Keeping Up with Technology: An Introduction to Creating Apps for Clinical Research Using ResearchKit

Sacha McBain, M.S.
Sacha McBain, M.S.

Sacha McBain, M.S .
University of Arkansas for Medical Sciences, Pre-Doctoral Clinical Psychology Intern

With approximately 77% of Americans owning a smartphone (Smith, 2017), the potential positive impact of mobile technologies on health and access to care are exciting and varied. Mobile health apps, also known as “mHealth,” represent a growing sector of apps available in the Apple App Store and in Google Play, with a wide range of utility including the augmentation of care and facilitation of patients’ self-supervision of chronic illness. The use of mHealth also has implications for reaching patients who may not otherwise access care or do so on an irregular basis. Numerous studies have explored the impact of apps as an intervention for a wide range of health behaviors (e.g., treatment adherence; Pereira-Salgado et al., 2017) and mental health (Langarizadeh et al., 2017). Replacing traditional physical diary cards or therapy workbooks with mHealth apps may create greater ease of use of therapeutic tools and has the potential to increase accessibility and motivation to engage in therapeutic homework assignments, especially self-monitoring. In recent years, more researchers have begun exploring the potential role of mHealth in transforming how clinical research is conducted. The use of mHealth in clinical research provides a means of innovation in participant recruitment methods as well as intervention development and dissemination. In 2015, Apple introduced ResearchKit, a first of its kind open source framework designed specifically to support the development of apps for population-based medical research (Zens, Woias, Suedkamp, & Niemeyer, 2017).

Defining Open Source Tools

An “open source” tool, including ResearchKit’s open source framework, is software that is publicly accessible at no cost to the user. This means that anyone may view, change, or share the source code. The benefit of having access to the source code of a tool or program is that the community of developers who use the tool have the ability to correct potential errors, improve the code, or modify it to their unique needs and share it with the rest of the community without having to wait for updates from the original creators. This means innovations can happen at a faster rate compared to proprietary tools and there is generally more flexibility in what can be accomplished using the tool. To simplify, ResearchKit can be roughly defined as a freely (i.e., openly) available coding template specifically designed with development of medical research apps in mind. Although ResearchKit is novel in its focus on supporting development of apps specifically related to medical research, there are numerous open source and proprietary options to support app development including FileMaker, Xamarin, Adobe PhoneGap, Intel XDK, Ionic Framework, and many others.

ResearchKit Components

ResearchKit’s core framework (as originally designed by Apple) provides examples of code in two different coding languages that developers can use to create surveys, create consent forms, customizable templates for app content, and include Active Tasks. Active Tasks are within-app activities that users engage in while iPhone sensors collect data. These Active Tasks are organized into six categories including: Motor Activities (e.g., using the accelerometer to collect device motion data to capture range of motion), Fitness (e.g., using gyroscope to collect heart rate data), Cognition (e.g., using accelerometer to capture device motion to measure reaction time), Voice (e.g., using microphone to capture uncompressed audio recordings), Audio (e.g., recognition of tones and minimum amplitude), and Hand Dexterity (e.g., using multi-touch display to record completion time). For example, mPower, developed by the University of Rochester and Sage Bionetworks, utilizes the Motor Activities Active Task to have participants tap on the screen which is recorded using the multi-touch display and becomes data that is used by researchers to assess hand tremor.

Data Management and Extraction

After data collection, the ResearchKit framework includes code to automatically encrypt the study data, which is exclusively available to researchers and is never seen by Apple. Data is de-identified by replacing identifying information with randomly generated code and is HIPAA compliant. There are options to add passcode or touch ID to control access to the app, however, there is the potential risk of others accessing the app if they are aware of the user’s password. Data sharing is based on specifics of each app which is included in the initial consent process. Data analyses are completed outside of the app after data extraction. Data extraction code examples are included in the framework.

ResearchKit in Action

Since ResearchKit’s launch in 2015, numerous apps have been released that may have clinical relevance for practicing health psychologists, ranging from LGBTQ health, microbiome, hepatitis C, cardiomyopathy, HIV associated neurocognitive disorders, epilepsy, endometriosis, autism, depression, COPD, concussion and many more.

As ResearchKit developed apps are uploaded to the app store, researchers gain access to millions of potential participants to potentially accelerate enrollment and completion of clinical studies. University of Rochester’s Parkinson’s Disease app mPower has enrolled over 10,000 participants to better understand factors related to symptom reduction, such as sleep and mood, in Parkinson’s disease (Bot et al., 2016). Strikingly, 93% of participants enrolled in mPower reported having never taken part in any research study, highlighting the powerful role technology can play in conducting large scale medical research (Apple, 2018).

In addition to expanding recruitment pools for medical research, ResearchKit can facilitate development of apps that can also increase access to care. Duke University and University of Cape Town’s Autism & Beyond utilizes the front facing camera of the iPhone and facial recognition technology to assist in the detection of early signs of developmental issues allowing researchers to collect valuable data about early detection while providing users with access to earlier diagnosis and treatment.

Apps developed by ResearchKit can also provide valuable augmentation of existing care and improve patients’ self-supervision of chronic illness. Johns Hopkins University’s EpiWatch is an app developed for Apple Watch that is designed to collect data from the Apple Watch’s accelerometer and heart rate sensor to determine how to predict seizures. The app enables users to track the onset and duration of seizures. Users can collect data to create correlations between episodes and medication and set up automatic alerts to be sent to a family member or caregiver when seizure activity is detected.


There are several key limitations of ResearchKit including concerns regarding accessibility both in terms of who is able to develop ResearchKit apps and who is able to access apps. Unless one has significant personal knowledge of app development and coding, those interested in utilizing ResearchKit will likely need support from software developers. For those with limited resources, it may be possible to utilize undergraduate or graduate students studying software design to reduce costs. Despite the intent of ResearchKit to increase diversity in medical research participation through population based recruitment, participants continue to be primarily young, white, and male. The high cost of Apple products may also influence who has access to ResearchKit-developed apps, resulting in potentially skewed research and an exacerbation of health disparity gaps. Given the open nature of the framework, it is possible for developers to create ResearchKit applications for Android devices, however, this represents a very small percentage of content creation.


ResearchKit is a relatively new tool developed by Apple that supports the development of apps designed specifically for clinical research. It includes templates for developing consent forms, content design, and guidance on maximizing the iPhone’s built in sensors to collect action oriented data. Since its inception, it has facilitated the development of a wide range of apps that may be relevant for health psychologists including hepatitis C, COPD, depression, cancer-related PTSD, and more. However, there continues to be a significant opportunity for further development of behavioral health oriented apps in which health psychologists can lend content expertise.


Apple (2018). ResearchKit. Retrieved from

Bot, B. M., Suver, C., Neto, E. C., Kellen, M. Klein, A., Trister, A. D. (2016). The mPower study, parkinson disease mobile data collected using ResearchKit. Scientific Data, 3(3). doi: 10.1038/sdata.2016.11.

Langarizadeh, M., Tabatabei, M.S., Tavakol, K., Naghipour, M., Rostami, A., & Moghbeli, F. (2017). Telemental health care, an effective alternative to conventional mental care: A systematic review. Acta Inform Med, 25(4), 240-246. doi: 10.5455/aim.2017.25.240-246.

Pereira-Salgado, A., Westwood, J. A., Russel, L., Ugalde, A., Ortlepp, B…Schofield, P. (2017). Mobile health intervention to increase oral cancer therapy adherence in patients with chronic myeloid leukemia (the REMIND system): Clinical feasibility and acceptability assessment. Journal of Medical Internet Research mHealth and uHealth, 5(12), e184. doi: 10.2196/mhealth.8349.

Smith, A. (2017, January 12). Record shares of Americans now own smartphones, have home broadband. Pew Research Center. Retrieved from:

Zens, M., Woias, P., Suedkamp, N. P., Niemeyer, P. (2017). “Back on track”: A mobile app observational study using Apple’s ResearchKit framework. JMIR Mhealth Uhealth, 5(2), e23. doi: 10.2196/mhealth.6259.