- ResearchKit was hailed as an innovative mHealth tool when it debuted in 2015. But as one of the first studies to use the clinical trial platform begins to show results, competitors and mHealth analysts are taking shots at Apple’s shine, questioning whether the digital health platform has outgrown its usefulness.
ResearchKit was founded on the premise of enabling healthcare to collaborate with consumers in the digital health ecosystem, through their smartphone, laptop or Apple Watch. But some say the clinical research space needs more specific and reliable digital data.
“ResearchKit was great for providing usability and feasibility studies,” Ryan Rossier, vice president of platform studies for market research firm Medullan, says. “That’s great for ‘feet-in-the-water’ studies, but the paradigm in healthcare is shifting. You’re going to want to do more than what [ResearchKit] offers.”
“When you get down to it, they’re libraries, not platforms,” he adds. “You use [ResearchKit] to create the framework and get consent and enrollment down. But once you get into the specifics of a study, you need to get creative. You need to build custom code. … That takes a lot more time and effort than what you can find on ResearchKit.”
A crowded market
That’s where companies like ePatientFinder, ProofPilot, KlickHealth and Medidata are coming in. They’re marketing themselves as the next wave of mHealth clinical trial platforms, offering providers access to new sources of consumer and medical grade data and the tools to create custom apps that drill down a lot feeper.
ResearchKit offers “really simple observational tools [and] really basic data,” says Tom Dorsett, CEO of ePatientFinder, “but the reality is it hasn’t gone very far.” It’s not deep enough for interventional trials, he says - which health systems and the pharma industry are looking for.
Dorsett says clinical trial platforms like ePatientFinder drill down into EHR data to find the right patients for the right trials, rather than collecting and sifting through vast amounts of consumer data to find the matches. This type of platform requires more of a collaboration with the healthcare provider, he says, and it produces more reliable clinical data.
ProofPilot, launched out of New York last year, offers a different approach. The company markets its research platform on a software-as-a-service basis, and argues that healthcare researchers can launch their own targeted studies if given the right tools and templates.
“We made an assumption that researchers knew what study they wanted to run, and based on our [previous] experience, that these researchers were pretty good at using technology,” CEO Matthew Amsden told Mobihealthnews in a May 2016 interview. “The reality is they weren’t. These researchers didn’t know how to create the research study they had in their head and create a study that was going to be automated and use technology efficiently. And they had a hard time. … It was really problematic and we felt like we were a consulting firm that masqueraded as a tech firm.”
Medidata, meanwhile, touts a medical device data system (MDDS) platform powered by mHealth integrator Validic, capable of gathering data from hundreds of devices and apps and in what’s called the Medidata Clinical Cloud.
“[W]e recognize that the rapidly expanding ecosystem of mobile health technologies requires openness and interoperability,” Glen de Vries, Medidata’s president, said in a July 2015 press release announcing the Validic partnership. He called the deal “a big step toward realizing the potential of mobile health in clinical research because it offers life sciences organizations the flexibility to select the mHealth tools that provide the most clinically meaningful information for specific patient population.”
ResearchStack joins the fray
ResearchKit is also facing competition from the Android universe. Earlier this year a team of researchers at Cornell Tech launched ResearchStack, the Android equivalent to Apple’s IOS-based platform. Within a few months the first app – MoleMapper – appeared in the Google Play store.
Some proponents argue that ResearchStack offers even better opportunities for clinical trials because roughly 70 percent of the world’s smartphone users have Android devices, and underserved and remote ppopulations – who would best benefit from these studies – are more apt to own inexpensive Android-based smartphones than costly iPhones.
To its credit, ResearchStack’s developers say they’re aiming to create a platform where Apple-based and Android-based clinical studies can work together.
“(A)n overriding goal of ResearchStack is to help developers and researchers with existing apps on iOS more easily adapt those apps for Android,” officials from the non-profit Open mHealth, who helped develop the tool, said in a 2015 blog. “Though the correspondence of features between the two SDKs isn’t one-to-one, the two SDKs will offer enough shared functionality and a common framework and naming scheme to greatly speed up adaptation of ResearchKit apps to Android (and ResearchStack apps to iOS) and the procedural aspects of running a study on a new platform (such as IRB approval and secure connectivity with a data collection backend).”
Dorsett and Rossier say ResearchKit rode a wave of popularity caused in part by the Apple name and in part by the high-profile health systems that signed on to launch studies. They see the platform as a “starter kit” of sorts for health systems looking to make a big splash in population health and public health. But they wonder if those studies will lead to meaningful clinical outcomes.
“It’s still somewhat of an experiment,” says Dorsett, who wonders if healthcare providers are aware of the complexities of an interventional study and think observational studies will accomplish the same goals.
“My fear is that there’s going to be actual impact on the well-being of people,” says Rossier, noting the difference between data fed into a research study by a consumer and clinically validated information gathered from devices. “There are risks involved in using consumer information in clinical trials.”
MyHeart Counts shows results
While Rossier says researchers “are still playing the waiting game” with ResearchKit, at least one study on the platform is starting to show results.
A report published in December in JAMA Cardiology found that the Stanford University School of Medicine’s MyHeart Counts app, one of the first studies released on ResearchKit in 2015, “demonstrates that a mobile app can be used to consent and engage a large population across the U.S. in a clinical study.”
“The ultimate goals of the MyHeart Counts study are to provide real-world evidence of both the physical activity patterns most beneficial to people and the most effective behavioral motivation approaches to promote healthy activity,” said Michael McConnell, MD, a professor of cardiovascular medicine at Stanford and co-lead author of the study, in a Stanford press release.
“Traditional research on physical activity and cardiovascular health has been based on people writing down what they remembered doing,” he added. “Mobile devices let us measure more directly people’s activity patterns throughout the day.”
More recently, Stanford announced that it was applying AI and analytics tools in a new iteration of the MyHeart Counts app to apply precision medicine to the study.
“[This] lets us begin to customize feedback to users, and also discover which types of information might be most useful or motivating for different groups,” Abby King, PhD, a professor of medicine and health research and policy at Stanford who designed the coaching module, said in a press release.
“The most unique thing about the new version is its ability to randomize patients and intervene,” added Euan Ashley, DPhil, MRCP, an associate professor of cardiovascular medicine and genetics at Stanford and the principal investigator for the MyHeart Counts study. Those interventions, he said, “will give researchers a handle on whether a particular intervention for an individual is prompting a change in behavior for the better.”