Mobile technology and wearable sensors are broadening the limits of research and changing how IRBs view privacy.
The voluminous data can point to health strategies previously unimaginable.
“More data mean more solutions and better solutions for pressing health questions,” says Megan Doerr, MS, LGC, principal scientist, governance at Sage Bionetworks in Seattle. “The challenge comes in what are legal protections for those data and in what ways people consent for those data. Also, in what ways do those data implicate other people who have not consented?”
The challenge for IRBs and researchers is twofold. First, data collection can be huge and granular; secondly, there are multiple implications for privacy. “You name it, and it can be collected using a mobile device,” Doerr says. “Our phones have more sensors in them than the planes used to navigate the Pacific in World War II.”
Touchscreens, microphones, and other sensors are great at capturing continuous data, and they are improving.
Privacy Is a Challenge
Privacy is more difficult to protect in the era of big data. “I think the categorization of data as health data or not health data is no longer serving us,” Doerr says. “We’re understanding health data includes all different categories of data, like buying patterns at the grocery store, or walking patterns on mobile devices. I don’t know if HIPAA is fit for modern purpose.”
Informed consent should include an explanation about how data are used and risks that might be unrelated to the study. For example, researchers might have a plan to keep data private, but third parties also could be collecting the information. These include cellphone companies and internet companies that might collect everything on the device and use the information for marketing and other purposes.
Also, when more data are collected, identifiability is easier. IRBs should push researchers to justify the data elements they require for their study. “The researcher needs to be responsible and parsimonious in their approach to the data,” Doerr says.
Mobile technology makes it possible for researchers to study people’s daily experiences as they live them. “These changes are a potential boon for health research. If we don’t understand people’s lived experience in the world, we don’t truly understand their health,” Doerr says. “We’re seeing some recognition of that as communities start to recognize racism as a health concern. People are recognizing the multidimensionality to health.”
Researchers have always wanted to capture these data. Now, everyone has a cellphone that can collect data, or already is collecting, she notes.
Even more troubling from an IRB’s perspective is determining whether a proposed study with mobile technology might inadvertently capture data without consent. When IRBs review studies involving wearable and mobile technology, there will be potential risks they had not considered. For example, Doerr was involved in research of the app-mediated mPower Parkinson’s disease study, which was designed to enable participant-centered research. The app collected data from participants and asked them to perform tasks. One task involved their responding to a notification by holding the phone to their face and saying, “Ahhh,” holding the sound as long as they could, Doerr explains. (For more information, visit: https://parkinsonmpower.org/your-story.)
What the researchers did not anticipate is that people would think of the notification as an emergency prompt. They would stop whatever they were doing, including shopping, working, or other activities, and perform the task. This meant the data collection included background sounds and voices. If someone completed the task on the bus, then it could pick up other passenger’s words. Plus, people treated it as though it was urgent, which resulted it in interrupting important daily tasks, Doerr notes.
“It was through the unstructured data, where people could type in notes to us, that we learned people were saying, ‘It’s embarrassing when I get a notification during a work meeting and I had to step out to do this task,’” she explains. “They felt like it was urgent because it was on their phone, but they were not required to answer immediately.”
IRBs can anticipate these types of problems by inviting the study’s technology experts to speak about how the tool works and by asking more questions of researchers.
“IRBs can ask researchers about the kinds of instructions they are going to give them and the direction the researchers are giving,” Doerr says.
For instance, with the mPower app, investigators learned they should have told people to respond to the prompt when they had time and could complete the task in a quiet room, she says.
“It’s really essential when an IRB is considering a study that has mobility technology that the IRB call on all of the experts they’ll need to understand what is being collected,” she says. “The technologists understand these devices. But they’re a couple of steps removed from the IRB, with the researcher as the middle person.”
IRBs need to ask technologists and researchers who will collect data, how information will be transferred, how it will be stored, and how it will be used.
“It’s not like researchers want to do bad things and IRBs want to be unobservant or technologists want to create deviant technology,” Doerr says. “But they need to work together, talk, and ask questions.”
Researchers might not be experts on the technology they use. IRBs will need a technology expert there to help them conduct more homework on the details necessary for better informed consent.
“I would stress to IRBs that they can’t be afraid to say they don’t know something, and they can’t be afraid to drill down on researchers to make sure the IRB really understands what is going on with these data and how they’re being used,” Doerr says.
Examples of essential questions: How are data encrypted as they are collected? Are there unintended vulnerabilities?
“You might imagine that a massive trove of data might be a bigger target,” Doerr says.