provide company-based phones or iPads”
- Balancing flexibility and scalability
- Picking your battles
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO DOWNLOAD THIS PODCAST AND SUBSCRIBE TO OUR FEED AT iTUNES
I’ve got 17 different iterations of devices I have to support and try to maintain and keep track of. How do you do that? I don’t see it. It doesn’t scale.
We threw everything we could possibly throw in there to have variety and let them self-select. We spend too much time, I think, trying to come up with the perfect answer and there is no perfect answer. People are on a bell curve. People have different ways of practicing.
We went to the ends of the earth in designing and laying this out directly with the anesthesiologists. We had three of them on board helping us build the screens and make this thing extremely usable and fast, because they need to be focused on patient care, not messing around with the computer.
I said, ‘This is totally unacceptable. This is going to kill somebody.’ If this thing locks up and you’re in the middle of the case and it takes three-and-a-half minutes to get back in, we can’t have that.
I think you take every case and you’ve just to deal with it in that context. To sit there and say, ‘This is our standard device’ or ‘this is our standard monitor and it’s going to go everywhere in the health system,’ that might be easy to do and easier to deploy and support, but it’s certainly not going to work from their perspective.
Richardson: We’ve got a ‘bring your own’ strategy from a mobile device perspective. We don’t hand out or provide company-based phones and we don’t provide company-based iPads. If you have a smart phone, and from a wage and hour perspective you can get your e-mail remotely, we will push your email out to you. You need to have policies. We have to push policies out to the devices to make sure that you have a password to protect that. We’ll make sure the laptops are encrypted. We provide access in for iPads and the slate devices through Citrix so that we don’t have any footprint left on those devices if you’re accessing the medical record. So we certainly take our jobs very seriously on the security side and make sure we have good processes in place.
The alternative is providing iPads, for example. Who gets one? So then you have to have a policy on who gets them and which kind they get. You have to standardize because are you really going to support all of them? And as soon as you roll out the iPad 1, what’s going to happen when iPad 2 comes out and everybody who has the iPad 1 wants to turn it in and get iPad 2? So now I’m spending twice the amount. Some people are going to want WiFi, some people want 3G, some people want it to synch up with their Verizon account, and some people want it on AT&T. So now I’ve got 17 different iterations of devices I have to support and try to maintain and keep track of. How do you do that? I don’t see it. It doesn’t scale, just like HIEs don’t scale.
And then they have one for personal use and now they’re carrying a personal iPad around because if it’s going to be my device then we’re going to have policies that say you can’t load your own stuff on it because that’s good practice. Now they have two devices. I mean, who wants that? They only want one, so why not let it be theirs, and figure out how to do it securely. So that’s our strategy.
Guerra: Yeah and I think that’s certainly an even greater wish than having the organization buy the devices to use their own, but some CIOs would even prefer if the organization bought the devices so they could be locked down and whatnot. So certainly it seems to be a ‘bring-your-own’ strategy, and then a matter of figuring out how we empower the physicians to use their own devices. Would you say that’s a lot of what you’re seeing?
Richardson: That’s what I’m seeing from my seat. I spoke at a mobile health conference in Boston with the folks up in Ottawa. Dale Potter at The Ottawa Hospital just rolled out 3,000 iPads. He bought them and rolled them out. He brought in 125 consultants and wrote a frontend for their EMR to be run on in the iPad. I mean, good for him, they must have been a lot more money in Canada to support their health systems. If I said I was going to buy 3,000 iPads and hire 125 developers to put in a frontend — first of all, I don’t have that much money. And I don’t want the headaches of the device.
And yeah, it would be easy to standardize and say, ‘This is the device we will have.’ But that flies contrary to where we’re going with technology in that not everybody wants the same device. Not everybody likes the iPad. They want the Samsung Galaxy. People want different things— it’s no different than with devices throughout the health system.
I’ve been a part of rolling out three electronic health records and there is no silver bullet. Everybody’s trying to figure out what we’re going to do for hardware. Are we going to put devices in every room? Are we going to put in centralized PCs? Are we going to put in computers on wheels and make them mobile? Are we going to deploy laptops? What about notebooks? What about iPads? Everybody’s looking for one silver bullet, and there isn’t one. These devices become more function-based; at a bedside terminal, you’re going to do barcode administration. Physicians and nurses may look up some information. If we think a doctor is going to stand there by the bedside and do all of their documentation, we’re nuts. Some will, but the majority is not going to want to sit there and dictate or do their notes at the bedside.
Guerra: No, not in front of the patient.
Richardson: Some want to be mobile. Some want a slate. We have an anesthesiologist who brings his own iPad. If he takes somebody directly up to an ICU out of the OR, he logs on with his iPad and he does all the PACU stuff on his iPad up in the ICU. Others wouldn’t think of owning an iPad. They’re going to do it at a standard desktop PC right in the PACU. Another one will bring up a COW. So we threw everything we could possibly throw in there to have variety and let them self-select. We spend too much time, I think, trying to come up with the perfect answer and there is no perfect answer. People are on a bell curve. People have different ways of practicing. And we’ve got to stratify the devices across for them. It’s just a microcosm of the slate tablet. Not everybody wants an iPad. And then you have those at the end of their career and with the iPad, the form factor is too small for them. They want some great big thing with big clown buttons on it so they can see it. It really varies.
Guerra: There is a big difference between viewing information and inputting information and there are different types of information. There are things that I can do and see on my iPhone and there are things that I can see but I can’t input the information. There are certain web pages that have graphs on them or charts that you need to look at and they’re not really suited for a small screen. So there are all different scenarios, and different form factors work in different cases. My question is, you talk about providing variety of options, which makes a lot of sense, but then don’t you run into these scalability questions — issues that you’ve brought up before?
Richardson: Scalability in what sense?
Guerra: Just in managing a program that’s got too many variables.
Richardson: I think there is. A good example is we ruled anesthesia out. Here’s a case where you have physicians who are kind of like a pilot — takeoff and landing is what kills you when you’re flying a plane. When you’re putting the patient to sleep and when you’re bringing him back, that’s where that patient focus needs to be.
We’re trying to bring a system in now that takes the place of what they used to do on paper. Speed was of the essence; the layout of the screens, the building of the macros to bring in the various cases — if it’s a general anesthesia case, the meds that you’re going to use versus an ortho case, a neuro case, or a heart case. Those meds are relatively standard across the specialties. You can create a cardiac macro so that there’s one big clown button and when you punch it, it brings in all your meds that you’re going to probably need for that up on your screen.
We went to the ends of the earth in designing and laying this out directly with the anesthesiologists. We had three of them on board helping us build the screens and make this thing extremely usable and fast, because they need to be focused on patient care, not messing around with the computer, finding the mouse, and figuring out where they have to go to multiple screens. We took the device itself — I worked with my technology folks at the desktop level — and we brought in a 21-inch wide screen so that they can really lay out, like you’re saying, what the information is that they need so that they can track trends and look at the vital signs as they’re tracking through the case.
When we first brought it up, by the time we loaded all the standard garbage that we load to get on our network, it was three-and-a-half minutes from turning it on to being ready. And I said, ‘This is totally unacceptable. This is going to kill somebody.’ If this thing locks up and you’re in the middle of the case and it takes three-and-a-half minutes to get back in, we can’t have that. So we started to streamline it. We took this thing down to the point where we had it MAC address filtered by each device with a generic login so that only that login would work on that device. That way, we could control it. So we didn’t have a generic that could be used anywhere in the health system.
We put solid-state drives in these things, and by the time we were done with it, my guy had it down to 25 seconds from turning the machine on to hyperspace. Hyperspace is the icon for Epic. So from turning it on to ready to login to Epic it was 25 seconds. Now coming from three minutes and 30 seconds, that was light years. We put those things in with these nice big 21-inch touch screens. They can take care of the patient. They can turn around to their monitor, hit these big clown buttons, and it was slick. We put the devices in a couple of months before we went live with the anesthesia module itself to let them get them used to the devices. I had one lady who just wanted to hug me. She said, ‘Oh my God, thank you so much. My eyes are getting old as I’m getting older.’ And she said, ‘This is fantastic.’
So I think you take every case and you’ve just to deal with it in that context. To sit there and say, ‘This is our standard device’ or ‘this is our standard monitor and it’s going to go everywhere in the health system,’ that might be easy to do and easier to deploy and support, but it’s certainly not going to work from their perspective.
When you deal with anesthesia on that module, you’ve got to deal with them. And there’s even a microcosm level within that group and within one module. If one guy wants to bring his iPad in, let him do it. As long as it doesn’t compromise the system, let it go. You fall back on security concerns; you don’t compromise on those things. But if it’s stuff that in the big picture doesn’t matter, then why focus on it? If it’s not going to sink the ship, let it go.
Guerra: I was afraid when you were telling that story you were going to say that you rolled it out and they thought 25 seconds was unreasonable — that they thought that was too long.
Richardson: Well that’s one strategy, make it a three-minute-and-30-second login all the time and then when you go to 25 seconds, it looks like you’ve just arrived.
Guerra: Yeah, and make sure you show up unshaven and say you’ve been working all night, right? Because it’s all relative.
Richardson: It’s all relative. And I guarantee you that a year from now that with the 25 seconds, I’m going to hear complaints. Hopefully, technology will continue to push and we can make it faster.
Richardson: We round in the ORs every morning. I have my help desk guys and desktop folks round just to make sure everything’s up and running, it’s functioning, and they can get to the Internet — that they’re up in running form. We’re talking about patient care, and when I say patient care, we use these devices in patient care, but it’s different seeing an eight-year-old kid in your office for otitis media and calling that patient care, than it is putting somebody to sleep, cutting them open, and doing surgery on vital organs. You have zero room for error there, folks. I need systems that are highly, highly reliable, available and the integrity cannot be compromised.
We can’t have doctors dinking around with these systems and trying to get them turned on. We’re paying them big money and they have lives at their hands. I need these systems and I can put systems in place to make sure everything’s working. It’s not going to be a 100% but I’ll tell you what, as I get ready to leave to go out for Christmas, they don’t complain. They believe that we’re doing everything we can. Now, when it hiccups, they’ll call me and it will rumble up through the force. There will be a disturbance and we’ll figure it out and we’ll fix it, but these guys believe now that we are absolutely there to support them in whatever they need.