Apple’s ResearchKit: Game Changer or Gimmick?
While new details about the Watch may have been the most anticipated news to come out of Apple’s recent “Spring Forward” media event, one of the biggest rounds of applause from attendees came after an announcement that was not directly related to the upcoming wearable.
In a surprise announcement, Apple introduced ResearchKit, an open source software framework specifically designed for creating medical research apps for the iPhone. “With hundreds of millions of iPhones in use around the world, we saw an opportunity for Apple to have an even greater impact by empowering people to participate in and contribute to medical research,” said Apple SVP of operations Jeff Williams. “ResearchKit gives the scientific community access to a diverse, global population and more ways to collect data than ever before.”
At least one of the lofty ResearchKit goals expressed by Williams already appears to be coming true. A cardiovascular disease research app that Apple developed with Stanford University acquired 11,000 participants overnight after the study was highlighted during the iPhone maker’s media event, reported Bloomberg. “To get 10,000 people enrolled in a medical study normally, it would take a year and 50 medical centers around the country,” Stanford Cardiovascular Health medical director Alan Yeung told Bloomberg. “That’s the power of the phone.”
As CEO Tim Cook noted during the “Spring Forward” event, Apple has sold 700 million iPhones to date, so there’s no question that ResearchKit has the potential to provide researchers with an enormous group of study participants. On the other hand, there is some debate in health research circles about the quality of the data that can be derived from iPhone users. So does Apple’s ResearchKit represent a revolution in how medical research is conducted or is it just another marketing ploy to sell more iPhones? Here’s a rundown of the potential benefits and pitfalls associated with Apple’s newest software framework.
What Apple says
As noted by Williams, developers can utilize many of the hardware components found in the iPhone to create a wide variety of medical apps to conduct studies on conditions such as asthma, breast cancer, cardiovascular disease, diabetes, and Parkinson’s disease. For example, the mPower app developed by Sage Bionetworks tracks the symptoms of Parkinson’s disease patients by having users perform a variety of tests on the iPhone, including a finger tapping exercise on the device’s touchscreen.
While the testing methods employed by these apps are essentially the same ones that doctors already use in conventional medical studies, Apple argued that the use of the iPhone as a medical research tool provides several additional advantages. Since the iPhone can collect data from participants without having them appear in person at a research facility, Apple noted that it will be easier to recruit more participants for large-scale studies. Apple claimed that this larger pool of participants will also provide researchers with a broader cross-section of the population.
Additionally, since most people take their iPhones with them wherever they go, medical study apps that collect data throughout the day will be able to get a more comprehensive and reliable picture than traditional studies that rely on self-reporting or observable behaviors during brief doctors’ visits. Finally, Apple noted that the built-in consent forms provided by the apps will allow researchers to spend less time on paperwork and more time on actual research.
What some medical researchers say
Although ResearchKit will give researchers access to an incredibly large group of potential participants, some health experts noted that having more participants does not necessarily lead to better quality research. “Just collecting lots of information about people — who may or may not have a particular disease, and may or may not represent the typical patient — could just add noise and distraction,” Dartmouth Institute for Health Policy and Clinical Practice professor Lisa Schwartz told Bloomberg. “Bias times a million is still bias.”
Additionally, as multiple studies have revealed, there are significant demographic differences between the typical iPhone user and the typical Android phone user. A 2013 survey conducted by the Pew Internet & American Life Project found that iPhone users tend to be wealthier and more highly educated than Android phone users. Apple also had a considerable lead over Android when it came to older users. All of these discrepancies could potentially distort the information gathered through medical study apps on the iPhone. So while it’s true that ResearchKit could allow researchers to more easily tap into a larger group of participants, those participants may not necessarily be representative of the population as a whole.
It should be noted that although Apple announced that ResearchKit would be “open source,” it is still unclear how the iOS-based tool could be implemented in Android. Commentators such as Fortune’s Dan Primack and MobiHealthNews’ Jonah Comstock have noted that ResearchKit is likely to present multiple problems for Android developers since it is so closely integrated with other iOS-based programs. In any case, ResearchKit will remain an exclusive iOS tool until at least next month when the API is publicly released.
Pocket dialing and other problems
Other data accuracy problems are directly related to the use of a mobile device as a data collection tool. Just like mobile phone users have been known to inadvertently make phone calls, it is quite likely that there will be a significant number of medical study participants who will skew results by unintentionally activating an app. Inaccurate data could also be collected if an iPhone is shared between multiple users, or if a child accidentally accesses a medical study app and plays it like a game.
There is also a risk that unforeseen programming issues with the apps will distort the information collected. Apple had to temporarily disable the blood glucose feature in its native Health app after it was discovered that it was incorrectly displaying glucose values for users in countries with alternate measurement systems. While this particular issue shouldn’t affect ResearchKit since it is currently only available for users in the U.S., this incident still raises questions about the reliability of medical data gathered through mobile apps.
Besides potential issues with the accuracy of the data collected by the iPhone, some medical researchers have raised questions about the ethics of conducting medical research via mobile devices. While Apple touted the reduction of paperwork as a benefit of ResearchKit, some of the app developers admitted that the informed consent process for medical study apps could be improved. “That interaction is pretty crappy,” Sage Bionetworks CCO John Wilbanks told The Verge. As noted by Wilbanks, participants in traditional medical studies usually have the opportunity to meet with a researcher face-to-face in order to have all their questions answered, something that is not possible with consent forms in apps.
On the other hand, there are some ways in which mobile app-enabled medical research could be superior to traditional studies. As noted by Stanford’s Yeung, one problem with traditional medical studies is the overreliance on self-reporting. “People don’t want to say they did zero exercise — they want to say they did something,” Yeung told Bloomberg. “They don’t really tell us the truth.” Since a mobile device would constantly collect activity data on a user without relying on their willingness to tell the truth, medical studies conducted through the iPhone could actually be more accurate than traditional studies in this respect.
The unique capabilities of the iPhone’s hardware could also enable studies that could not be conducted using traditional methods. Apple’s Williams described an asthma study planned by the Icahn School of Medicine at Mount Sinai in collaboration with Weill Cornell Medical College that relies on crowdsourced research performed by iPhone owners. The asthma study will provide participants in New York with spirometers and Bluetooth inhalers in order to document areas in the city where they are most susceptible to asthma attacks. Using the GPS data from participants’ iPhones, the researchers will then compare the areas where asthma triggers were found to a “pathogen map” created by taking surface swabs at various locations around New York. By taking advantage of the iPhone’s GPS hardware and Apple’s large user base, researchers are able to conduct a type of study that would not be possible using traditional methods.
Apple’s ResearchKit also has the benefit of being tied into Apple’s HealthKit health data platform. If a user gives permission, the information collected through a ResearchKit medical study could be shared with their health care provider in order to offer a more comprehensive portrait of their overall health. So not only would the information provided by a participant in a ResearchKit-enabled study help overall research on a larger medical issue, the information they provide could also be immediately beneficial to them personally.
ResearchKit is a mixed bag
Overall, ResearchKit appears to be a medical research tool that has both good and bad qualities. While there may be some issues with the accuracy of the data collected through mobile apps, there are also some impressive advantages afforded by the iPhone’s hardware and the size of Apple’s user base. Although Apple presented ResearchKit as a purely altruistic initiative, it’s clear that some of the data accuracy issues with the tool are directly related its initial exclusive availability on the iOS platform.
In this sense, ResearchKit is another important selling point for the iPhone and not completely unrelated to business as CEO Cook portrayed it during Apple’s annual shareholder meeting on March 10. “[F]or those looking for an ROI [Return on Investment], there’s not one,” said Cook, according to Apple Insider.
However, it seems clear that the company intended ResearchKit to be primarily used as an iOS-based program. So while ResearchKit may not directly benefit Apple with a return on investment, the only device currently compatible with ResearchKit apps is made exclusively by Apple. Still, even with its limited availability and some possible data accuracy issues, ResearchKit has the potential to become an important medical research tool, if not a game changer.
Follow Nathanael on Twitter @ArnoldEtan_WSCS