In my blog last week, I touched upon how Apple is once again attempting to fundamentally impact and disrupt an industry that is very change averse with a lot of entrenched interests and challenges. Apple’s ResearchKit (an open source software platform)in many ways promises to democratize medical research by making it easier for the masses to participate in it while at the same time making it efficient (and cost effective) for the researchers.
In today’s blog I am going to touch upon at a high level what’s included under the hood in ResearchKit as well as more functional details on one of the built-in modules – Surveys.You can get into more implementation level details here.
At its core,ResearchKit provides three modules that can be customized and built upon by the developers.These modules provide base functionality for items commonly needed in medical research such as surveys, participant consent and simple active tasks that users perform such as walking, sitting and talking.
ResearchKit uses the concept of tasks and steps to handle all app functionality including user interaction and data collection(research and meta data).A task is essentially a collection of steps – each step handles either a single or collection of actions within the app with an associated result when the user completes that step. The framework inherently supports handling some of the common user behaviors when using an app – such as giving an ability to save progress in middle of a long task (such as taking a multi-page survey) and restarting it later or saving and restoring UI state if the user had to switch out of the app suddenly (to answer a phone call for example). The results data can be either saved locally in a secure fashion or uploaded to a remote server securely for further processing and analysis – this being the crux of medical research.
The Survey module provides multiple types of steps such as:
- Instruction step – provide information/instruction to the user at the beginning of the survey. For example,a survey may need for user to not have eaten for an hour before taking a survey to get accurate responses.The instruction step can be used to reiterate that requirement at the start of the survey.
- Question step – allows the app to ask user a single question with multiple possible answer formats such as numeric, text etc. For example, the survey may ask the user to describe the food that she ate last or ask to make a choice from a list of food options
- Form step – allows for a multi-question survey to be presented as a single step on same page. For example, the survey could ask through a series of questions on the form breakdown on the type of food the responder ate last.
The Survey module also supports multiple types of answer formats such as:
- Scale – which can be used to ask users on a scale of say 0 to 10 how much they liked the last food they ate
- Boolean – this is used for a &’yes’ or &’no’ question
- Value Picker – used for selection from a list of values
- Image choice – my favorite ☺️can be used to ask user how they felt after eating their last meal
- Text and numeric – single or multiple choice
- Date and time selections
So as you can see the Survey module provides a lot of functionality out of the box already that the app developers can use fairly quickly to construct the survey component of any medical research app. In the next blog post, I will touch upon the other two modules provides by ResearchKit – Consent and Active Tasks.
Is your company building apps using ResearchKit? If yes, I would love to hear about your experience.
Learn more about ResearchKit. Click the below image to watch our hugely successful webinar –