HealthKit

By Nick Pelletier

As we’ve evolved from Hunter-Gatherers to desk dwellers, we have become less active and consequently more unhealthy. Sixty percent of the US population is overweight and the average US male, ages 30 to 39, has a body mass index of 29 (one point shy of obesity).  Most deaths are related to cardiovascular diseases that stem from these very weight issues, most of which are preventable. All of the treatment to care for these issues add up to billions of dollars spent each year for diseases that could have easily been prevented.

Thanks to mobile and cloud technologies, healthcare is becoming more personalized. Individuals can track their health on a daily basis and make adjustments to their diet or fitness accordingly. This puts the responsibility on the individual, not their physician, to make health changes.

HealthKit is Apple’s attempt to carve out a niche in the healthcare market.

HealthKit is a data collection and distribution framework that stores all the health data collected from your phone, apps, and accessories. It is essentially a data store with a frontend, Apple’s native Health app, for users to access and manage the data. Apple’s goal is to standardized apps and accessories to communicate with iOS HealthKit Framework.

The are two major aspects of HealthKit that make up its framework: health data types and data access. The health data types are used to categorize all data into common and manageable units related to health. Data access is a standard to help developers to query the different data types and calculate statistics from them.

Health Data Types

The data types are divided into two categories: characteristics and sample:

• Characteristic data such as date of birth, sex, or blood type are static and don’t change over time.
• Sample data like steps, calories, or heart rate are not static but rather change over time, updated by other apps or accessories. Sample data has a start and end (even if you’d like to track one point in time, the start and end are the same). The sample category is more dynamic than characteristics as it can also be used to make up composite samples.

Here are common data types within HealthKit. It includes samples and characteristics.

As seen in the diagram below, in the family tree of health data types, composites (such as correlation and workouts) are the children of samples. Correlation data of a banana would be Vitamin A, Potassium, Vitamin B6, etc. Essentially all of the nutrients that make up the banana.

Respiratory Rate, Steps, Distance and Heart Rate are Workouts data.  They are activity based health metrics. Workouts data is the summary of activity from duration, distance, energy burned to related data points such as heart rate.

 

In total Apple has more than 60 different health data types. This way, app developers and users can easily access, collect and manage health data.

Data Access

Apple has opened up privisioned access of this data to software developers.  They can query data that is stored in HealthKit from other applications or devices. Some available methods for gathering data are sorting, setting time frame, observing changes, and calculating statistics (sum, min, max and average).

Developers, for example, can sort out all of the Potassium I’ve consumed, set time frame on steps taken yesterday, and notify the user when existing data has changed in this case when I have taken more steps. Finally, developers can calculate statistics with all or a particular set of data.

Privacy

The most critical aspect of HealthKit is keeping the collected data private. Apple has taken more than the necessary steps to instill confidence into their users. All data is saved locally and encrypted when locked. Data is not saved to cloud or synced between other devices. To access or share data, the app must have the users express permission for each data type.

The types of permissions are read and write. Write data simply means sharing your data with HealthKit. To read data from another fitness app, you must enable reading permission in HealthKit. For example, the user gives MapMyRun permission to write calorie data to HealthKit. The user can then allow MyFitnessPal to read this calorie data from HealthKit.

Essentially, the user can explicitly allow certain data be collected by HealthKit but not allow data pulled by another app from HealthKit.

It’s worth noting that an app developer will not know when a specific type of information is blocked (or in other words when the user does not allow it to be read) because that knowledge in and of itself can compromise privacy (i.e. blocking blood sugar might suggest diabetes).The app (or developer) will interpret it as the data not existing within HealthKit.

For better user experience,  this mean developers should not assume all data is available to query but guide user on deciding whether or not they would like to disclose the information for another app.

Apple has taken its privacy one step further by restricting the types of apps that can use HealthKit:

• Only health and fitness apps that use the data for improving health and medical research are allowed, no apps for marketing or advertising
• Developers are not allowed to share data with 3rd party without express consent from user.
• The app must include policies from HIPPA or Personal Health Record within its own privacy policy and developers are also responsible for clearly stating its integration with HealthKit in marketing text and its UI.
• Apps that provide diagnosis/treatment or controlling hardware must provide written regulatory approval.

With HealthKit, Apple has established their framework for developers to leverage when building products for iOS. HealthKit will make development easier and allow for innovation, but it comes at a price of being tied to iOS. We have seen some companies, like FitBit, openly admit that they won’t work with Healthkit as they feeling it is limiting. That said, they are also building an Apple Watch competitor. HealthKit, overall, is a great start to standardizing health data and hopefully results in individuals becoming more health-conscious.

---

Filed under: Product Engineering | Topics: apple, Design, Healthkit, iOS, IOS 8, ios developer, ios development, UX