52 Things About ResearchKit That Every Medical Researcher Should Know

Challenges in medical research

• Medical researchers are faced with many challenges today - from lower participations rates, selection biases driven by demographics of research sites, high costs for collection of data to lower volumes of data.

That’s great ! But, just how exactly does ResearchKit live up to this hype?

Here is how!

ResearchKit

1

• ResearchKit is an open source software framework that provides core functionality for developers to build mobile apps to use for medical research .

2

• ResearchKit apps can take advantage of the many powerful processors and sensors ( for example, tracking movement) that are standard with today’s smartphones and that enable researchers to gather new types of data.

3

• By making the transition from a traditional pen and paper questionnaire to an engaging, fill-where-you-are app right on their smartphone, ResearchKit apps enable researchers to exponentially increase the number of participants.

4

• The ResearchKit is open source and enables app developers to build apps both on iOS and Android devices, making its potential truly far-reaching..

5

• For apps built on iOS, ResearchKit works seamlessly with Apple’s HealthKit. HealthKit is a framework that collects data from various health and fitness apps and makes that data available to Apple users through the Health App.

6

• It encourages higher participation in research studies by leveraging the ‘any where any place’ premise of the mobile to collect data thus helping researchers to overcome the resource and logistical constraints often faced by them.

7

• Higher participation means higher more robust volumes of data (Does Big Data come to mind? ) available for researchers to analyze and draw inferences.

8

• Any app built on the ResearchKit platform has to be IRB approved before it can be downloaded

ResearchKit Modules

9

• ResearchKit provides software developers a set of pre-built, customizable modules that can be used to build quality research apps aligned to the research’s purpose and designed to make data collection that much more effective

10

• ResearchKit framework provides three modules – Surveys, Informed Consent and Active Tasks

11

• ResearchKit uses the concept of Tasks and steps for all user actions and data collection.

12

• Task is a collection of one or more steps - each step in turn handles either a single or a collection of actions within the app - each such action is associated with a result (data) when the user completes the action.

13

• The ResearchKit framework inherently supports managing disruptions caused by some of common ‘app usage’ behaviors. For example, it allows for saving progress in the middle of a long running task so it can be stopped and started again or for restoring the UI state if the user had to suddenly switch out of the research app (to answer a phone for example)

14

• The data generated by the user actions can either to saved locally in a secure fashion (so it can be downloaded at a later time) or can be uploaded directly to a remote server.

Informed Consent 

15

• Considering the sensitive nature of the data collected through research studies, the Informed Consent module is a pivotal one.

16

• Researchers can use this module to create a model (structure) of a consent document with customized content,display it for visual consent and allow for review and signature by the user. Images, videos and quizzes can be used to turn a normally drab interaction to a more engaging one.

17

• It is important to note that this module does not currently support digital signatures.

18

• Researchers can easily integrate pre-approved IRB consent templates thereby ensuring compliance.

19

• The ResearchKit framework provides some of the core sections like that are usually included in most of the consent documents. These include Overview, Data Gathering, Privacy, Data Use, Time Commitment and Withdrawing.

Survey

20

• The Survey Module consists of a pre-built user interface, that lets developers quickly build surveys by specifying the questions and the preferred answer types.

21

• Instruction step - Used to provide information/instruction to the user at the beginning of the survey. It can be used to reiterate any survey requirements to the participants at the start of the survey.
• Question step– This is used to to ask the user a single question with multiple possible answer formats such as numeric, text etc.
• Form step – This allows for a multi-question survey to be presented as a single step on the same page.

22

• This module supports multiple answer formats to capture user responses. Examples include Scale, Boolean (for ‘yes’ or ‘no’ answers) ,Value Picker (selection from a list of values) ,Image Value, Text and Numeric (single or multiple choice) and Date and Time selections.

22

• This module supports multiple answer formats to capture user responses. Examples include Scale, Boolean (for ‘yes’ or ‘no’ answers) ,Value Picker (selection from a list of values) ,Image Value, Text and Numeric (single or multiple choice) and Date and Time selections.

23

• The survey module also allows for questions to be customized based on the geographic location of the user.

Active Tasks

24

• With the Active tasks module, users can perform certain fitness and cognitive activities and the data captured by those actions are collected through the iPhone’s sensors. For example, users could walk a short distance and the iPhone’s accelerometer would collect the ensuing data.

25

• Active Tasks module currently includes five different tasks encompassing these actions ranging from fitness to audio and balance as explained below.

26

• Fitness Task– This typically involves users doing a physical activity and collating the data collected such as accelerometer, device motion, location to the app.

27

• Audio – AVFoundation framework is used to collect data from audio recording of the research subject making a sustained sound for a period of time.

28

• With the Active tasks module, users can perform certain fitness and cognitive activities and the data captured by those actions are collected through the iPhone’s sensors. For example, users could walk a short distance and the iPhone’s accelerometer would collect the ensuing data.

29

• With the Active tasks module, users can perform certain fitness and cognitive activities and the data captured by those actions are collected through the iPhone’s sensors. For example, users could walk a short distance and the iPhone’s accelerometer would collect the ensuing data.

30

• Spatial memory – By asking users to perform memory recall activities, this task allows researchers to assess the executive function and visuospatial memory of the participant.

Data Privacy

31

• One thing that goes hand in hand with consent is data privacy. The one big question in everyone’s mind is “Will Apple see my data?” .And the short answer is ‘No”.

32

• Now that it's established that ResearchKit will not be relying on Apple’s servers for data storage, the onus falls on the developers to ensure 100% compliance in both data storage and transmission.

33

• As guidelines state, the app itself would not be subject to HIPAA if the user permits the app to send information, but such transmitted information would be subject to HIPAA once the research organization receives it. This blurring of the lines is something that needs to be dealt with caution by study sponsors and developers , keeping in mind the extremely sensitive nature of the data.

34

• ResearchKit works with Healthkit to seamlessly collect data from the Health App, if the user consents to this.

Potential Challenges 

35

• Fear that research populations would include only iPhone users for now resulting in limited race and gender diversity.

36

• High Attrition potential – as the participants sign up in high numbers, they may also drop off in equally high numbers if they are not kept engaged and motivated to continue the study.

37

• Participation bias – Since participants self-qualify (especially in cases where the app is publicly available) they may tend to fudge some of the qualifying facts if they are very motivated to participate.

38

• The ‘Minor’ Predicament - One of the ethical dilemmas that ResearchKit is facing is that of confirming eligibility criteria of participants. With a significant section of the smartphone users being minors, this remains a sticky issue, though researchers feel that impersonation might not be a huge problem considering the length of time and involvement that goes into these studies.

39

• Reliability of data – Concerns about reliability of the data collected through the app stemming from concerns around selection and participation bias as well as varying degrees of participation/engagement throughout the study period.

40

• Ethical and Privacy concerns – Making sure that the participants truly understand the scope of the study and its objectives and what the data from the study will be used for.

Value add to -participants 

41

• Putting People Front and Center - User friendly ResearchKit apps make it incredible easy for people to participate, engage in and gain key insights into symptoms and managing their health conditions.

Case In point

• Over 10K participants signed up for Stanford medical trial after the ResearchKit debut. To get 10,000 people enrolled in a medical study normally, it would take a year and 50 medical centers around the country.

42

• Participant Empowerment: Traditionally, sponsors made the decision on how data from research studies could be used. But with ResearchKit, a lot of that power rests with the participants. They can choose to make their data available only to the researchers and sponsors of the study, or they can choose to share it among a broader audience, paving the way for any other studies related to the area of research. This clearly restricts who has access to the data.

Value add to Researchers

43

• There is a higher volume of data available for research due to greater participation and more precise data as it is captured by iPhone sensors based on user actions.

44

• Participation from wider geographic locations may help to avoid selection bias that may be introduced due to demographics surrounding a physical research site.

45

• Making the research process more efficient and cost effective will encourage more research activity.

46

• Reduces the mundane and repetitive tasks from the researchers allowing them to focus on core research functions.

Adding Pizzazz to ResearchKit Apps

47

• Now that we know what ResearchKit can do, its time to bring on the bells and whistles. ResearchKit apps have the potential to drastically reduce the cost of medical research studies.

48

• Developers can use the ResearchKit framework to create user- centric apps that can be engaging and enable participants to fill in data more intuitively, rapidly, efficiently and accurately.

49

• Custom branding of apps can bring in more study sponsor visibility and can drive adoption, loyalty and brand awareness.

50

• Additional push notifications can be built in to drive up the participation rate in the study and the level of engagement. Usability of the app will be critical to keep the participants engaged in the study and to control the attrition rate.

51

• ResearchKit apps can be taken a step further and be integrated with backend Health information systems to increase the value of the meaningful data being collected.

52

• Developers can build interactive dashboards that can transform data collated into a rich, visual and animated interface for researchers to get an easy to read snapshot of data.

There You have It!

For more visit
www.boston-technology.com