Blog

Posts Tagged design

Know Your User: User-Centered Design in Healthcare

Designing software is both an art and a science. Software developers by nature think about every possibility and edge case, while designers focus on the most common user paths. Handling unexpected failures gracefully is a hard problem to solve.

I’ve been re-reading the classic book The Inmates Are Running the Asylum, about the natural tension between design and functionality in software. While it’s a bit hard on developers (they don’t all think in ones and zeros and many have fantastic design instincts), many of the problems the book outlines are still prevalent today, and you can also see the architectural problems that are inherited by lack of user-centered design in software that’s been around a while.

User-centered design puts the end-user front and center, and it’s not just about thinking of what ‘jobs’ that user needs to accomplish, but also the users purpose for those jobs and mental state while solving them.

It’s often easier to illustrate a point by what not to do, so let’s start with two recent examples, one from my experience and one from an opinion piece in the New York Times.

Our accountant at Wellpepper uses QuickBooks. QuickBooks is designed for accountants and business owners. It’s the grand-daddy (or mommy) of accounting programs, and has been around in one form or another since 1984, so they should know their users and what they are trying to accomplish pretty well by now. Guess what? They don’t.

Here’s the error message I see when QuickBooks isn’t able to pull up my data.

quickbooks error message

What have you done with my data, Quickbooks?

Isn’t that cute? No, no it’s not. The job I’m trying to do is look at bank balances, profit and loss, expenses, accounts payable, and accounts receivable. QuickBooks has the data that runs the financial side of the business. I’d like to think they know where that data is, and take both the delivery and protection of that data seriously. This isn’t a situation like the now retired Twitter Fail Whale when a tweet isn’t sent. Even Twitter realized that when someone is trying to accomplish a task a cute error message is more frustrating than appeasing. Add QuickBooks’ inappropriate error message to the fact that they can’t display my business data, and that I get this message at least once per day, I start to worry about whether I can trust them.

Now let’s look at healthcare, where trust and empathy are really important. We’re all familiar with the whimsical Verona campus of Epic Systems. People clamor to visit during Epic’s annual conference. It’s a source of pride for employees. Some of that whimsy comes across in cow-splotch login screens. Is that appropriate in a healthcare setting? Well, we have a dog as a logo, so I’m not going to judge. What is never appropriate is making your users feel bad, as this physician outlines in her New York times opinion piece.

But on a recent Monday morning when I logged into so-called Epic Hyperspace for the first time, I was greeted with a pop-up box and an urgent message: “You currently have deficiencies that are either delinquent or will become delinquent within one week. Please complete at your earliest convenience.” https://www.nytimes.com/2019/11/01/health/epic-electronic-health-records.html

First run user-experience is so important. It sets the tone for how people will feel about using your product. Doctors have trained for years to be able to care for patients. Why make them feel lacking, and especially not on the first time they meet you?

We know it’s possible for software to be context aware, and deliver appropriate levels of empathy, and also we know when to sound the alarm. At Wellpepper we take user-centered design very seriously, and work with patients and providers to make sure the messages we send, and the overall experience is appropriate for the people using the software. Our patented adaptive notification system is just one example of this. It changes the type of messages people receive based on how they interact with the application.

Since our application is used by people in both acute and chronic conditions, standard messages also provide encouragement but don’t imply that someone will “recover” or “get better.” In the event that a patient records a symptom that indicates a life threatening emergency, we clear the screen of any other tasks, and display a short and clear alert. (We also alert the care team, but we make sure the patient clearly understands what they need to do next.) These are all examples of being aware of not just what the person needs to do, but their potential frame of mind when doing it.

We think a lot about what to leave out of the software. Someone who is recovering from surgery, or newly diagnosed with a chronic disease has enough to worry about. Figuring out a demanding interface shouldn’t be part of that. We like to think about the minimum information we need to help a patient-self manage, and to help the provider track the patient. As a result, we have what we call a deceptively simple user interface, both for patients and providers.

If you’re interested in how Wellpepper patient engagement platform can help support your patients, get in touch!

Posted in: Healthcare motivation, Healthcare Technology, M-health

Leave a Comment (0) →

Simple Patient-Centered Design

At Wellpepper, we work hard to make sure our software is intuitive, including working with external academic researchers on randomized control trials for people who may have cognitive or other disabilities. This is both to make sure our software is easy-to-use for all abilities, and to overcome a frequent bias we hear about older people not being able to use applications, and also to provide valuable feedback. We’ve found from these studies, the results of which will be published shortly in peer-reviewed journals, that software can be designed for long-term adherence, and this adherence to programs can lead to clinically-meaningful patient outcome improvements.

User-centered design relies on three principles, all of which can be practiced easily, but require continual discipline to practice. It’s easy to assume you know how your users or patients will react either based on your own experiences, or based on prior knowledge. There’s really no substitute for direct experience though. When we practice user-centered design, we think about things from three aspects:

Immersion

Place ourselves in the full experience through the eyes of the user. This is possibly the most powerful way to impact user-centered design, but sometimes the most difficult. Virtual reality is proving to be a great way to experience immersion. At the Kaiser Permanente Center For Total Health in Washington, DC, participants experience a virtual reality tour by a homeless man showing where he sleeps and spends his days. It’s very powerful to be right there with him. While this is definitely a deep-dive immersion experience, there are other ways like these physical therapy students who learned what it was like to age through simple simulations like braces, and crutches. Changing the font size on your screens can be a really easy way to see whether your solution is useable by those with less than 20/20 vision. With many technology solutions being built by young teams, immersion can be a very powerful tool for usable and accessible software.

Observation

Carefully watch and examine what people are actually doing. It can be really difficult to do this without jumping in and explaining how to use your solution. An interesting way to get started with observation is to start before you start building a solution: go and visit your end-user’s environment and take notes, video, and pictures.

Understanding what is around them when they are using your solution may give you much greater insight. When possible we try to visit the clinic before a deployment of Wellpepper. Simple things like whether wifi is available, how busy the waiting room is, and who is initiating conversations with patients can help us understand how to better build administrative tools that fit into the clinician’s workflow. Once you’ve started with observing your users where they will use your solution, the next step is to have them test what you’ve built. Again, it doesn’t have to be complicated. Starting with asking them how they think they would use paper wireframes or voice interface testing with Wizard of Oz scenarios can get you early feedback before you become too attached to your creations.

Conversation

Accurately capture conversations and personal stories. The personal stories will give you insight into what’s important to your users, and also uncover things that you can’t possibly know just by looking at usage data. Conversations can help you with this. The great thing about conversations is that they are an easy way to share feedback with team members who can’t be there, and personal stories help your team converge around personas. We’ve found personal stories to be really helpful in thinking about software design, in particular understanding how to capture those personal stories from patients right in the software by letting them set and track progress against their own personal goals.

Doctor’s often talk about how becoming a patient or becoming a care-giver for a loved one changes their experiences of healthcare and makes them better doctors. This is truly user-centered design, but deeply personal experience is not the only way to learn.

To learn more:

Check out the work Bon Ku, MD is doing at Jefferson University Hospital teaching design to physicians.

Visit the Kaiser Permanente Innovation Center.

Learn about our research with Boston University and Harvard to show patient adherence and outcome improvements.

Read these books from physicians who became patients.
In Shock: My Journey from Death to Recovery and the Redemptive Power of Hope, Rana Adwish, MD
When Breath Becomes Air Paul Kalanithi, MD

Posted in: Adherence, Aging, Behavior Change, Clinical Research, Healthcare Technology, Healthcare transformation, patient engagement, Patient Satisfaction, Research

Leave a Comment (1) →
Google+