Better datasets and tech will benefit patients


screen shot 2015 05 21 at 3 41 09 pm
Photo credit: 
Saul Bromberger


The explosion of big data is transforming the way those in health care are diagnosing, treating and preventing disease, panelists at the Big Data in Biomedicine said on its opening day.

During a five-member panel on population health, experts outlined work that is currently being done but said even bigger datasets and better technology are needed to ramp up the benefits from digital data and to save lives.

"Using the end-of-millions to inform care for the end-of-one – that is exactly where we're going," said Tracy Lieu, MD, MPH, director of research at Kaiser Permanente Northern California, a health-care network that includes 21 hospitals, 8,000 physicians and 3.6 million patients. "And we think that in a population like ours, in an integrated system like ours, we are in an ideal setting to do personalized medicine."

Stanford Medicine professor Douglas Owens, MD, director of the Center for Health Policy and Center for Primary and Outcomes Research, led the panel on Wednesday. He said big data is also changing how research is being conducted.

"There's been an explosion of data of all kinds: clinical data, genomics data, data about what we do and how we live," said Owens. "And the question is how can we best use that data to improve the health of the individual and to improve the health of populations."



Lieu said two key trends are central to medical researchers: informatics and genomics. She said Kaiser utilizes a "virtual data warehouse" with the digital data of 14 million patients dating back to 1960. But Lieu cautioned that the data are not always the means to an end, particularly if the findings are not tested and implemented.

"Sometimes we fail. And we fail when we identify a problem of interest, we make a decision to study it, we assemble the data, we analyze and interpret the results - and then we send them off to journals. So we fail to close the loop," she said, because researchers typically don't go beyond the publication of data.

She said Kaiser is now focused on trying to close that loop. "To do that, we need the kinds of tools that you in this group and the speakers at this conference are developing," she explained. "We need better and better technology for rapidly analyzing and aggregating data."

Joining Lieu on the panel were Mark Cullen, MD and professor of medicine at Stanford; David Atkins, MD, MPH and director of health services research and development at the U.S. Department of Veterans Affairs; David Maron, MD and clinical professor at Stanford and Ruth Travis, DPhil and associate professor at Oxford University.

Maron presented the Baseline Study, a clinical research collaboration between Google Life Sciences, Stanford and Duke University.

The baseline study was launched last year with an initial 175 healthy volunteers in the pilot program. Eventually, 10,000 thousands patients are expected to undergo physical exams and provide samples of blood, saliva and other body fluids that can be examined using new molecular testing tools, such as genome sequencing, to look for risk factors or the presence of cancer or cardiovascular disease.

Maron called the baseline study "the Framingham Study of the 21st century," referring to the landmark heart study under the direction of the National Heart Institute that began in 1948 and ushered in a new era of heart health.

He believes the massive amount of data generated by the new study will eventually create biomedical information that will enable scientists, clinicians and policymakers to work with data scientists on revolutionary medical tools and treatments.

Each participant in the study would eventually generate 4 terabytes of data, he said.

“It's going to be a lot of data and we're going to leverage the analytic capacity of our partner, Google, to analyze this data and make some sense out of it so that we have a better understanding of human health and transition to disease," he said.

Kathy Pham, a software engineer and a member of the Digital Service at the White House, asked the panel about efforts to open up the data to patients who want to know more about their own diagnoses and have a greater say in their care.

Lieu said that Kaiser was partnering with the Silicon Valley start-up, Smart Patients, an online community for patients and their families to learn from one another about scientific developments related to their conditions.

“Patients and their caregivers are starting to make these requests to get connected, can we actually have access to the knowledge and wisdom of not only other patients, but maybe even downstream, the health care systems.”

Atkins said that the Veterans Health Administration is relying more and more on big data to help with medical predictions and to better understand clinical variations.

The VA is the United States’ largest integrated health-care system with more than 1,700 hospitals and clinics serving some 8.75 million veterans each year. It pulls from more than 120 data systems around the country.

“Big data is great at improving predictions and if you’re providing care, what you really want to think about is: predicting for what?” he said.

Atkins emphasized: How you use data is as important as what data you use.

“Health-care systems are complex, social organizations,” he said. “We need to apply the data carefully and conservatively to make sure that we do good and avoid some unintended harm.”