How to Make Data Meaningful for Users of Connected Health Devices and Apps

Key Insights for IoT Health Apps

Brett Webb, Vice President

One of the most rewarding projects I’ve worked on in my 10 years with Favorite Medium was an app for a sensor that continuously measures blood glucose levels. You stick the sensor to your arm, scan it with your phone and get your glucose reading. No finger pricks; no dropping blood onto a strip.

That was around seven years ago, before the explosion of the IoT health devices market. Hundreds of thousands of diabetes patients have since used the app to manage their sugar intake, know when their glucose levels spike, and remember to give themselves insulin. Today, we’re developing a similar app for people who simply want to experience the benefits of continuous health monitoring for things like athletic performance.

Helping to save lives and empowering people to improve their health has been the best part of these two projects. But tackling the technical, creative, and UX challenges has also been rewarding. Our biggest question: How do you make medical data meaningful to specific groups of users?

Here are the answers we’ve found.

1. Frame data within a narrative

If the only thing users wanted to know was their blood glucose level at that very moment, we could have built a single-page app that showed that information alone: a single mg/dL number, updated every five minutes or whenever the user opened the app.

But health metrics need context. As there’s no cure for diabetes yet, patients and their doctors need to track how the disease affects their daily lives and progresses over time. For athletes and everyday users who want to understand their bodies better, presenting their health data within a narrative is also beneficial.

To address this issue, we built out a couple of features into the glucose monitoring app and its subsequent updates.

• Activity notes: These allow users to log activities like eating or exercising at specific times, so they can connect each activity to periods where their blood sugar went high or low. For example, running for 15 minutes or walking for an hour would affect a user’s blood sugar levels differently than eating an apple or a candy bar.
• Historical data: These reports show users’ glucose levels over time — say, the last 24 hours, seven days, or three weeks. We also incorporated a trend arrow into this view, so a user can look at it and identify when their glucose level is heading up or down and act accordingly.

2. Consider how the IoT app fits into the user’s life

In the app for people with diabetes, just giving glucose measurements is enough. For athletics and general health monitoring, though, the context is different.

Hence, it is important to understand the specific needs of your app’s target audience — medical or otherwise — and how it fits into their daily reality. Subsequently, making data meaningful for them becomes a matter of identifying simple, relevant, and intuitive approaches to displaying their health information.

For instance, the science on ketones and lactate and bodily performance is not as established as in diabetes care management. But users want to know this information, so catering to their needs requires extensive research and development (R&D).

Initially, we patterned the second version of the glucose monitoring app a lot like the one designed for people with diabetes. But we soon found that the consumers with no direct health implications were less inclined to fill out activity notes. This problem was further exacerbated by the fact that there are already a lot of existing apps that track user health, from exercise and hormones to diet.

On the user’s end, not logging their daily activities in the app means that they don’t get enough context on why their health metrics look the way they do. But when building IoT health devices and apps, you also need to consider how much precision and context users actually need.

By logging activities like eating and exercising at specific times, users can establish connections between their blood sugar levels and these activities.

For example, in the medical use case, there is little to no room for error because the data presented will be used to make healthcare decisions. In the case of everyday fitness users, though, some ambiguity is still acceptable as the context is more about making lifestyle changes and less of a life or death situation.

3. Consider the emotions triggered by the data

People with diabetes using IoT health devices and apps may need to act on the data presented instantly. Meanwhile, the average everyday user likely sees the tools more as nice-to-haves than life-saving technologies. Hence, the approach to building for each customer profile should differ since the data they view on the app can trigger varying emotions.

So how do you alert the user to changes in their blood sugar levels without causing undue distress? Incorporating sound design, removing data specificity, and reviewing data indicators within the app are some solutions we explored.

• Incorporating sound design: A lot can be conveyed via the feedback sounds in a product, like indicating whether a user’s glucose levels are spiking or crashing. We worked with a sound designer to create and test different sounds that users eventually began to associate with specific types of data so they could act accordingly. When we began building the new apps for everyday users, we also mimicked sounds from the legacy hardware.
• Removing data specificity: To avoid alarming users unnecessarily, we worked on approximating health metrics within the reporting dashboard. That way, minor spikes or drops in users’ sugar levels weren’t easily noticeable unless they looked into the data granularly — say per hour, day, or week. Instead, users saw relatively even mg/dL numbers unless the upward or downward change was significant.
• Reviewing data indicators: Within the app’s health monitoring dashboard, we changed the trend arrow from red to a softer, neutral color so it looked less worrisome when users’ checked their records.

Building IoT health apps is challenging but highly rewarding

So far, the technology we helped build is enabling hundreds of thousands of people with diabetes to better manage their blood sugar levels and improve their overall health and wellbeing. It has also had a massive (multi-billion USD) impact on our client’s business.

As we help our client cater to a new set of users (athletes and everyday users looking to improve their health and fitness), the user experience question we asked seven years ago is the same one we’re considering now: How can we continue to make data meaningful for users?

We relish challenging technological projects at Favorite Medium. Pairing that work with a product that has such a massive impact on people’s lives and health makes this project an honor to have worked on. Add to this the financial impact of this technology on our client’s business, and this is the type of work we’ll tell our grandchildren about.

If you have any questions or would like to learn more about us, feel free to get in touch with Favorite Medium!