Medicaid’s Data Gets an Internet-Era Makeover

Jini Kim, a Nuna founder, at work with her autistic brother, Kimong, a Medicaid recipient whom she cares for with their parents. Credit Jason Henry for The New York Times

New York Times

By Steve Lohr

Jini Kim’s relationship with Medicaid is business and personal.

Her San Francisco start-up, Nuna, while working with the federal government, has built a cloud-computing database of the nation’s 74 million Medicaid patients and their treatment.

Medicaid, which provides health care to low-income people, is administered state by state. Extracting, cleaning and curating the information from so many disparate and dated computer systems was an extraordinary achievement, health and technology specialists say. This new collection of data could inform the coming debate on Medicaid spending.

Andrew M. Slavitt, acting director of the Centers for Medicare and Medicaid Services, described the cloud database as “near historic.” Largely because Medicaid information resides in so many state-level computing silos, Mr. Slavitt explained, “we’ve never had a systemwide view across the program.”

This week, for the first time, Nuna’s executives are talking about the company’s funders, business strategy and work for Medicaid, starting on Monday evening at the J. P. Morgan Healthcare Conference in San Francisco.

Health data on its own — billing, diagnostic and treatment information, typically recorded in arcane, shorthand codes — is not very useful. But if it can be aggregated and analyzed economically and quickly, that data is seen as a vital ingredient in transforming health care.

The health care marketplace in the traditional fee-for-service model values volume. More doctor visits, hospital stays, operations and pills mean more revenue and profit for health care providers. But the push in recent years has been toward what is known as value-based health care. In the value model, medical groups are paid for outcomes: patients treated more efficiently and people who are healthier.

But that transition only becomes possible with accurate, reliable data as the raw material for measuring outcomes and discovering what works and what does not. That is the reason for the excitement about and investment in health information technology, and the goal of the Medicaid work with Nuna. The data set will be stripped of identifying information before it is released to researchers.

Ms. Kim, 35, a former Google product manager, has firsthand experience with Medicaid. Her brother, Kimong, who is a year younger, is severely autistic. When he was 8, Kimong started having monthly grand mal seizures, which are characterized by loss of consciousness and violent muscle contractions. The ambulance, hospital, doctor and therapy bills piled up. As a 9-year-old, Ms. Kim helped her Korean immigrant parents complete the Medicaid application forms.

“Our family would have gone bankrupt without Medicaid,” she said. “It saved us.”

The Medicaid system covers millions of working families, older people, children and people with disabilities. In fact, 40 percent of Medicaid spending goes to the disabled. Half of long-term care in America, mainly for older people, is through Medicaid. And nearly half of the children born in the United States are in the Medicaid system.

The trouble with traditional health data warehouses, specialists say, is that they resemble digital vaults. It is difficult and time-consuming to get information in or out, and only people with specialized skills can use them.

But the new cloud-based technology, using internet-era software, is flexible and interactive. It opens the door to real-time monitoring of emerging disease clusters, billing patterns and program effects. For example, did the percentage of low birth-weight babies decline after a Medicaid program was put in place? If so, how much?

“This kind of data can help move health care policy from a partisan ideological debate to one informed by knowing who the people affected are and what will likely happen to Medicaid recipients,” said Drew Altman, president of the Henry J. Kaiser Family Foundation, a nonprofit health policy research organization.

Though young, Ms. Kim has spent years learning hard-earned lessons about health care. At Google, she was a product manager on Google Health, a failed effort to attract millions of people to use its free, online personal health records.

“Health care is hard, and humility is important,” Ms. Kim said. “You can’t just put technology on something and assume it’s going to work. You really have to understand the ecosystem in health care.”

In late 2013, Ms. Kim got a call from Washington and became one of the small cadre of Silicon Valley technology specialists called on to fix HealthCare.gov, the application website for health insurance under the Affordable Care Act. That was six months of 18-hour days, seven days a week through Thanksgiving, Christmas, New Year’s and beyond, until the end of the spring 2014 enrollment period.

That government experience, Ms. Kim said, “paved the way” for Nuna’s work with Medicaid data. It demonstrated, she said, that internet technology and software methods could be applied to government health data from the mainframe era.

Her brother is still on Medicaid, and caring for him is a family affair. Ms. Kim lives with her parents and Kimong in Hercules, Calif., in the Bay Area, about 10 miles north of Berkeley. When excited, Kimong waves his arms. When frustrated, he bangs his head against a wall. He can be loud and disruptive when he is anxious, and the family has been escorted off a few plane flights. “He’s a challenge,” Ms. Kim said.

To relieve her parents, Ms. Kim takes Kimong with her to work some days. He can only say a few words, and one of them became the start-up’s name — “Nuna,” or “big sister” in Korean.

Nuna, founded in 2010, had only one full-time employee, Ms. Kim, until 2014, when it got an early round of venture financing. And no one was paid until then. David Chen, the company’s co-founder and chief data officer, holds a Ph.D. in bioinformatics from Stanford University, but he also was a data scientist at Netflix for three years while Nuna was getting off the ground.

Today, Nuna has 110 employees. It has raised $90 million in venture capital led by Kleiner Perkins Caufield & Byers and prominent individual investors, including John Doerr, Kleiner’s chairman, and Joi Ito, director of the MIT Media Lab.

Nuna is applying the technology and lessons learned from its Medicaid work to pursue business from private health plans, especially at large companies, to help them improve the efficiency and quality of care for workers.

Nuna’s technology achievement, specialists say, is impressive. Government research agencies and big health providers have put some data on Amazon Web Services, the leading cloud-computing supplier, which hosts the Medicaid data set. “But no one has tried to do something as audacious as Nuna has done,” said Andy Jassy, chief executive of A.W.S.

Ms. Kim expressed pride in her start-up’s work, but also a sense of personal satisfaction. “This is my love letter to Medicaid for saving my family,” she said.

A version of this article appears in print on January 10, 2017, on Page B3 of the New York edition with the headline: Start-Up Gathers Far-Flung Medicaid Information in Cloud Database.

Image: Jini Kim, a Nuna founder, at work with her autistic brother, Kimong, a Medicaid recipient whom she cares for with their parents. Credit: Jason Henry for The New York Times