91av

THERE is life in data, if you know how where to look: the number of steps from your apartment to the coffee shop, the cost of a latte, the levels of pollution in the October air you’re breathing in. As we move through the world, we leave thousands of other signals behind us that speak volumes about our health and well-being. Now, a new project in New York City will gather and process those signals on an unprecedented scale.

The ambitious scheme, named the Kavli HUMAN Project, will track every last scrap of data generated by 10,000 New Yorkers. Taken together, that data will weave a detailed narrative of each person’s life – where they go, what they do, how they’re feeling, how they walk the line between sickness and health.

The idea is to dive deep into everyday situations, says Dennis Ausiello, director of the Center for Assessment Technology and Continuous Health in Boston, who is advising the HUMAN team. The medical community is good at diagnosing disease, but struggles to understand what it means for people to be in good health. When it comes to our health, he says, it doesn’t make sense to wait until something is wrong to start paying attention.

Instead, we can learn what good health looks like and how to maintain it by tuning in to the fluctuating rhythms of life – diet, exercise, sleep, medication. “Medicine works by you setting a date on the calendar for a doctor’s visit,” Ausiello says. “Real life works 24/7.”

HUMAN plans to start recruiting in early 2017. A randomly selected group of New Yorkers will be asked to hand over their personal data for the next 20 years. Among the long list of things the scientists hope to track, outlined in a paper published earlier this month (Big Data, ) are: electronic medical records; credit card transactions and banking documents; school and employment history. Then there are emergency calls, arrests or other interactions with law enforcement, as well as location, as captured by a smartphone’s GPS tracker.

“A randomly selected group of New Yorkers will hand over their personal data for the next 20 years”

Beyond that, there’s a wealth of environmental data to consider, like city records on the building the person lives in, demographic details about the neighbourhood, and environmental measurements such as air quality and noise levels.

And that’s just the stuff that’s already out there. The researchers plan for participants to undergo an extensive physical examination, including microbiome samples; genetic tests; and interviews with psychological professionals. Mental status will be regularly assessed through voluntary surveys, pushed to a smartphone. Participants may also be asked to don a wearable device, like a Fitbit, to capture their physical activity, or a silicone wristband to track their chemical exposure. Bluetooth sensors placed around the home could quietly collect information on activity and family interactions.

What will this truckload of data turn up? It’s hard to say, says Steve Koonin, director of the Center for Urban Science and Progress at New York University. That is what makes it so exciting.

“What did Galileo think he was going to see when he turned his telescope on the heavens? He didn’t know,” he says. “This notion of discovery is very powerful for me.”

Studies that take the long view are not new. For example, the , running since 1948, has discovered many risk factors for cardiovascular disease by following three generations of adults in Massachusetts. In the UK, the on Ageing has followed the changing health, lifestyle and economic well-being of older adults since 2002. But while these studies check up on people once every few months using traditional doctor’s tools like CT scans, HUMAN will be able to track every minute or second, polling data from the genome to the commute. Smartphones – now in almost every pocket in the rich world – are a driving force behind this new way to study human health. They act as both sensors and filing cabinets, allowing physicians to gather real-life health data.

One slow-moving mystery HUMAN hopes to tackle is obesity, a growing issue in the US. Without detailed information, it’s difficult to unpick the mesh of biological, psychological, and economic factors that might push us to eat too much cheap, processed food that’s high in fat and sugar. Credit card records and location data could help establish what kind of food a participant eats, while information about their genetic make-up or demographic surroundings might suggest additional pressures.

The project will also study cognitive decline. Sixteen million people in the US suffer from some form of cognitive impairment, such as Alzheimer’s disease or dementia, according to the Centers for Disease Control. Studies provide some insight into who might be at risk – those who retire later in life seem less likely to develop dementia, for example – but there are probably other behavioural or environmental factors that have yet to be sussed out.

“We’ve come to understand that it’s an entire life history that gives rise to the various wounds and bruises that make for all types of mental decline,” says the director of HUMAN’s scientific agenda, Andrew Caplin, “but how this is unfolding, we really don’t know.”

The HUMAN project is a massive undertaking – and one that may be tricky to get right. As the design of the study is being finalised, a panel is exploring how best to protect everyone’s data. One crucial step, says Koonin, will be to ensure that participants understand what it is that they’re signing up for. Encrypting the data and restricting researchers from accessing the entire set at once could also help maintain privacy.

But with such detailed data sets it may be difficult or even impossible to maintain files that can’t be traced back to individuals. A study published in January of this year, for example, found that people could be identified in 1.1 million “anonymised” credit card records using just four clues to what they’d purchased or where they’d been. Another experiment from the same lab did a similar trick with cellphone location records, recognising about 95 per cent of users.

Samuel Volchenboum, director of the University of Chicago’s Center for Research Informatics, sees the potential benefits of the project, but worries that it may get tripped up by data collection long before it’s time to crunch the numbers. Most studies, he says, still struggle to combine disparate and mismatched data sources, or to get patients to keep up with a single new device or app.

Caplin compares this new way of studying health to the famous documentary series Up. The films followed the lives of a group of British children, checking in with them every seven years.

“You watch that and say, how on earth can people change this way?” says Caplin. “Every one of us goes through a story that, if you had been tracking it in its richness, would have had its own logic.” HUMAN will attempt to capture that richness, logging snapshots almost continuously of everything from DNA to travel patterns, to understand how and why humans change.

(Image: Matthew Pillsbury, courtesy Benrubi Gallery)

This article appeared in print under the headline “Tracking the health of 10,000 New Yorkers”