Patient Health Records with Dean Karavite

Posted on | Leave a comment

  • A Podcast for Everyone coverMedical records are moving online. That means they have to be accessible, but it’s also an opportunity to improve them – especially personal health records intended for patients to use to monitor their own heath. Dean Karavite explains how a project to design a personal health record for people with disabilities led to some innovative ideas that could make them more useful for everyone.

    Transcript available.   Download file (mp3).   Duration: 26:36 (17.3MB)

    Dean joins Whitney Quesenbery for this episode of A Podcast for Everyone to talk about the work he’s been doing in a project to create usable and accessible health records.

    • What are personal health records?
    • Why are they important for people with disabilities?
    • What have they learned about the accessibility of electronic health records?
    • What can you learn about any product by testing it with people with disabilities?
    • How does the design of the Cuisinart connect people with disabilities and the Mac?

    Photo of Dean KaraviteDean Karavite is a Human Computer Interaction Specialist for The Children’s Hospital of Philadelphia (CHOP). He has fifteen years experience in clinical informatics, was a usability specialist at IBM, and has experience developing assistive technology solutions for people with disabilities. He got his masters degree in HCI from the University of Michigan, School of Information. About the project collaborators.

    Resources mentioned in this podcast

    A Podcast for Everyone is brought to you by UIERosenfeld MediaThe Paciello Group, and O’Reilly.

    Subscribe on iTunes  –   Follow @awebforeveryone   –   Podcast RSS feed (xml)


    Whitney Quesenbery: Hi, I’m Whitney Quesenbery. I’m co-author with Sarah Horton of “A Web For Everyone.” Today, I’m talking to Dean Karavite from the Children’s Hospital of Philadelphia, where he’s a human interactions specialist in clinical informatics.

    That sounds like a mouthful, but it really just means that he works on making medical information usable for both patients and health professionals. One of his current projects is designing personal health records that are useful for people with disabilities. Hi, Dean. Thanks so much for joining us.

    Dean Karavite: Hello, Whitney. Thank you so much for inviting me to talk about our work.

    Whitney: Let’s start by talking about what you mean by a personal health record for people with disabilities. What kind of problems are you trying to solve?

    Dean: First, I think we can start with some definitions. They’re actually important for a number of reasons. If you were to go to the website, there’s a page that provides precise definitions for different systems used in health information technology.

    What’s interesting about these definitions is they’re largely based on system interoperability.

    What that means is the ability for these systems, or the lack of ability, for these systems to share information between different healthcare providers or organizations, as well as to patients.

    Whitney: This is all technical. This is “Can one computer talk to another?”

    Dean: Yes, in effect. If I could go through some definitions, and then I can explain to you why it’s important to people with disabilities. First, we have the electronic medical record, or the EMR.

    This is defined as the electronic version of a patient chart, but exists only within a single healthcare organization. For example, Children’s Hospital of Philadelphia or another hospital in town, or even your doctor’s office around the corner.

    Corresponding to that on the patient side is what’s often referred to as the patient portal. This allows patients or family members to log in, access some of the record, and perform some basic tasks, like prescription renewals, lab results, things like that.

    Again, like the EMR, they are confined or limited to a single healthcare organization. You might be receiving care from different providers, and each might have their own EMR, and therefore, you might have multiple patient portal accounts, none of which are connected.

    Whitney: That doesn’t sound very good.

    Dean: No, it’s a big problem, and there are many people working on this. Next, we have another definition. The EHR, or electronic health record. This refers to, again, a clinician system, but unlike the EMR, this has information about the patient from any and every provider they’ve ever seen.

    Whitney: Now we’re getting a little more patient-centered.

    Dean: Yes, but the problem with the EHR is that, at least in the United States, and at a practical level, it really doesn’t exist yet.

    Then, we have the PHR, or personal health record. This is defined as a system that is managed by patients, where they can collect, compile, organize all of their information from multiple healthcare providers in one place, but the problem there is it’s not easily connected to healthcare providers.

    Whitney: We have a mythical EHR, the thing that’s going to pull all of our data together, and then we have to make up for it, something patients can use on their own systems.

    Dean: Right.

    Whitney: We have a mythical EHR that will pull everything together, but doesn’t quite yet. Is it filling a gap to say we have a Personal Health Record, which the patient can manage to solve that gap?

    Dean: Yes, it allows them to manage their own information, but it doesn’t allow them to easily share that information with new or existing healthcare providers.

    Whitney: Another gap appears.

    Dean: Yes, but in our project we decided to use the term PHR somewhat generically, and imagine hopefully a near future state where interoperability is solved.

    Whitney: This sounds like a pretty big generic problem, and I’m sure you’re right. There are probably thousands of people in the US working on it, if not more, around the world. Tell me where this connects to accessibility.

    Dean: As part of our project exploring accessible Personal Health Records, one of the methods we have applied was performing a survey with 150 people with different disabilities. In that survey, we had our participants’ rate over 20 health topics in two ways.

    First, in terms of how important the particular topic was to their health and healthcare, and second, their current level of satisfaction with a particular issue or topic.

    The number one, most highly rated issue in terms of importance was the ability to share medical information between different providers’ offices, and hospitals.

    Whitney: This isn’t just technical…you’re just looking at a chart, and saying here’s a gap, let’s fill it. This is actually coming from people. Are you talking about people with various kinds of disabilities, here?

    Dean: It certainly applies to people with severe disabilities, but also anyone who’s a complex patient and receives care from multiple healthcare providers, because the real underlying issue here isn’t just the transfer of data, but care coordination, which is the collaboration in not just communication, but collaboration between multiple healthcare providers.

    Whitney: When we were talking before, you said something that I thought was really important, and I want to make sure we bring it out early, which is that disability doesn’t necessarily mean poor health.

    Dean: Right. The source of that idea and that concept, which I think is really a very valuable resource that your listeners might be interested in, is the Surgeon General’s Call to Action to improve the health and wellness of persons with disabilities.

    It’s really available online. I’m sure you’ll be providing a link to that on your website.

    Whitney: We sure will.

    Dean: In the report, which was published in 2005, it doesn’t really cover healthcare technology, but it does a wonderful job, I think, of describing the overall landscape, you could say, of people, disabilities, and health, and healthcare. The authors make a number of important points.

    First, and foremost, is that people with disabilities can live a meaningful and fulfilling life, but the key is health and wellness. However, they go on to present a number of studies and data demonstrating that people with disabilities encounter myriad disparities with almost every aspect of health and the healthcare system.

    Whitney: Give me some examples of things that might challenge them in managing their health.

    Dean: The report cites a number of measures, for example, higher risks, poor outcomes, access, really almost you can think of. What’s, I think, important about that is now we know people with disabilities are underserved by technology that’s not accessible, but they’re also underserved by the healthcare system.

    Whitney: Tell me about some of the kinds of health issues, in specific, someone with a disability might need to manage.

    Dean: This is a pretty large topic, and there’s different ways to think about this. As you stated before, we should not equate disability with poor health, and not think of disability as an illness.

    One thing that the Surgeon General’s report stresses is that we should treat not the disability, but the whole person. This was reflected with many of the participants in our study.

    For example, we met plenty of people who were disabled, but their disability wasn’t necessarily a health issue, meaning it didn’t require ongoing care or any type of treatment, nor did these people think of their disability as a health issue.

    At the next level, there are people who have a disability, and while their disability isn’t necessarily an ongoing health issue, there are secondary issues related to their disability that need to be managed.

    I have a few examples. For example, a person who’s blind, they might have to take some eye drops to maintain their tear ducts. Or, it could be a person who’s deaf and has a baby, and undergoes genetic testing to determine whether or not their baby has a particular variant for hearing loss.

    Then, there might be a person who has a spinal cord injury and is using a wheelchair, and with that, comes a number of secondary issues. For example, first, we need to prevent pressure sores or bed sores. There are special cushions that can do that, as well as checking and maintaining skin integrity.

    They also might have a urinary catheter. Along with that, we have to prevent infection, and also measure daily fluid intake and output.

    At the next level, we met plenty of people where their disability was the result of an illness that required ongoing treatment, for example, multiple sclerosis or spina bifida.

    Then, you had other people who had these types of issues along with other more common chronic illnesses, for example, diabetes, heart disease, high cholesterol, etc.

    What we found is that we’ve learned a lot from, of course, all of our participants, but from these more complex patients, they often provided us with the most detailed, sophisticated, and even innovative ideas on what an accessible PHR should do.

    Whitney: It’s interesting. Shawn Henry, who wrote the book “Just Ask” about doing usability testing with people with disabilities, has been talking about, questioning whether, we ought to be thinking about people with disabilities as extreme users.

    That is people who might be particularly stressed by websites that are written to code or by information that isn’t very clear or by information that doesn’t appeal to all their senses, so that if they’re relying one sense more than other senses, they’re more likely to find a problem.

    That’s a pretty interesting idea. I think what you’re saying is very much the same.

    That people with disabilities use the healthcare system a lot and in many different ways, both as the people with many different chronic health problems to people for whom their health isn’t really an issue at all to people who need some monitoring of things like what you said about monitoring infections and fluid intake and outtake.

    That suggests that you have a population that represents, in a very small group of people, a wide range of what a good personal health record needs to do.

    Dean: Absolutely, I actually had a chance to talk to Shawn on the telephone about a month ago about our project, and she was fantastic. I agree with this. I think that many people who promote this idea of the extreme user. I have two sources where I’ve learned about this method or this concept, you might say.

    First, I met a gentleman from IDEO a number of years ago. He told me that this was a foundation of their approach to design in many ways. Maybe not specifically to people with disabilities, but any extreme user who represents the tail end of the normal curve. They’re more interesting, and we can learn more from them.

    Another great example of this method in practice was the industrial designer, Marc Harrison. He was both a teacher and designer at the Rhode Island School of Design. Way back in the mid-70s, he applied this approach specifically by thinking about people with disabilities in the design of the Cuisinart food processor.

    Whitney: Way before the Good Grips, huh?

    Dean: Yes, some people say that Harrison’s work and teaching actually influenced people who went on to design all these wonderful kitchen devices.

    He actually influenced technology, too. I have an interesting little story here that I love to tell.

    If you read the Steve Job’s biography by Walter Isaacson, there are two references to the Cuisinart, where Steve Jobs told the original designers of the first Mac to go out and buy Cuisinart. “It’s great. You have to see this thing.”

    If you put the original Mac in 1984 side-by-side with an early ’80s Cuisinart, the influence on the physical design of the Mac is immediately obvious. Not only is the Mac designed with software for accessibility and more universal design, but its physical design had this perhaps unknown influence as well.

    Whitney: This goes along with the idea I’m hearing a lot in design circles which is that, the broader your influence is, get out of your little box and look for inspiration all over the place.

    In fact, IDEO’s OpenIDEO project really espouses that, because they start with inspirations as their first step where people are encouraged to go out and get as broad and crazy, as many ideas as they can, and bring them into the group.

    Can you give me a story about how one of your extreme users helped you see something interesting that you wrapped into the project? I think we’ve been talking a little theoretically, and I’d love to bring it down and ground it in some of your research.

    Dean: I have many interesting stories, but there’s one in particular I’d like to share with you. One of our participants was a woman who, a number of years ago, was in a horrible accident. She was a pedestrian hit by a drunk driver.

    This left her paralyzed from the neck down, paraplegic. Not only did she survive her injury, but she insisted on staying at home, though confined to a bed or wheelchair, with home health aides, in order to raise, I think, her three or four children to adulthood.

    Once her children were fully grown and out of the house, she had more time to pursue her own interests. One thing she did, for example, she’s always interested in art. So, she had learned to paint both realistic portraits and landscapes by essentially holding the paintbrush in her teeth. That’s pretty amazing.

    She also had never used a computer, but wanted to learn. With help from one of our project partners, Inglis Foundation here in Philadelphia, and their assistive technology experts, she had learned to use a computer, but the only way she could interact with a computer was using speech recognition, Dragon Systems.

    She had to learn both a PC and navigating a PC and all the different applications while simultaneously learning how to use Dragon systems. She told me a story that brought together both healthcare and technology.

    A little while back, she had seen her physician who, I believe, specialized in people with spinal cord injuries. They did some lab work, and there was some particular level, I’m sorry, I don’t remember what the level was, that was low.

    The doctor prescribed her some supplements to bring this level up. I believe she went home and she thought about this, and she wasn’t happy about this prescription for at least two reasons. First, like a lot of people with disabilities, she has a very fixed, limited income.

    This represented an additional cost that she really couldn’t afford. Second, she just didn’t like the idea of taking a pill. She went online and did some research and identified a number of foods that could help increase this blood level.

    On her own, she changed her diet and then created a spreadsheet where she tracked, not only her diet, but these different blood levels over time. When she finally reached the desired level, she went to her doctor and said, we’ve met the level.

    Here’s what I did and here’s my data. The doctor was so impressed that she not only agreed with this treatment course. She now prescribes the same dietary changes to other patients who have the same issue.

    Whitney: I think what I love about that story is not the personal overcoming of difficulties, but how she used essentially a very small PHR, she created her own little PHR, to manage this. She was able to use data to solve her own health issues, but also to suggest things that could be used for other people.

    Dean: What this really demonstrates, what I think is really interesting about it, is it demonstrates some of the work we’re doing at Children’s Hospital in research on shared decision making, using patient portals and different systems to support decision making and collaboration between patients and providers.

    One of our research assistants’ focus, his name is Dr. Alex Fiks. What I’ve learned from Alex is that decision making is really founded on, first, defining and communicating a person’s preferences and goals.

    Then you come up with a plan to meet those goals, and then you develop some measure that can be used to track the progress of meeting that goal over time.

    This woman’s story perfectly represents a shared decision making model, and everything’s there except for a PHR that can support the collection and communication of this information.

    Whitney: We’ve been talking about this project for a long time, and we haven’t given you a chance to talk about who’s involved and how you brought together the team. I understand it’s three organizations, the Children’s Hospital of Philadelphia, the Inglis Foundation, and NCAM at WGBH.

    Why don’t you just tell me a little bit about who’s involved and how the collaboration comes together.

    Dean: The project is led by WGBH and the National Center for Accessible Media. They received the grant from the Department of Education National Institute on Disability, Rehabilitation, and Research, NIDRR.

    Both Inglis and I and my boss, Dr. Grundmeier, are collaborators on the project with WGBH. The grant is referred to as a field-initiated project, so unlike a more traditional research project, its real goal is to demonstrate some process towards improving the life or every aspect of people with severe disabilities.

    Whitney: It’s like what we would call a pilot.

    Dean: Exactly. Now we couldn’t build an actual PHR, but the project was designed to demonstrate the application of both usability and accessibility methods, including subjects of people with disabilities, and essentially going through a whole series of usability methods towards the final product of interactive prototypes that we user-tested.

    Whitney: At the end of this, NIDRR, which is a federal agency, will end up with not just a bunch of concepts, but some good, user-centered design supporting the results and hopefully take it further. Is that the goal?

    Dean: Right, and all of our methods, results, and even the prototypes are freely available to EMR, EHR vendors, developers, anyone in the community.

    Whitney: That’s really great. I know that you have a website. I think it’s

    Dean: WGBH is managing that, yes.

    Whitney: We will, of course, as always, have links and all the other names and links that we’ve been mentioning on the website. I know that a couple of your collaborators asked if they could be on this podcast, but we think it’s an easier conversation when we can’t see each other to do it one-on-one.

    I’d like to just give them a nod and hope that you have represented them well.


    Dean: I hope so.

    Whitney: What else have you learned? Has there been anything that’s surprising? Actually, my favorite ideas are ideas that seem obvious once you’ve seen them happen in the research. Anything that’s come out that really jumped out for you?

    Dean: Yes. Beyond all the data and the results and the analysis, there’s something that I actually learned a long time ago but continued to relearn, and I think it’s very important. You referred to this many times.

    In technology, UX, many people are striving to create accessible technology, and then other people are striving to create innovative technology. Unfortunately, there are still people, though not in the UX community, who feel that accessibility can be a barrier or somehow inhibit innovation.

    I believe the exact opposite is the case. People with disabilities can be the source of many innovative ideas. Going back to the mid-90s when I started working with people with disabilities and assistive technology, I remember a few things.

    First, I was forced to learn about and apply very rigorous usability methods. There was really no way around that, so it’s a great environment for that. At the same time, I was extremely selfish and thinking, repeatedly, “Boy, this stuff is so computer, I wish I had this on my computer.”

    For example, touch screens, on-screen keyboards with word prediction, zoomable displays, speech recognition, text-to-speech. Think about it. It took about 10 to 15 years, and now we all have it on our computers, our phones and other devices, and we absolutely love it.

    Whitney: That’s a really, really great example. Sometimes, it takes awhile to figure out. We start with a new technology.

    For a while, there’s the technology looking for a problem. Often, extreme cases are the problem they solve. Then, it slowly migrates into general technology.

    Dean: I had the same thoughts in this project. By taking all these different participants and their experiences and their stories and their ideas, and creating these PHR prototypes, I found myself thinking repeatedly, “Boy, this is a pretty cool PHR. I wish I had one, not only for myself, but my family, friends, and, of course, all the physicians and patients we work with.”

    Whitney: I want to touch on one thing that relates to methods. If people are interested in what you did, there are great write-ups of all of your surveys and your prototyping and your usability testing.

    I don’t want to take time to go through them here because they’re much more comprehensive on the Web than we could cover. But one of the things that a lot of people find as a challenge is finding people with disabilities who they can work with, and setting up usability tests to be able to include people in their research.

    Can you talk a little bit about how that worked for you?

    Dean: We had a distinct advantage in that our project partner, Inglis Foundation, manages the care of hundreds of people with disabilities. Not only were they a collaborative in the project, they were a source of many of our participants.

    However, they specialize mainly in people with severe physical disabilities. We also found another organization, here in Philadelphia, Liberty Resources, who helps people with all types of disabilities live independently. They provided us with other participants, too. They were a great help.

    There are certainly a lot of people out there with disabilities and all they seem very appreciative and enjoy being part of the project. Another resource might be your local university.

    Most have some type of assistive technology and lab and/or faculty and students who are interested in accessibility. That might be a good resource, too.

    Whitney: It sounds like one of the pieces of advice I heard was to not wait until you are ready to do your first usability test to get to know people. Don’t put out a call and announce that you’re ready for the doors to open, but actually do a little bit of outreach in advance.

    Start making those connections early. Find out who’s in your community, what organization are there, what schools are there, so that you’re building those relationships before you ask people for their help.

    Dean: Absolutely.

    Whitney: I want to end with one other thing, which I was reading the website, and it said one of the things you did was an audit of three personal health record websites. Of those three, one only met half of the basic accessibility guidelines.

    That sounds really terrible, but the other two only met two of the 12 guidelines, but then it goes on to talk about an inverse relationship between accessible and functional. Can you tell me about that?

    Dean: Yes, it was really an interesting experiment. We reviewed these three different systems for both usability, accessibility, and also functionality.

    I really felt like I was in the “Goldilocks” scenario. Though, of course, it wasn’t one that was ‘just right.’

    For example, the first system, they seemed to have made an honest effort at meeting accessibility guidelines. Yet, the system functionality was really very limited, and usability was also a problem. There was very complex navigation, very confusing.

    The other system that did the worst in terms of accessibility actually had the most functionality, and a lot of very sophisticated functionality matching some of the ideas from our participants.

    The third system was somewhere in between. It was a PHR type system. It was pretty good in accessibility except for one particular issue. It relied on patient entered data, and with every piece of patient entered data, you had to enter a date. Well, the date widget was neither accessible for screen readers or keyboard only. I thought it was interesting representing this one little problem that could be such a huge barrier.

    But, overall, I think what we learned from that exercise is that if you were to put these three systems together, you’d have a pretty great PHR. I think what’s fundamental here and interesting is that the technology isn’t the problem. There are no constraints.

    There’s no issue with making an accessible PHR. It can be done. The real issue here with these systems really comes down to good requirements gathering, and then the application of a solid set of both usability and accessibility methods.

    Whitney: Was the focus of your work really about understanding what a PHR needed to do because you already knew how to make things accessible and usable?

    Dean: Yeah, we actually talked about that quite a bit. WGBH, people like yourself, many are promoting all types of accessibility methods. There’s plenty of that out there.

    We didn’t feel like we could add too much to the excellent works that already available. Instead, we wanted to go in a different tack, and think about, or demonstrate in effect, “What if we designed a PHR specifically for people with disabilities? What would that be like? Would it be usable? Would it be accessible? Would it be functional?”

    I think we answered that. I hope we answered that. That no only would be accessible and usable, but also highly functional and more innovative than existing systems.

    Precisely because it’s based on people who face a lot of challenges, both with their health, but also access, mobility, finance, everything, and as a result, have many more interesting and detailed ideas of how we can better engage with the healthcare system.

    Whitney: I think that’s an awesome place to wrap up at actually. Dean, this has been fascinating.

    I think we could talk for hours, but I’ll say again that there’s lots more information and some great research stories up on the Health IT Access site.

    Tell me a little bit about some of the people who really helped make this project sing.

    Dean: At WGBH, we had a number of people. Larry Goldberg who was the PI of the project, who’s the Director of NCAM. Madeleine Rothberg who’s Project Manager. Geoff Freed who’s their Accessibility Expert.

    At Inglis Foundation, their CEO Gavin Kerr, who is fantastic. Lea Frontino who’s their VP of Health IT and Innovation. Don Waller, who’s the Head of their Assistive Technology Lab and did a lot of work in not only connecting us with participants, but building test stations that we could use for the participants we tested at Inglis.

    At CHOP, myself and my boss, Dr. Grundmeier who was one of our most amazing researchers, physicians, developers, analysts, statisticians.

    Whitney: That’s a really big team. That’s kind of cool. I’m used to hearing more about projects that are a couple of people, in a skunkworks project, but this sounds much more well rounded.

    Dean: Yes. And…I forgot one person. I’m sorry, Jason Withrow who’s a developer. He’s a contractor, and he also has a degree in HCI and experience with accessibility.

    Whitney: We’ll give a shout out on the website, and I’ll get some papers from you that might be interesting for people who want to dig a little more deeply in the research. To wrap up, if you’re interested in learning more, the project website is, again,

    To find the podcast page on, look for the page for “A Web for Everyone.” That’s our book and follow that link to A Podcast for Everyone.

    Thanks so much to you for listening in, and a special thanks for our sponsors, UIE, Rosenfield Media, The Pacielo Group, and O’Reilly for making this series happen. Be sure to follow us at @awebforeveryone on Twitter for information about future podcasts.

    About the Project Collaborators

    CBMi and The Children’s Hospital of Philadelphia. The Children’s Hospital of Philadelphia was founded in 1855 as the nation’s first pediatric hospital. The Center for Biomedical Informatics (CBMi) at The Children’s Hospital of Philadelphia is the home for the development of innovative solutions to healthcare’s immediate and long-term informatics needs. CBMi provides informatics-focused services, applications, and educational programs to Children’s Hospital clinicians and researchers.

    The Inglis Foundation is a community-based foundation that is committed to using technology and healthcare information to empower people with severe physical disabilities to live life to the fullest. Inglis maintains a skilled-nursing care facility and independent-living facilities as well as care management, adult day and employment services for people with disabilities living in their own homes.

    NCAM and WGBH (Project Leader). WGBH Boston is America’s public broadcasting powerhouse—PBS’ largest producer of TV and Web content. The Carl and Ruth Shapiro Family National Center for Accessible Media at WGBH is a research and development organization that works to make existing and emerging technologies accessible to all audiences. NCAM is part of the Media Access Group at WGBH, which has been providing captioning and video description services for people with disabilities since 1972.