Apple ResearchKit: Can patient-generated data be trusted?

From the mHealthNews archive
By Eric Wicklund
08:11 am

Apple's ResearchKit platform promises to change the way clinical trials are conducted, giving medical researchers access to patient-generated data on a global scale.

Whether that will actually work remains to be seen.

After it was tucked into Apple's recent press conference to unveil the Apple Watch, ResearchKit was hailed as an open-source platform upon which apps could be developed that allow consumers to enter their data and researchers to collect biometrics in real-world settings, outside the constraints of a lab or clinic or some such controlled environment. This would, in essence, open up medical research to everyone with an iPhone.

"This is an inflection point for medical research," said Corey Bridges. CEO of LifeMap Solutions, which helped develop the Asthma Help app – one of five already available for ResearchKit  and is engaging in a study with Mount Sinai Health's Icahn School of Medicine. "This will lead to larger and much more interesting studies."

Bridges said ResearchKit eliminates the "brick and mortar" issues that have plagued clinical trials in the past – namely, that they're restricted to subjects who can physically visit the testing site for interviews. Now, armed with e-consent forms and following Institutional Review Board standards, medical researchers can fashion a study that draws data from any number of participants and environments.

Others, however, questioned whether the platform will enable researchers to conduct better tests or just inundate them with questionable data.

"I was able to sign up as a diabetes patient, someone who has Parkinson's disease and someone who has asthma on the ResearchKit apps, yet I have none of those conditions," Partners HealthCare's Kamal Jethwani recently told Politico. "How will researchers separate noise from signal? And even if they can do that, the overhead involved in cleaning the data may just not be worth it. I think more needs to be built into the process to keep only qualified patients in it."

“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,” Lisa Schwartz, a professor at the Dartmouth Institute for Health Policy and Clinical Practice, told Bloomberg. “Bias times a million is still bias.”

There's no doubting the news struck a chord among the healthcare community. While more than 2,500 have signed up so far for the Asthma Help study, Stanford University reported that 11,000 people had signed up for a cardiovascular study less than 24 hours after ResearchKit was unveiled.

“To get 10,000 people enrolled in a medical study normally, it would take a year and 50 medical centers around the country,” Alan Yeung, medical director of Stanford Cardiovascular Health, told Bloomberg. “That’s the power of the phone.”

Bridges, meanwhile, says LifeMap will be working with Mount Sinai on other clinical care apps through the ResearchKit platform.

"This is one of the biggest advancements in how medical science is done," he said. "It's there to reduce difficulties. It's there to reduce challenges."

"It's observational data," Patricia Ganz of the UCLA Fielding School of Public Health told Politico. She's working with the Share the Journey app on a medical study that aims to understand the symptoms of breast cancer treatment. "By having large numbers of respondents, we could generate new insights into the experience of the patient population."

Apple officials say they're working to make sure privacy and security concerns are addressed. In guidelines posted on the App Store, the company said ResearchKit applications will be rejected "if they don't provide a privacy policy, store user information in iCloud and provide diagnosis or treatment advice. App developers must also agree not to share data with third parties or sell data."

"Apps conducting health-related human subject research must obtain consent from participants or, in the case of minors, their parent or guardian," the guidelines state. "Such consent must include the (a) nature, purpose and duration of the research; (b) procedures, risks and benefits to the participant; (c) information about confidentiality and handling of data (including any sharing with third parties); (d) a point of contact for participant questions; and (e) the withdrawal process."

In addition, data collected from those first five apps won't be used commercially, nor will those apps be able to pull any personal data from a user's smartphone, according to the company handling that data.

“We are as careful as we can be about keeping data as confidential as possible,” said Christine Suver, principal scientist and head of open data governance at Sage Bionetworks, in an interview with ITworld. Sage, a nonprofit biomedical research organization based in Seattle, will be gathering, de-identifying and storing all health data gathered from those apps.

Sage told ITWorld it would handle data from the five apps for the "foreseeable future," but since ResearchKit is an open-source platform, the potential exists for different data management capabilities down the road. There's no requirement that future apps use Sage, according to the article, and Apple hasn't endorsed the organization. 

“Sage would be happy to discuss how our services could be used by other groups to build mobile study apps,” Suver said.