All you need is 15 minutes with any researcher, scientist, or physician to realize that data is a key component of their work. Spend another 15 minutes with them, and you’ll also come to understand how hard this data can be to come by (especially if the data must come from a human population). It’s because of this, that an entire multi-billion dollar industry exists, running studies, trials, and population tests for anyone with a stake in healthcare.
Apple introduced ResearchKit at their March 9th, 2015 event. While it certainly took a backseat to Apple Watch news for most of their customers, ResearchKit was important enough for Apple to give it a prominent mention during their live event: http://www.apple.com/live/2015-mar-event/
What is ResearchKit?Simply put, ResearchKit is an open source framework that allows for easy collection of data used in medical research. The data ResearchKit can collect is only limited by hardware, sensors, and the iPhone itself. With ResearchKit you can track movements, take measurements, and collect data for your studies. It can even be used as an additional recruitment channel to locate study participants. But the real beauty is its convenience, as participants can now supply incredibly valuable data with medical researchers instantly through their personal devices.
But given we can already develop apps to do this with the current iOS, why is ResearchKit important at all? At it’s core, ResearchKit provides a framework that makes all of the aforementioned activities (e.g. population data collection, etc.) much simpler. It accomplishes this by providing software developers a set of pre-built, customizable modules that remove a lot of work from the equation when it comes to building high-quality clinical research apps. It’s also designed in a way that enables developers to focus heavily on researchers’ needs, and as an open source framework, allows them to give back to the greater development community through updated and improved modules.
To start out, Apple created three customizable modules they feel address the most common elements of clinical research studies: surveys, informed consent, and active tasks. I will touch on these at a high level in this post, but please read their technical overview for more detailed information: https://developer.apple.com/researchkit/researchkit-technical-overview.pdf
Traditionally, asking your research population direct questions has been a huge undertaking involving quite a bit of engagement through phone calls, emails, mailers and more. Obviously, an expensive and time-consuming set of activities to complete. With ResearchKit’s built-in Survey framework, researchers will now have the ability to leverage a very intimate and trusted device to ask study participants questions. It also allows researchers to localize their questions based on a specific population’s localization settings. In other words, if a study participant speaks Spanish primarily, the module would be ‘smart’ enough to recognize that, and provide the survey to the participant in Spanish.
The data typically collected in clinical research studies is highly sensitive. For this reason, it’s essential that participants clearly understand what information they are sharing, where the data they share is going, and who will have access to it. ResearchKit’s Informed Consent module allows researchers to insert approved language and design the content flow they want a participant to follow in a study. If needed, researchers can also insert tests to ensure participants fully comprehend the details of a study. Lastly, the Informed Consent module allows researchers to obtain signatures, and deliver those as signed PDF’s to a server, email address, or the app itself.
For many studies, researchers will need data that is not available through the general capabilities of the device. To address this, Apple’s ResearchKit includes five Active Task modules to help gather the following data categories:
- Motor Activities - Gait, Tapping
- Fitness - 6-minute walk
- Cognition - Spatial Memory
- Voice - Phonation This allows researchers to put together the tests they need to improve the data from their participants. A great example of this is in Apple’s ResearchKit video demonstrating how a Parkinson’s patient can tap on a phone’s screen to help gauge their current motor skills: https://www.youtube.com/watch?v=cMt0Q06UeFY
Potential Limitations of ResearchKit
While all of this seems quite promising, ResearchKit also has limitations. First, it does not give developers “automatic compliance” out of the box, which means you still need to think through HIPAA requirements and any research guidelines that need to be adhered to. Second, it does not ensure secure communications between the app and the backend server; this is the developers responsibility. Third, there is no “standard” or data format for how data is serialized. Again, this puts the burden on the developer.
Also, to date, ReseachKit has not been officially released. While announced, it is in closed beta with select researchers who have released five apps to test the framework. We expect that it will be released sometime in Q2 but do not know exactly. When it does, we predict that the medical research community will release apps using this framework. What will be interesting to see is how well adopted it will be by the general population.