The Problems and Promise of HealthKit

After not one, but two false starts, HealthKit has arrived. Announced back at WWDC in June, HealthKit (the API) and Health (the user facing app) were tentpole features of iOS 8. The incredibly ambitious plan is to have Apple offer a single location for all your health related data, and to offer an easy way to share this data with other apps, services, and even with your doctor. Fitness tech has exploded in the last few years. I have seen an ever increasing number of people in the real world wearing Fitbits, Jawbones, Misfits, and other fitness devices. Fitness apps like RunKeeper, Nike+, and MapMyFitness on smartphones have been a staple of the races I have run for years.

But the connection between these apps, and the uses of its data, have so far been confusing and limited.

Let me give you an example. Take three platforms I use, Fitbit, RunKeeper, and MyFitnessPal. Each offers an API allowing other services to read and write data. Fitbit would write to RunKeeper, but not read. MyFitnessPal would read and write to both. RunKeeper accepted data from both Fitbit and MyFitnessPal. So if I logged food in MyFitnessPal, it was synced to both RunKeeper and Fitbit, but then Fitbit would turn around and sync the same information to RunKeeper, resulting in double data. Try the same thing with Jawbone and the opposite problem existed, it would read from RunKeeper, but not write to it. UGH!!!

What ended up happening was a convoluted series of connections, cross connections, duplication, de-duplication, and endless attempts to find the one chain of events that made everything sync just right, which usually didn’t exist. Remember that article I wrote about syncing Jawbone sleep data into RunKeeper? It remains the most read article I’ve ever posted to the site. I think that really speaks volumes to the level of difficulty people have using this technology.

Then there is the bigger question. Why? Why collect this data, and what purpose does it serve? I have found that really paying attention to my fitness goals has helped me lose weight and maintain the loss. But there is so much more this data can do.

Health app screenshot
The main screen of Health. You decide which data sets appear here.

The promise of HealthKit was to solve these problems and answer these questions. This is no small task. But Apple would seem to be the one player in the game who could actually do this. They have the reach, the talent, and the capital. This in addition to the soon to be launched Apple Watch, a device that puts fitness tracking front and center. HealthKit may end up being the perfect solution.

But it is not there yet, and it looks to have quite a ways to go.

Right now I see three major problems with HealthKit: stability, usability, and buy-in from the major players.

Stability

Let’s ignore the fact for a moment that on launch day for iOS 8, just hours before most people got it into their hands, every single app compatible with HealthKit had to be pulled from sale due to an undisclosed bug found very late in development. Let’s also ignore that the update put out to fix this problem ended up causing far bigger issues for all iPhones 6 owners who had the misfortune to apply it. So far attempting to actually use apps that have been updated to support health has been a buggy experience at best. For example, MyFitnessPal was one of the first to support HealthKit. Unfortunately it’s first version had a habit of deleting data from Health just as quickly as it was entered. This seems to of been fixed and the subsequent update but it still leaves a bad taste in my mouth, no pun intended.

SleepTrack, which is an app that adds sleep information into the RunKeeper HealthGraph, was also updated to be able to both write to and read from Health. Yet so far it’s been completely incapable of actually reading data that is posted there from other apps.

These kind of issues can really be the undoing of this system. Health data is extremely personal and people are unlikely to trust it to something that isn’t a rock-solid platform. More than privacy issues, more than the layout of the app, it has to work as expected every time. So far this has not been the case.

Usability

How do you enable Health integration in an app? This turns out to be a much trickier question than one would have assumed. Showing deference to user privacy, apps are not able to just began accessing Health data without permission. This, in and of itself, is a good thing. The problem is that Health is completely blind to any compatible apps until the user take specific steps to enable it within the app. Some apps prompts for Health integration on launch. Others require the user to dig into the settings. Still others will only prompt when very specific steps are taken. I would not be surprised to discover that most users have no idea that many of their apps are compatible with HealthKit in the first place.

Once the connection between Health and the app is granted things can actually get somewhat more confusing. Each individual item that Health is capable of dealing with is a separate toggle switch. With some apps this could mean as many as 20 or more on off switches to enable integration. And then you have to consider the difference between reading data and writing data. This may come off as options overload for the average user.

Health app options. Lots of switches for on and off.
That’s a lot of on/off switches.

The way the data is laid out is clean and pleasing to look at but it can feel somewhat incomplete. I can see how many steps I’ve taken per day and roughly when they were taken, but I don’t get the nice breakdown that I would see in something like Fitbit or Jawbone. Sleep data is even worse as health merely supports whether or not you are asleep and completely ignores data such as light sleet and deep sleep that most other fitness apps are able to track.

Buy-In

The third problem for HealthKit maybe one of the biggest, and that’s buy-in from third party apps. So far some of the biggest players in the space such as Nike, Fitbit, RunKeeper, and Withings have refrained from adding support for the Health app. Apple needs support from these players in order to make this a viable platform. Even some of the companies who have supported HealthKit have done so tepidly. Misfit’s latest update adds support only to add steps into the health app with no option to read data or work with sleep, calories, or workouts. Lose It, a popular food logging app, has only added support to read from Health, not to add any data to it.

An even bigger problem maybe delivering on the promise of integrating with the actual real world healthcare system. It’s too early to say a lot here as doctors offices have not yet had time to really consider how they can integrate with this platform. But consider that most offices still have not moved beyond pencils and clipboards. There is a real opportunity here for a true breakthrough in healthcare. But I think it’s going to be sometime before we see that.

The Good

But there is a lot to like about Health. As of last month I was actually wearing three different fitness trackers on a daily basis, a Fitbit Flex, a Jawbone UP24, and a Withings Pulse. This in addition to having some apps running on my phone. Now, using the iPhone 6 with the motion co-processor, I can track steps and physical activity without wearing any sensors on me at all. It’s been much more comfortable than having to strap all of these devices on me every morning.

Even more though, I’m looking forward to the syncing capabilities of HealthKit. Earlier I mentioned the usability issues having so many on/off switches for reading and writing of data. But there is a benefit to this. An awful lot of the sync issues I mentioned in the introduction are taken care of by having this level of control over what data is written into other apps. This is something I always wished the fitness apps APIs would allow for.

I really want HealthKit to work very badly. I went so far as to elevate it to the first page of apps on my phone, which may not sound like much but for me that is a major endorsement, if only of its potential.

There is no question that this platform got off to a very rocky start. And the unfortunate reality is that this can have long-lasting effects. I will bet most of you have not used Apple map since it’s botched introduction two years ago. This despite the fact that the service has gotten many times better since then. I think Health is still too small a segment of the market to have that kind of an effect at this point. But I really want to see Apple step up their game now and bring some real improvements to this quickly. Otherwise it really runs the risk of being moved into that folder of apps you wish you could delete but can’t.

The Problems and Promise of HealthKit was last updated October 8th, 2014 by Michael Truskowski