April, 2010 – Healthcare Podcast

Jeff Horvath: Hello everybody, this is Jeff Horvath, Vice President of Human Factors International. I am joined today by Susan Weinschenk, our Chief of UX Strategy for the Americas and by Tom Suther, Human Factors Engineer at the Mayo Clinic. HFI is planning a series of podcast on healthcare UX over the next several weeks or months where we will be talking with a series of outside experts in the field of healthcare UX and Tom is our first guest today, Tom from the Mayo Clinic. So we will be talking today about electronic health records and the usability of them. Tom will be talking to us about a series or about a project that he's been working on as well as the field in general. So we will, well, Tom, why don't you say hi to us all and Susan say hello.

Tom Suther: Hey Jeff, thank you. Hi Susan. Good to hear from you and thank you for this opportunity.

Susan Weinschenk: Good to have you, Tom.

Jeff Horvath: Great. So Tom what can you tell us? Can you give us a little background about EMRs and EHRs at Mayo Clinic on the space there and what you guys are working on?

Tom Suther: It's interesting that you said EMRs (laughter), because we do have, we have a huge complex system that we need to display to physicians and nurses and other care givers. And this complex system happens to be our body. And somehow through tests and procedures and observations and assessments, all this complex data needs to be viewed by the care givers to make some assessments and then go on with may be care plans and how to help the patient. I've seen displays of nuclear power plants and they basically show their all their information up on one huge wall. We don't have that luxury here because we've got patients all over the place. So we try to do that through the computer and it seems to me that the progression of the electronic medical record where as we had a good system, a fantastic system here at Mayo developed by Dr. Plummer of the paper record and at that time each patient had one record and only one record so that we could get all of the information in there. But of course, now what we are viewing is well, that record isn't quite exactly what I need, I need something else. So we created a database over here for the transplant center or the wound care needs something special. And with all these despaired data bases now, it's hard for a physician to get a good view of the data or a nurse, because they're looking in so many multiple places. I have to see what the orders are. I want to see what that care plan is. I want to see okay now why did we do that care plan? I better go over here to the labs tab. Oh may be I'd better look in the clinical notes. Oh what was I looking for? (Laughter) I mean it is almost that serious with our care givers and the reason it is serious for us is because of they added here at Mayo to our main slogan or main logo is the needs of the patient come first. So how in the world, you know, we have this all this data that doesn't come to us in an integrated view. In addition, it sometimes doesn't even help us in our work flows. You know data, the Japanese had a great term. Just in time. Just in time what? What was the manufacturing term at that time I've spread that now to training and education and data. When do you need that data that's when you should serve it up.

Jeff Horvath: Tom, you said it started long ago and there was a wonderful single system of paper based system was that?

Tom Suther: Yes.

Jeff Horvath: Yeah, so it sounds like that was working pretty well. You know, I'm sure there are lots of good reasons for moving to digital. But how would you say what caused the things to kind of diversify and end up with all those different systems instead of keeping that initial, you know, probably very well thought out paper system?

Tom Suther: Um hm, well, also I wanted to point out that the paper system was also, we have a fairly large campus here. But the paper system was also augmented by a trolley system, because we knew we had to get the record to the patient, but then the patient had to sit there for 2 hours now. You always got excuses, but you never knew really why, because the record was coming down. The reason why is because we purchased a fairly good at that time it was a fairly good EMR. But the EMR just wasn't quite what we needed. It wasn't complete. It did an awful lot of data recording, and that's what EMRs are doing now, good job of getting the data in there and letting you record the data, right, but then what data is it recording, what data will it allow you to record? And then will you give me all that transplant information? So the transplant folks said, well, I am sorry but this is just not satisfying us as much as we need to be satisfied. So now we need to create our own system. And the orthopedics, not the orthopedics but ophthalmology is a great example. The system that we purchased was not able to allow us to have a picture of the eye in there. So you could draw a little squigglies to show what was wrong with the eye or to represent the eye in different ways. So we had to come up with an application that would allow us to do that. So it might have been with the bringing up of our EMR, because it didn't fit all those work flows that we talked about before.

Jeff Horvath: Got you. So you mentioned that the wonderful complex system that we are trying to work with here, the human body, and all of that information and you've got different systems now with the EMR and all these augmented systems attached to it with all that information, what are you guys doing to wrestle with that issue? How are you trying to get that information back into a usable form or matching the work flow of the people that need it?

Tom Suther: Well, there was one move a little while ago in which we created a program called this was our first at templates thing, a program called Synthesis and by the name it kind of tells you what it was. Basically we took pertinent data from several of these major applications and now we display it in this program called Synthesis. So if I go to the labs tab, I can see all the labs as they come from all these different systems. Or I can see the vitals as they come from all these different systems. But that's just the first step, because it's still not integrating the data. When you get a – we have a little system called the inbox. This is a great example. The lab sends over an inbox message and says your lab for a patient XYZ is done. Okay. It's done, good. Now I got to get the patient XYZ don't I (laughter)? Then I got to find that lab. So why not just show the lab value right there in the email so that you have it? And that's sort of the idea of getting the better augmentations to the work flow. But as I am saying that Synthesis is a great start, but we still have to develop that further. Because now we got all the data there, now we have to identify these work flows and try to help those work flows out.

Jeff Horvath: Got you. So you have a data, you have a complicated work flows, lots of different users with lots of different demands, and their time, and attention. It's a complicated problem to solve, so what are you guys...

Tom Suther: Excuse me, one of the things that we're doing, I wanted to mention Synthesis, but also the other thing that we're doing is working within this system, our current EMR, the nurses have an awful hard time with as a matter of fact, our nurses come up to 30 minutes before their work schedule to get prepared for the work shift.

Jeff Horvath: Oh boy.

Tom Suther: And that's on their own time. They create these little, we're not supposed to call them that, but they're little to do lists with all the pertinent information so they are integrating this stuff on paper, and...

Jeff Horvath: Because the system's not meeting their needs. It's too complicated or it's not matching their flow.

Tom Suther: It's not matching their flow.

Jeff Horvath: Yeah.

Tom Suther: And not matching the needs that they need to go care for the patient. Exactly, you hit it on the nail.

Susan Weinschenk: We used to say many years ago and Tom and Jeff you've probably have experienced this too that if you want to know what changes need to be made in a computer system that will really be usable for humans just go see what work arounds they created on their own.

Jeff Horvath: Absolutely.

Tom Suther: Yeah.

Jeff Horvath: We kind of talk to users go look for those sticky notes on the side of the monitor.

Tom Suther: Exactly.

Jeff Horvath: Yeah. And we got a lot of sticky notes.

Susan Weinschenk: It does sound, Tom, though that one of the things that you have gone for you there at Mayo is that is this culture that was already established about, you know, very customer centric, patient centric culture. It would seem that that's a really good environment for making your suggestions and ideas knows as a human factors expert.

Tom Suther: It really is in not only that but remember that we carry three shields: clinical which is the practice, research, and education. And with just having those three shields on everything we look at it seems to help because you know what's the worst thing about doing a usability study. I got to get representatives, users, don't I? Well, at Mayo Clinic I am blessed with 20,000 people (laughter). I can find actual users of the software we're going to develop. And everybody says what, isn't it hard to get people there? No. People are more than willing to come in and help us out with these usability problems. They see the problem also. I mean it's a kind of an I don't know it's kind of a old adage when I was working at a big major computer company one time we always had to take a satisfaction data right. And I never had any idea, I don't know why I'm doing this. And now I know why I'm doing, taking satisfaction data. Because if our users were are not satisfied with what they are doing, they're going to have a little frustration. And I can tell that frustration might just show itself to the patients. Now the patients' are saying you're confused with what you're doing and you are caring for me. I don't like that. So there's an awful lot to what our profession does to try to help in the big picture that we're not even realizing.

Jeff Horvath: Sure, I mean like you said you're motto there if I, I don't know if I remember it verbatum, but you know the patient first. You're there to serve the needs of the patient. And if nurses are spending time wrestling with the system or the physicians are spending you know extra time typing or searching or what not, that's time that they are not spending with the patients and you know that's counter to your vision and your goals.

Tom Suther: It is very counter and I don't like to say this but it is kind of counter to the bottom line too.

Jeff Horvath: Sure.

Tom Suther: If you take a look at it, if a nurse has to prepare for 30 minutes, well, I guess they are doing okay, because they're not charging Mayo for it (laughter). But we just put out a pilot of an Apple application for the EMR and I had doctors, yes it is hard to take it and estimate a time. But I said well, how much time do you think you are saving? It's 10 to 15 minutes. One doctor said 30 minutes a day. While if you find out what a doctor makes per hour, you'll say oh that's a lot of money saved. But also this doctor can go off and see other patients and help other people just as well. Now one of the problems that we've also had and talks to a specific project that I'm on also is that nurses were kind of forgotten in EMRs. At least the EMR that we had it was very physician centric, which is very important. I don't take anything away from that. But the nurses were all struggling in there. And in one of our little projects said well, how can we help the nurses with their plan of care? And the regulatory industries are saying an awful lot about what you need to do for the patient. And one of them is a plan of care stuff. And we started out in trying to find out what is a plan of care, we were finding the terms, and then what do you need to do a good plan of care? And well, we went out and did a year of study of nursing and what they needed and then came out with a little prototype for them.

Jeff Horvath: Great. Well, Tom, you got a human factors group there in the group that you're in. How are you guys involved in this product redesign effort, a product design effort. What role do you guys play in the process here of understanding these issues and trying to address them?

Tom Suther: Well, this, I'd like to may be use this nursing plan of care as an example. We got around and tried to do a competitive analysis and found that there aren't any other EMRs doing this either. At least reported by the nursing leadership and nursing list serves and direct questions to nursing. So we set out and we didn't know it was going to take a year, but we set out and just poured usability techniques in to and it's interesting because some of these techniques were already there. They had already been asking nurses this, and asking nurses that. As you all know, it's how you ask those questions that are very important as well.

Jeff Horvath: Right.

Tom Suther: We had a group of nurses who are always there on Monday afternoons for two and a half hours in which people just go to them and say what do you think about this, what do you think about that. That turned out to be somewhat of a fairly good sounding board, but in group focus groups like that, you basically get a lot of yeses. Because people just don't interact that much.

Jeff Horvath: Sure, yeah. Focus group is very different from an in depth interview or an data gathering session where you are talking to one person and really digging and understanding the context and all that, sure.

Tom Suther: Absolutely. And so we didn't rely on that particular one, but we did take our ideas to them to get some idea of what's going on. But we used structured interviews knowing exactly what we're going to ask them. We used informal interviews. Come on Ann, we've been here doing here doing this for an awful lot of long time, what do you think? What is the major thing that you want? And she told us the biggies. And she described what the biggie was. And we said wow, we can do something with that. And so those informal interviews are really good. We did an awful lot of observation of nursing. And yeah, we understood they had to go all over the place to get their information. So we looked at their work flows and tried to find out what types of information they needed. We did an affinity sort of tell me what all the groupings are of these data content, data pieces that we have. One of the main things that helped us an awful lot was that we had a team of 20 to 30 staff nurses and CNSs. And they really helped us out because they were our main drivers. We could go to them with questions and talk to them and really get down deep into these things. But even the CNSs said you shouldn't listen to me, this is for the staff nurse, which is something that I would tell them who is the audience, the primary audience being the staff nurse, the secondary audience being the CNSs. CNSs are super nurses, I guess you could say, they've gotten some special registration and information. So we got all this, but we still weren't able to get the idea across to them. You know, it's that communication that you need to have between the two. We were dancing around the solution. We were literally dancing around until we talked to Ann, and then I said well, I can mark that up somehow. So I created a series of markups. And these markups is my team mates like to say a two way diagnostic. We had a little bit of Apple in there, we had a little bit of Windows in there, (laughter) we had a little bit of my coloration in which I am no designer to speak of, and it didn't look like anything that we would want to present or want to give them as a final product. But it sure did give us ideas and more and more requirements. So we didn't start with the prototype because we didn't know what it should be or the markup. We started with visioning exercises and interviews, watching them do the work, and those types of things. And that's how it finally came out. And then we said, okay now we get a fix, we got to decide on what kind of technology. So we decided on the technology and we decided to do it, try to do it in our current EMR. And we had some really bright really bright programmers who knew the software and could get there and give us a first pass at the iteration that, quite frankly, I'll sit on my laurels here, I was overwhelmingly accepted by the usability participants. We ran 10 or 12 I think through this, which was quite a few. But we still kept coming back with little tit bits. But the idea was that these nurses said, well, it's not the idea of what. Now it's the idea of when are you going to give this, I need it now (laughter).

Jeff Horvath: Can I have it right now, please?

Tom Suther: And you know what, it's really great is that I have done usability studies for 30 years now, and this usability study was different, because we had done all that homework, we knew what they wanted, we knew what they needed, and we served it up to them, and they said oh wow. Thank you (laughter). And so it was the most happy time for everybody instead of sad time we got all these problems, we got to go and try to fix them. It was more of a happy time.

Jeff Horvath: Sounds like a pretty good user center design process, doesn't it Susan?

Susan Weinschenk: Yeah, it does. It's a, I just think it's really exciting and inspiring to hear about everything going on at Mayo. I am guessing that some of the people there are listening in are probably a little jealous about (laughter) all the support that you're gathering. But I know you, Tom, I've known you for a while and I know that you have worked in other environments that were not as supportive and you probably paid your dues in that regards.

Tom Suther: Oh I did, but you know, it's interesting because when I came here to Mayo 9 or 10 years ago, I came as a usability tester. But I knew that it was a foot on the door. And once I got here, I said my gosh! There's this work for 30 life times (laughter). But once again, you cannot always and to say there's still push back in some arenas right, until you start showing that these processes work that these tools that we have work. We introduced affinity sorting 9 years ago and now everybody does affinity sorting. I can't imagine. It's great because I'll come in and say we're going to do an affinity sort and we will do all the work. I said great (laughter). Now they are catching on. They like these tools. They love them. And when we have such good results as we are having with this plan of care work, other people take notice and they say I want to be in on that too. So it wasn't a rosy picture at the very first when we came here, but it's working out to be really good, because the process that usability engineering, it is just a winner. That's all it is, just a winner. And some people say it's good to talk at least talk to one customer if you can't talk to any, at least talk to one. We have a little bit better support in talking to multiples.

Jeff Horvath: That's awesome. Tom, if you could give some advice to some other folks in similar organizations about how to make user experience more strategic and how to make it a core part of the culture at their organization? Do you have any words of wisdom about that especially in the healthcare field?

Tom Suther: Wow, it's tough because it's really tough to get to the leadership and get them to understand it because they won't work in these tools. They only see the results. And then you got to have people work as an act of faith, you know, this guy knows what he is doing, I think, (laughter) you know. So it's I think that the old adage gets some short projects, big projects under your belt so that you can then share it with others and then they understand the value of it also. The one thing that I tell people is that I would much rather have happy users. And the only way I can get happy users is to know what they want and need and what their requirements are and only way I can do that is do these usability processes. Well, I imagine there are other processes too, but one way, a good way to do it is all these usability tools that we have at our hands. And you know, it's interesting because some of these tools anybody can use, but they still need training in how to use them. What is a good interview? Well, we got to help these folks in learning what a good interview is. So I guess the major wisdom here is to use these tools even if you have to use only a little one. You'll show people and you'll teach people that yes, I would rather my user do the work instead of me having to do that work. So and I am still struggling in our little group trying to get full acceptance of our program. We're only two people also. So that makes it a little harder, but we do have groups that are like consultants, systems and procedures, and I've talked to them and I've worked with them, and I've shown them and they become evangelists like I guess of some of these techniques. Just like affinity sorting all of a sudden it just took off like crazy. So I think it goes to one of the other adages here at Mayo, teach everybody. Just teach them. And you know give them, teach them how to fish, basically, and give them the tools, and then they'll be asking you to come back and say how do we put these tools together and when do we use these tools. And that's where your value will come in also.

Jeff Horvath: That's great. Yeah, I mean, at HFI we try to help organizations institutionalize user experience all the time, and you know some of the things you talked about there, Tom, are key to having the support from the senior folks, having the right people with the right skills. If they have a small team there, which sounds like a good team and you're making some progress so.

Tom Suther: Yes.

Jeff Horvath: Finding those show case projects, showing the success, really making it part of the culture in the process these are all key things.

Tom Suther: Also one thing that helped us is we had some consultants come in from the outside to teach a lot of us and to understand what some of our problems were. And those are consultants with their suitcases (laughter) and their brief cases you know, it's usually you have to have a brief case come in and tell you what you need to do. And it does, it worked, because these folks came in. The biggest part about that, the reason why I think it was worth the money is that our senior management saw this and said, my gosh in the way that they were talked to, they learned some of these concepts. And they learned yes, it's not good to just throw stuff out there and see what happens. So it sometimes it's needed to have someone from the outside come in and do a study on your EMR and say you've got problems here (laughter) and here are a couple of people who can help you out with all those problems. As a result of that, we've become very integral to a lot of decision making that goes on here at Mayo. And also one of the things that we are, we are pretty much run by a committee. So it's very important for Human Factors people to get in in those leadership areas and those committee areas also which we've done.

Jeff Horvath: Absolutely. Excellent, well, it sounds like you guys are off to a great start. And beyond the start you're heading down the path in a great way. Susan, any other thoughts here as we begin to ramp up?

Susan Weinschenk: I was just thinking as Tom was talking about how some of the things he is talking about remind me of the basic principles of persuasion and the things we know about people, for instance using the – doing projects and then talking about the projects. It's a form really of social validation or social proof. The idea is that other people in the organization start to realize that other people inside the organization are doing these things. And then it becomes more accepted. And then the idea of bringing in an outside expert which of course would be the whole principle of you know authority and experts and the idea there being that I always tell people here's what's going to happen, don't feel bad. You've been saying this stuff for years inside the organization. And now someone from the outside is just going to come in and say the same thing and everyone will ooh and aah and (laughter) you know you want to say wait a minute, I've been telling you this for 10 years, you know. (Laughter) but it's just the way it is. And use these things you know as part of the user experience tool kit to just use any tool that you can find.

Jeff Horvath: Great.

Tom Suther: I have to say that happened with us. (Laughter) we had, you guys, we had a big presentation at the very end and all of a sudden everybody was getting ah ha movements (laughter).

Jeff Horvath: We love ah ha movements.

Tom Suther: Yeah, we love ah ha movements, but hey we can do that for you also (laughter). So you're absolutely right. Keep that mouth shut, Tom, let him talk (laughter).

Jeff Horvath: As long as they are agreeing, right?

Tom Suther: Yep.

Jeff Horvath: Great. Well, Tom, do you have any other thoughts for us before we ramp up here today?

Tom Suther: No, the only other thought that I have is that I'm hoping that the EMR developers, there are several really great vendors out there including our vendor. They'd given us a great data entry tool. They'd given us a great way to look at all that data. Now they need to start thinking about how can we integrate this data. How can we put assessments and orders together so that the nurse can say oh you got a skin problem, here's all the assessment's, here's all the orders related to that skin problem. Hopefully, the EMRs of the future will help the nurses and physicians with those kinds of questions and it will help the patients.

Jeff Horvath: Great Tom. Well, we certainly appreciate your time today and everything you've been able to share with us. It's great to know how you've been really able to involve the users and get them involved in the design process to make a better product for everybody there. We wish you guys the best of luck going forward. This concludes our first podcast in the healthcare UX series. For those of you listening, there are a few things you can do from here. You can go to humanfactors.com and look for a white paper that we've put together that will be a companion to this series talking about some of the themes in healthcare UX. Or you can go to connect.humanfactors.com and join the conversation. Tom will be there with us and we'll continue the conversation about electronic health records and healthcare user experience in general. I'd love to hear from you and get your questions and your thoughts as well. Thanks for joining us today and we look forward to talking to you all again.