This is the second of a series of blog pieces I will do to sum up my experience as a judge for the Ensuring Safe Transitions from Hospital to Home Challenge. Developers were asked to come up with a mobile app that would empower patients and reduce preventable hospital readmissions. Each of my blog pieces will look at one or two design issues that were not explicitly covered in the evaluation materials, but which struck me as important.
Education is not enough
Quite a few of the apps included some form of patient education in their design. I think patient education is a good thing, so I was pleased to see that many included the feature. But some of the teams botched the job badly. I doubt that some of the developers ever tried to consume any of the education content on their own devices. The majority of the education features were text-based, using presentation methods that were state of the art in the late 1990s. Some seemed to be trying to replicate a handout on a sheet of paper. For that, I need a tablet computer? Others included various web links that took me to health education available on the Internet. There is a place for that, but seriously folks, is Wikipedia the best health education source you can think of?
In contrast to those close-but-no-cigar solutions, the best of the apps made good use of well-produced videos to explain health conditions on demand. Some had video tutorials to train the patient in self-care tasks, or show how to take complex medications. The state of the art in on-demand education is short video segments that are easy to absorb, targeted at the right level of health literacy, and well-segmented so you can take them in small doses as needed. Some of the apps showed features that allowed the user to adapt their education content to their preferred literacy level. Several of the apps indicated that they could accomodate insertion of video content provided by a healthcare system or other third party. But wouldn't it be nice if there were a central source for high-quality, peer-reviewed education videos that can be streamed to devices for applications of this type? What we really need is a sort of Netflix for health education videos. That looks like a business opportunity for somebody out there.
Some of the applications were so gung-ho on patient education they seemed to think that just giving more and better information would somehow reduce hospital readmissions. Good luck. Simply giving more information is not going to reduce readmissions. If that were true, all you would need to do is send the patient home with more discharge paperwork.
The problem is that information does not equal understanding. And even understanding something does not guarantee compliance with a care regimen. The education focus needs to shift from quantity to quality, with an emphasis on immediate relevance and positive reinforcement of behavioral change. What is delivered needs to be adjusted based on how the patient or User understands the situation they are in. Since some of the education is targeted at changing behavior patterns, maybe the app designers should look at some of the models available for consumer health support such as weight loss or quitting smoking for design ideas. For example, a friend of mine is currently quitting smoking. Good for him! He has known for years that smoking is bad for him, so his decision to finally quit was not just a matter of information. Now that he is acting on what he knows, he also needs moral support and encouragement to help him continue with a difficult behavioral change and physical addiction to nicotine.The apps must be good at delivering "Just in Time" education when it is really needed and when there is a teachable moment. For that purpose, less is often more. To push the envelope just a bit, visualize Siri running on top of a large database of paragraph-length education segments, many of them attached to videos that are perhaps one or two minutes in length. Tell the device "Oops, I forgot to take my morning pills. What should I do?" Does the system know what your morning pills were, and what you should do if you forgot them? Here we cross over into the realm of intelligent systems that can determine what you need to know when you need to know it.
Only two or three out of the batch of 29 apps showed any awareness that not all patients are fluent in English. I live in California, and out here on the left coast 42% of the population sometimes speaks a language other than English when at home. [You can look up the stats for your own state via the Census Bureau.] That's twice the national average, which is 20%. Only one of the apps gave any details on their design that mentioned if they were capable of internationalization (sometimes called "language skinning"). Internationalization is a specific type of coding approach that enables a user to specify a preferred language as part of their settings. Since those for whom English is a second language are often among the medically-underserved, their exclusion from the education features was a disappointment. Some vendors, on the other hand, specialized in multilingual health content, but did not have as much focus on improving care transitions specifically. Maybe some of these teams should consider collaborating with one another to combine their respective strengths.
As I dicussed in my previous post in this series, the User of the app may actually be a concerned family member or caregiver rather than the patient. But none of the apps included demos of video education that was designed to meet the needs of a caregiver, such as how to take someone else's blood pressure or design meal plans that avoid salt. None of the apps mentioned training on how to do Advance Care Planning. And, of course, none of the demos ever suggested that the healthcare team might benefit from training in care transitions issues. Some of the apps had dashboards intended for use by discharge planners and care managers. Why not include some training videos on care transitions designed specifically for them? For a great series of short videos explaining the ins and outs of care transitions for healthcare professionals see the series provided by the Center for Elder Care and Advanced Illness.
I wish I could get all of the teams to go back to their whiteboards and create education features built around the following four use cases. Each case pushes the scope of the education problem in a different way. In each case, ask yourself who is the user, what does the user need to know, and how can I best deliver that information to the user? These cases are not based on real people. Education designers, start your engines!
- Your patient is Mae, a Chinese-speaking 79-year old woman who is currently in the hospital with her second admission in the past year for complications related to congestive heart failure (CHF) and chronic obstructive pulmonary disorder (COPD) exacerbated by continued smoking. She is expected to be stabilized and released within a day or so. Mae lives with her daughter, Julie, who is computer-literate and has a tablet computer, an iPhone, a home computer, and a home telephone (which her mother uses often). Julie's mother can speak English but is more fluent in Cantonese, which is what she speaks at home. Julie's mother generally receives care from her private physician, who accepts Medicare, and one hospital where her physician is affiliated. There is a Chinese-speaking case manager in the hospital who is familiar with the family situation. Once Mae is released she will need to check her weight each morning to see if she is retaining water. She is taking several medications related to her condition. Julie must travel often for her job, and when she is on the road she arranges for a Cantonese-speaking volunteer from a local community center to visit her mother once a day.
- Your patient is Ed, a 37-year old heroin addict who is HIV+. Ed does not have a regular physician but has been in and out of the ER four times in the past year for relatively minor health issues. He knows that he is supposed to take antiviral medications regularly but sometimes skips them. He gets his antiviral meds from a community health center where he is well-known and considered a "reliable regular". He has a cell phone but does not have a home computer. He is on Medicaid and SSI. He lives with his girlfriend, who is also addicted to heroin.
- Your patient is Susan, a 53-year old woman who is seriously overweight. She has been gaining weight steadily for the past several years, and her lack of success with dieting has contributed to her current depression. Recently she was diagnosed with diabetes after having been admitted to the hospital after a minor heart attack from which she recovered well. She lives alone and has a very limited social circle. She has a home computer and an Android phone. She takes insulin as well as oral diabetes medications. She has private health insurance through her employer and receives healthcare through a managed care system. She often complains that she never sees the same physician twice in a row.
- Your care provider is Ted, a 37-year old nurse who has recently been assigned as a discharge planner and discharge followup specialist in a rural hospital. Ted is new to discharge planning but wants to make a positive impact on the lives of his patients. He has been given a tablet computer for use with other hospital data collection applications and can access your care transitions application on the same device via a web connection. Some of the patients he is discharging may need special help to ensure they can stay at home. How can we help Ted understand the different types of home care needs that patients with different conditions will require? How can we help him facilitate support arrangements with community-based agencies in his town that could help his patients? Ted is addicted to his Android phone when not at work.
Next: Part 3: For mobile apps, size does matter
Previous: Part 1: The ONC Care Transitions Tech Challenge: Thoughts from a Judge