Apr 29, 2016
Behavioral economics, positive psychology, organizational behavior, econometrics… what do all of these fields have in common? A focus on the impact of mood on outcomes and behavior.
For too long, mood has been a largely subjective and often ill-defined target of study. Linked to a wide range of environmental factors and social determinants – such as pollution, sleep, social connectedness, and access to food – mood is the subjective, free-floating extension of our emotional experience. We know that mood is one of the keys to health; whether you are happy, depressed, stressed out, anxious – all impact your physical wellbeing. However, our knowledge of the relationship between mood and social, environmental, and economic factors remains limited, despite decades of study.
While many previous ResearchKit studies have focused on improving the patient experience for specific ailments and disorders, the recently launched Mood Challenge for ResearchKit, a New Venture Fund program funded by the Robert Wood Johnson Foundation, seeks to bring new insights and capabilities around psychological health and happiness to everyone with an iPhone.
Apps exist to help track mood, and there are clinical assessments used to measure it, but never before has there been an opportunity to combine self-reporting with background data analysis across thousands of subjects to paint a more nuanced picture of mood. The ResearchKit platform combines the standardization and reliability needed for scientific viability with the accessibility of a frictionless tool already in your pocket.
Beyond quantifying correlations between mood and social context, and translating analog clinical assessments into frictionless digital tools, we believe the most exciting aspect of the Mood Challenge will emerge from unmeasured and yet unimagined analyses. The most intriguing and successful studies will develop active and passive tasks that combine disparate data streams and user input to answer questions about what our phones know about our mood and mental wellbeing. The fact that these questions are yet unanswered (and, in many cases, unasked) is what makes ResearchKit – and the Mood Challenge – so exciting.
The Challenge will award $500,000 in prize money as well as expert mentorship and opportunities to pilot new studies. To learn more about and to submit a proposal, visit: www.moodchallenge.com.
- The Mood Challenge for ResearchKit Team
Apr 4, 2016
This week 23andMe introduced a module that allows researchers to seamlessly integrate genetics into their ResearchKit apps. The module and its documentation are available on GitHub.
"This new technology gives researchers a turnkey way to integrate genetics into their studies," said Anne Wojcicki, 23andMe CEO and co-founder. "This will enable research on a much broader scale. Incorporating genetics into a platform with the reach of ResearchKit will accelerate insights into illness and disease even further."
The 23andMe module will allow interested researchers who are using the ResearchKit platform to support the collection of genetic data for their studies from current 23andMe customers and beyond. Using the module, researchers can allow existing 23andMe customers to easily contribute their data to a study. Or, researchers could choose to offer genotyping services to their study participants through 23andMe, with the services being funded by the researchers. Participants will then have access to the full 23andMe service, and researchers will have a simple, low cost way to incorporate genetic data into their studies.
Once a participant agrees to share their genetic data with a study, the researcher can then use the credentials they receive from the module to access the participant's genetic data via the 23andMe API.
Data sharing within the context of the ResearchKit module has been structured to provide the same robust privacy protections as all 23andMe research. 23andMe only allows a researcher to access data when the user explicitly agrees to share that data with the researcher by providing informed consent, and the participant can disconnect from an app in their 23andMe settings anytime to prevent any future information sharing. The information 23andMe shares on our customer's behalf with third party research apps is de-identified and encrypted, safeguarding the data during transfer.
We also require that any partner using the 23andMe ResearchKit module have their study approved by an Institutional Review Board, which is an independent panel of experts which ensures that all research is conducted in accordance with all federal, institutional and ethical guidelines.
23andMe has created a development environment for researchers looking to include the 23andMe module in their app. Send an email to GeneticsModule@23andme.com for more information on accessing the development environment.
If your app will only accept data from existing 23andMe customers, then you may not need access to the development environment. You can create a development API account here.
- The 23andMe Team
Feb 2, 2016
Concussions are a form of traumatic brain injury that can lead to long-term consequences if not treated properly. They can happen at any time: a teen taking a tough hit during a football game, an older adult falling and hitting their head, or a driver suffering head trauma during a car accident. The goal of NYU Langone Medical Center's ResearchKit app is to obtain a better understanding of how activity, as well as rest, impacts recovery for those diagnosed with concussion.
A collaboration between physician researchers from NYU Langone's Concussion Center and clinical informatics experts with the Medical Center's IT department, our research team has started testing whether the new NYU Langone Concussion Tracker ResearchKit app can help those with concussions better track their symptoms during the critical six weeks following their diagnosis.
We're excited that with iPhone and Apple Watch, we can now evaluate a potentially large percentage of the concussion population across the country to gain daily insights into symptom profiles for patients. We used modules that capture symptoms (including balance issues, blurred vision, and drowsiness), a six-minute walk test, and tasks to measure concentration.
By tracking these measurements on a daily basis, instead of every one to two weeks at appointments, this app and the related research project will let us assess current treatment protocols in ways not before possible, including greater understanding of how patients' concussion symptoms improve over the course of their recovery. For instance, after being diagnosed with concussion people are primarily told to rest, but 50 years ago that's the same thing we said with heart attack. Now we know this guidance was incorrect, and the treatment has since evolved.
We took the open source ResearchKit framework and customized it a few ways, with a strong emphasis on user experience. For instance, we tailored the implementation of the design to utilize color schemes that are more accessible to concussion sufferers. Since individuals with concussions are more sensitive to low contrast, we used orange prominently because that color is easily identified by concussion sufferers.
Participant access to data was another important element we took into consideration when designing the app. For patients at our Concussion Center who are involved in the study, our ResearchKit app integrates with our electronic health record (EHR) system, allowing them to view their results online in their personal "MyChart" account within the Epic EHR. For those who download the app and are not an NYU Langone patient, they can review the results of their activities over time in the Concussion Tracker app, which they can then share with their physicians.
- Paul Testa, MD, Chief Medical Information Officer and Assistant Professor of Emergency Medicine at NYU Langone
Jan 19, 2016
"It takes a village...."
The Autism & Beyond app is a meaningful example of using a mobile device to conduct research that otherwise would have required a visit to a research facility. The video capture capabilities of iPhone combined with an advanced algorithm for recognizing facial expressions makes it easy for a parent and child to complete the study at home.
Although technology is an enabling factor, that wasn't enough. We learned that the Autism & Beyond study could be successful only with the cooperation of a large team of experts from various disciplines, over 30 experts in our case: child psychiatrists and psychologists, pediatricians, neuroscientists, epidemiologists, global health researchers, engineers, developers, designers, and computer scientists.
While the Autism & Beyond ResearchKit study has been ongoing for months, it is just the fruit of a much broader initiative that had been germinating for years. Specifically, the Child Mental Health Initiative was formed two years ago between the Pratt School of Engineering and Arts and Sciences at Duke University, led by Dr. Guillermo Sapiro and Dr. Helen Egger. Each week we met with the goal of creating affordable and accessible tools to screen for childhood mental health disorders.
We benefited greatly from having a large multidisciplinary team from the very beginning. Rather than one individual coming up with ideas for the team, each team member played a critical role in all areas of the study from beginning to end. This was particularly evident during the pilot testing phase. For example, as engineers and computer scientists were testing the feasibility of collecting facial expression data from the videos, the psychiatrists, psychologists, and neuroscientists worked directly alongside them to assist in capturing the precise data that would be most useful in the creation of an algorithm that could one day be used to screen for childhood mental health conditions such as autism. Figure 1 Automated Facial Recognition
The open source ResearchKit framework enabled us to focus on these scientific questions collaboratively rather than spend as much time refining the interface of the self-consent process or study questionnaires. It also allowed us to reach a larger and more diverse audience, which may enable our algorithms to be more accurate and generalizable in the future.
While we continue to collect data and to analyze it, two fundamental conclusions are already clear in the first months: first, the data quality is excellent, proving for the first time that we can observe behaviors in natural environments using ubiquitous devices; and secondly, parents and caregivers are very helpful in sharing data with responsible teams doing fundamental work to help their children. We hope that the collaborative and diverse team we've assembled will ensure that this work will touch the lives of children around the world for many years to come.
- The Autism & Beyond Team
Jan 11, 2016
Today we released version 1.3 of of ResearchKit. Among the many improvements, you'll find a new active task: the 9-Hole Peg Test (9-HPT). This is a digital version of a test that's widely used in rehabilitation medicine to measure finger dexterity and upper extremity function.
Two new modules make it easy to create and verify accounts for your participants and to secure the app with a passcode pin entry. Participants also have the option to use TouchID on devices that support it.
Perhaps most exciting of all, the release includes a sample app that not only shows how to use the main features of ResearchKit — informed consent, surveys, active tasks, account creation, and passcode pin entry — but it also shows how to architect a research app to ensure a great user experience.
The sample app has
- Placeholder pages for providing a preview of the study
- A dashboard page for displaying results to participants
- A profile page that shows key user data as well as providing an easy-to-access link for withdrawing from a study
- An activities page that provides a list of all the study tasks.
You'll be able to quickly prototype research ideas by modifying the sample app. Download the new release and try it out!
Dec 21, 2015
Over 3 million Americans suffer from hepatitis C and yet, we know very little about the impact the virus and medications have on people's daily lives. We developed C Tracker - an Apple ResearchKit study for people living with hepatitis C - because traditional clinical trials are unlikely to elucidate these correlations.
C Tracker introduces an innovation from the Boston Children's Hospital Computational Health Informatics Program, which we call C3-PRO: Consent, Contact, and Community framework for Patient Reported Outcomes. This "backend" connects any ResearchKit app to a widely-used clinical research IT infrastructure called i2b2. C3-PRO relies on an emerging health data standard called FHIR (Fast Health Interoperability Resources). We are making C3-PRO available to the broader research community to be used with their ResearchKit apps.
In the current version of C Tracker, we recruit subjects "in the wild" and maintaining complete anonymity. However, in a future version we're we anticipate "prescribing" C Tracker and other ResearchKit apps to study participants in a way that connects their study data to electronic health record data in i2b2. Through the SMART HeatlhIT apps project, we also are working to connect ResearchKit directly to the point of care.
- Ken Mandl, MD, MPH
Dec 2, 2015
ResearchKit makes it easy to create research apps that are both valuable sources of data and enjoyable to use. There are many things you can do when creating an app to ensure participants feel comfortable and stay engaged in your study - so we wrote a few guidelines to help you make your app delightful for participants. Happy creating!
Design Guidelines for Research Apps
- The ResearchKit Team
Nov 18, 2015
The National Institute on Drug Abuse (NIDA), part of the National Institutes of Health, announces a Challenge to develop novel mobile applications (apps) for future addiction research explicitly created on Apple Inc.’s ResearchKit framework. The apps should be designed to be used in future clinical research studies with human subjects to answer important scientific questions about the paths people take to avoid or to succumb to drugs and to improve the scientific understanding of drug use and addiction. NIDA is also interested in further understanding abstinence and wellness as it relates to drug addictions.
Apple’s ResearchKit lowers the barriers for medical researchers in terms of custom coding. New apps will be created by teams with a combination of IT skills and clinical research expertise. The Challenge submissions may not contain any data about real people, and must comply with all applicable laws and regulations.
NIDA will award up to $100k to the Challenge winners. The submission period ends on April 29, 2016. For information about the NIDA Challenge, “Addiction Research: There’s an App for That,” visit http://nida.ideascale.com.
- The ResearchKit Team
Nov 13, 2015
ResearchKit provides boundless possibilities for data collection. But just because you can pose this survey question or implement that sensor-based activity doesn't mean you should. Aggressively pare down to avoid app fatigue and try to make the activities of your study easy to explain and understand.
In Mole Mapper, we ask users to take photos of different areas of their skin; drop pins on their moles like on map; measure these moles with a photo next to a reference item; remeasure selected moles each month; and take a brief monthly survey. Explaining this process was cumbersome, but sage advice led me to follow the simplifying strategy in Apple's "Rip. Mix. Burn." campaign a decade ago. Thus the above directions were pared down to "Map. Measure. Monitor."
Following the tradition of fusing fields like desktop computing & publishing, or mobile technology & music, I think we are headed for an exciting future of app development & scientific research. I encourage you to get started!
- Dan E. Webster
Nov 11, 2015
Come prepared when meeting with a developer. Before you get together with a developer, you should do your research. Download existing ResearchKit apps and take screenshots to mock up what you would want in your app. If you have not seen this 'Fake it till you make it' WWDC video on prototyping, pause reading this blog post and watch it right now.
Give development a try. You most likely teach yourself unfamiliar protocols and skills all the time in the course of your research projects. Learning the basics of iOS app development isn't inherently different and there are many useful tutorials. While your PCR reactions are running, take some time to watch a lecture from Paul Hegarty's free Stanford video lecture series on Developing iOS Apps on iTunesU. Armed with an understanding of the basics, you will have more productive conversations with developers, and you can tinker with your own nascent ResearchKit app by using the plug-and-play components of ResearchKit from the Getting Started page. Interacting with the ResearchKit consent process and survey tools running on your own iPhone is tremendously motivating.
- Dan E. Webster
Nov 6, 2015
ResearchKit allows unprecedented numbers of people to participate in scientific research in real time. This potential to engage participants on a global level, brings together developers and scientists who may never have thought their fields would overlap.
I have worked in both fields: as a cancer biologist trained in the lab & as the developer of the Mole Mapper ResearchKit app. iOS development and research have more in common than you might think, but bridging the divide between these disciplines can be challenging. If you are a developer or scientist (or developer-scientist) who is just now joining this community, I have tips from my experience that I hope can be helpful as you get started. In my first post, I'll share tips for developers.
Refactor from AppCore. The union of functionalities and implementations for the first 5 ResearchKit apps was brought together to make the AppCore code base. Consider this a rich resource of example code, but not yet a plug-and-play solution. Learn from the code craft within AppCore and refactor pieces for your own needs where possible. I did this for the on-boarding process and the back end integration with Sage Bionetworks' Bridge Server. While an implementation exists in AppCore to integrate with Bridge server, you can directly use the nicely-documented Bridge iOS SDK for this purpose as well.
Maintain a local database. The data from ResearchKit apps will likely be encrypted and sent to a central repository for the research study. If possible, provide your participants with in-app access to their data and an export functionality. Consider keeping survey and activity results in Core Data or check out the APCKeyChainStore class in AppCore. In addition to providing a local backup if your upload fails, this approach enables participants to keep and interact with their contributions to research.
- Dan E. Webster
Oct 23, 2015
We're incredibly excited and humbled to see the medical research advancements that are already springing up from the ResearchKit community. Many of you have expressed a desire to find even more ways to connect with and give back to the community—so we've created a new space to do exactly that: The ResearchKit Forum.
The ResearchKit Forum is a place where you can connect with other community members to ask questions, discuss issues and share best practices and learnings. The Forum provides spaces dedicated to the below topics, and will be expanding to accommodate any new topics the community wants to discuss.
- Study Areas, where you can share and discuss ideas for research studies.
- Study Design, where you can talk about general study design, ethics approval, and engaging participants.
- Data Storage, here you can share and discuss ideas for managing the data collected by your study.
To join, go to https://forums.developer.apple.com/.
We can't wait to see what we can learn together!
- The ResearchKit Team
Jul 31, 2015
From the onset of the PRIDE study the researchers at UCSF envisioned leveraging technology to create the largest LGBTQ health study in history. The diversity of participants within the study population required setting aside traditional notions of study design, which typically involves researchers establishing study hypotheses and longitudinal questions in isolation.
- We immediately ran into a challenge: how could we pick questions to ask this community when
- a) the diversity of the LGBTQ community meant that not all topics would matter to everyone, and
- b) the gap in knowledge about this population's health is so great that any assumptions made would almost certainly lead to the exclusion of important health topics that matter to this diverse group.
To solve this problem, the research team decided to tap into their study participants to find out what health issues were most important to them. To this end, we created a community module that allows users to nominate a topic, explain its importance, and categorize it by health type and relevant demographics (See Figure 1). After the topic is posted, the community can establish a sense of agreement on top topics of interest through upvoting or downvoting topics and engaging in dialogue using comments and replies (See Figure 2).
Figure 1: Community Tab - Research Topic Nomination
Figure 2: Upvotes/Downvotes and Comments on a Participant Nominated Topic
You may be wondering if this approach was successful - and the answer is, absolutely! In just the first week, we had over 11,000 consented enrollees, and by the end of week two, we had over 13,200. The community module had 2,671 health topics suggested for research, with 35,092 upvotes and 2,716 downvotes to help establish community interest.
Surprisingly, critical agreement around issues was reached on enough topics within the first two weeks to start formulating longitudinal and cohort-specific activities. The initial evidence seems clear: there is great power in democratizing research by engaging an underserved yet incredibly willing community, and having them take personal ownership in advancing their own health.
We are incredibly excited about the future of this study and the potential it brings for this community and for other research projects.
May 7, 2015
With our partner Mount Sinai, we launched Asthma Health, one of the inaugural apps built on ResearchKit. The objective of the study is to gain greater insight into triggers of the disease and, in turn, help participants self-manage their asthma by avoiding areas where air quality could worsen symptoms.
The study and science are still ongoing, and we hope to publish more results in the future, but we wanted to share some of our preliminary findings on how engaging these apps are.
Before the launch of our Asthma app, we asked ourselves questions like: Would users be willing to step through the app’s entire e-consent flow? And would they continue to use the app after installation?
The good news is that the answer to each question is a resounding yes.- - -
The first significant unknown was at the very beginning of the participant's experience with a research study: consent. The consent process is obviously a critical component of any research study. It’s when participants learn what’s in store for them, and it's what sets their attitudes towards the study. In research studies, as in life, you get one chance to make a first impression. Transferring the paper consent process to a mobile one was a bold gamble— and we weren’t sure whether participants would even go all the way through the new e-consent process...
But the gamble paid off. Based on preliminary data for the Asthma Health app, over half of our users not only complete the e-consent process, they also come back the very next day to use the app. This is a very high rate of return for any app, let alone a health-related app.
Of course, consenting to a study and participating in one are two different things. Asthma Health needed to be as "sticky" as other popular apps in order to generate the required amount of study data. Excitingly, results have shown that users are as engaged (or more!) with Asthma Health as they are with games and social networks. Our working theory is that Asthma Health users are motivated by the goal of supporting research that helps the entire patient community. We plan to test this theory more extensively in the near future.
Significantly, our data shows that Asthma Health users respond to the same types of encouragement as users of other apps. In the chart below, we see the number of user sessions per day, with especially high activity on Mondays. Each weekly high point of user engagement (every Monday, as noted in the chart) corresponds to an automated reminder we pushed to users. The spikes in app usage are consistent with the type of behavior we see in mobile apps more broadly: users respond dramatically to regular reminders.
ResearchKit has already proven to be a revolutionary option for conducting scientific medical research, unbounded by the geographical constraints of traditional studies. The potential for future learning is enormous, especially as best practices are established.
- LifeMap Solutions
Apr 14, 2015
On March 9, Apple introduced the world to ResearchKit, a new software framework designed for health and medical research that helps doctors and scientists gather data more frequently and more accurately from participants using iPhone apps. Today, we’re excited to make ResearchKit available as open source so that researchers can more easily develop their own research apps and leverage the visual consent flows, real time dynamic active tasks and surveys that are included in the framework. Now that it’s available as open source, researchers all over the world will be able to contribute new modules to the framework, like activities around cognition or mood, and share them with the global research community to further advance what we know about disease.
This new blog will provide researchers and engineers all over the world with a simple forum for sharing tips on how to take advantage of the ResearchKit framework. It’s also a great place to check-in for news and updates from studies utilizing ResearchKit and hear from researchers directly about new studies and modules in development.
You can get started with ResearchKit by visiting GitHub, now available to all developers for free under the ResearchKit BSD license. Be sure to check out the Overview tab for great links to documentation and sample code to help you start building your research app today. You can even share your favorite modules with the global research community by creating an issue or claiming an existing one from our issue list and send us a pull request.
We can’t wait to see what we discover together!
- The ResearchKit Team