Verily’s Baseline: Homemade Wearable and Limited Data Sharing

Xconomy San Francisco — 

After three years of planning, a long-range health study from the biotech group at the former Google has begun.

Verily, the life sciences arm of Google’s parent company Alphabet, is funding what’s being called the “Baseline” study, which aims to collect extensive health data from at least 10,000 participants across four years.

The idea is to build a comprehensive database from a diverse set of Americans and gain insight into the biological changes underpinning the shift from health to illness. It could also point Verily and partners toward new products, although Verily officials made those products seem far down the road.

“We are taking a long range view in life sciences and healthcare and Baseline is a way to start to look for observations that could provide a reference set for a new set of tools and technologies,” said Verily chief medical officer Jessica Mega, in response to questions sent via e-mail.

Baseline is one of several health-related projects under Verily’s roof, which is now miles away from its corporate parent. Verily moved to South San Francisco last year into high-profile offices with room for 1,000 employees—reportedly the biggest biotech leasing deal in the Bay Area last year. It also has accepted outside investment, with $800 million pledged earlier this year by Temasek Holdings, which is affiliated with the Singapore government.

Baseline is no secret. Google started talking about it in 2014. It came under scrutiny last year when a Stat report revealed, among other things, that Verily CEO Andy Conrad owned the Southern California spa hosting the study’s pilot.

The idea germinated more than three years ago at Stanford University’s Canary Center for Early Cancer Detection, but after talks with Conrad and others (then running what was called Google X), the idea evolved into something “more ambitious” than simply a cancer study, said the Canary Center director and Stanford chair of radiology Sam Gambhir.

Gambhir is now one of several principal investigators overseeing Baseline. Stanford and Duke University are the first two main recruitment sites that will give participants full health workups of their genes, blood, saliva, urine, tears, and stool, with more daily data coming as they wear watch-like wrist devices (pictured) and have sleep sensors placed under their mattresses, said Gambhir. “We’d like to do even more, but it’s not practical or cost-effective at this point.”

He declined to say how much Verily is paying to run the study. “They don’t want me to comment, but I can say it’s very expensive,” he says.

Last year, Mega told Xconomy that her group might look for partners to share the costs for Baseline. That hasn’t happened yet, said Gambhir. Mega declined to discuss the costs of the study.

When asked if Verily plans to make money from the data it collects, Gambhir said an oversight committee will decide which outside researchers are allowed to mine the Baseline data. He did not rule out industry research conducting for-profit work, but he said there would be “many layers” of approval. Stanford, Duke, and Verily researchers will have two years of exclusivity on all data that enters the database.

Like its parent company’s projects to develop glasses-turned-camera and self-driving car, Verily has gained attention for an array of experimental devices, including contact lenses that collect data from tears. None have yet come to market. Verily will use only one of its homegrown devices in Baseline: a wrist device called the “Study Watch” that collects a volunteer’s blood pressure, heart rate, pulse rate, and electrocardiogram readings, says Gambhir. After years of journalists getting peeks and previews, Verily unveiled it last week in a blog post.

Baseline officials have decided to share data back with the participants, says Gambhir, only if it is certified by a lab: blood work and imaging, for example. Raw day-to-day data from the Study Watch will not be included, he said. If something highly abnormal crops up in a participant’s data, such as “massively low potassium” levels that could signal a cardiovascular problem, “we will work to notify them so they can see a doctor right away, although we can’t force them,” said Gambhir. (If the study crosses a line into medical advice, it would fall under a more complex set of regulations.)

Gambhir conceded that these days, with so many people comfortable having health-related information at their fingertips—or on their wrists—the Baseline edict of no sharing might turn off some would-be participants. (That might be countered with the modest stipend—amounting to a few hundred dollars over four years—Verily is offering for people to participate.) But with so much of the data expected to be in a “gray area,” Baseline wants to err on the side of caution, like in a clinical trial. “You don’t necessarily know the meaning of the data,” said Gambhir. “It could create all kinds of anxiety.”