Around March 2019 I had surgery to fix a respiratory issue in my chest. The surgery went very well. During recovery I had to keep track of a lot of medications. To track my medications, the hospital gave me a spreadsheet. Here is an example sheet:
To make it clear when I should take a medication, I would mark the pending medications with a circle. Then I would fill in the circle with a check once the medication had been taken. After taking each round of medications I would set the timer on my phone to remind me to take the next one.
During my recovery I spent a lot of time sitting around and thinking. Like any good technologist, I started to think about what an app for keeping track of medications would look like.
A couple of things jumped out at me:
So I started working on a prototype.
As I was working on the prototype there was a nagging question in the back of my head. "Is this actually better than paper?". My mind kept circling back to a quote from the UI design essay Magic Ink.
"I suggest that the design of information software should be approached initially and primarily as a graphic design project. The foremost concern should be appearance—what and how information is presented. The designer should ask: What is relevant information? What questions will the viewer ask? What situations will she want to compare? What decision is she trying to make? How can the data be presented most effectively? How can the visual vocabulary and techniques of graphic design be employed to direct the user’s eyes to the solution? The designer must start by considering what the software looks like, because the user is using it to learn, and she learns by looking at it.
Instead of dismissing ink-and-paper design as a relic of a previous century, the software designer should consider it a baseline. If information software can’t present its data at least as well as a piece of paper, how have we progressed?"
To see what issues an app would really solve, I drew up some pros and cons of paper.
When I dug into the cons, none of them seemed that bad.
"If you want to update what medications you are taking on the go you have to carry the paper with you." - Carrying a piece of paper is not that hard. Paper can easily be folded and put in the pocket. Also, I suspect that on average people who are on high amounts of medications spend less time outside the house.
"You could lose the paper." - This one would be annoying but not catastrophic. Since the majority of medicine is set on a schedule you could just try to remember if you took the medication on schedule and if there were any exceptions on the lost day.
"It will not automatically alert you when it is time to take medication." - It would be kind of nice to have an alarm set automatically. But, it is not that hard to set an alarm on your phone if you have a smartphone. Also manually setting the timer makes it easy to do things like: "I would prefer to sleep through this scheduled medication because I am low on sleep and I can just take it later without issue."
"It is not easy to aggregate information over a long period of time." - Since the schedule of most of the medication was fairly regular I do not think I would have gained much by aggregation. Every time I needed historical information I only cared about 1 or 2 days in the past. In those cases, paper beats digital because you can lay out the 3 days next to each other and look at the patterns. No need to condense the information into a small screen.
"Adding new medications requires printing a new sheet or writing the medications in." - Slightly annoying but if the medication leaves space for write-ins it is not bad. Additionally, if you have the template for the medicine chart it is simple to modify and print out a new sheeti
"In order to share your medication progress with your doctor you need to take a picture of the chart and show it to them or verbally tell them." - I only had to share my medical progress once in during a checkup 1 week after operation. The data that the doctor wanted to know was very high level. For example: "Which of these as needed medications are you still taking?". The only thing they really cared about was what I had done yesterday. But, I could see automatic sharing of progress with doctors to be useful for more complex cases.
"Every time you look at the sheet you need to seperately check the time to see when to take the next medications." - Checking the time is not hard. There are usually time keeping devices scattered around the house (stove, microwave, wall clocks etc.) and I almost always had my phone on me. But, It would be cool to see the current time slide across the chart. That way you would always know when it is time to take a medication. The time could move across the screen like in Mario Paint Composer.
When I examined the pros and cons, I realized an app wouldn't add much value. Maybe an app for easily creating medicine logs to print would be useful. But, a digital medicine log didn't seem like it was worth it. After deciding in this case paper was better than the screen. I started thinking about what it would look like if I could bring computation to the paper instead of the paper to computation.
In Oakland, CA there is a building that is a computer. This building is home to the Dynamicland research project. Right now, the building is outfitted with motion tracking projectors distributed across the ceiling. In the building there are special pieces of paper with code on them. The projectors can recognize these sheets of paper and run their code. The result of the code is then projected on the paper. I reccomend clicking the link above and checking it out. It is very cool!
I started imagining what my medicine chart could look like if my house was outfitted with Dynamicland like technology.
One of the simplest improvements to the chart I could think of was adding a line to track the current time. In this example, the projector would recognize the paper and project the current time as a red line. Now you don't have to consult your phone while looking at the sheet. The line can march across the chart as time advances.
It would be useful to highlight the times when a medication is ok to take. This is especially true for "as needed" medications. These are not taken on a schedule but they have minimum intervals they can be taken (ex: every 3hrs). So, before taking an "as needed" medication you need to look at when you previously took it to determine if it is too soon. If you highlight when the medication is "available" this is not an issue.
For certain medications like narcotics it could be useful/informative to see how many you have taken over time. Below is an example mock that charts the amount taken per day. Imagine the chart is projected on a blank piece of paper.
Other useful things to do in a Dynamicland like system:
It is interesting to see how augmenting the paper with computation allows for many improvements. What is cool about these improvements is they are all additive. They don't destroy what makes the paper useful. The paper can still act like paper.
It will probably be a long time before I have a house outfitted with Dynamicland like technology. In the mean time, I will keep using my paper medicine log.
Omar Rizwan's essay on designing a map application in Dynamicland. It is probably the best overview of how Dynamicland works.
May-Li Khoe's post on what makes the digital medium different. While designing, I kept on thinking if any of attributes of dynamic media listed were useful to my medicine chart.