A glimpse of Sanjay Basu’s work begs the question, “What doesn’t he do?”
From testing a San Francisco nutrition program, to breaking down some of the world’s largest data sets, to inspiring a national diabetes and obesity reduction program in India, Basu is a true interdisciplinary scholar.
An assistant professor of medicine and the newest core faculty member at Stanford Health Policy, Basu focuses on answering health care’s big questions about how to provide care that people can use.
Though he started his Stanford career at the Stanford Prevention Research Center, he has been drawn toward Stanford Health Policy over the past few years.
“Rather than always burning a hole in Eran’s rug, I figured it would make sense to make a physical move,” said Basu.
Eran Bendavid, an assistant professor of medicine and core faculty member at Stanford Health Policy, called Basu the “consummate collaborator.”
Brought together by malaria subsidies, Bendavid and Basu have worked on an array of projects. Bendavid particularly appreciates Basu’s openness to working with others and willingness to discuss new and different ideas.
“He's a delightful addition to the family,” said Bendavid. “We will benefit from his presence, and as this family does, reciprocate in kind and in kindness.”
“Sanjay’s outstanding research program is a great fit,” said Douglas Owens, a professor of medicine at Stanford Health Policy. “We also look forward to Sanjay’s participation in our teaching programs — our trainees will greatly benefit from Sanjay’s expertise.”
Better Health for Communities
Basu takes a global approach, tackling health-care questions that could improve outcomes for people around the world. But he also recognizes the value of customizing programs with community in mind.
Working with the city of San Francisco, Basu is testing a government-funded food voucher program to see what payment system works best to keep people healthy. Low-income residents in four San Francisco communities receive vouchers. Everyone gets the same amount, but some receive the full subsidy at the beginning of the month while others have it doled out weekly.
“There’s a lot of theory that even if you have the same amount of money but distribute it differently, people might be able to smooth their consumption over the month and eat more fruits and vegetables instead of processed foods,” said Basu.
Basu’s team will investigate whether spreading out funding keeps participants from running out of nutritious food toward the end of the month. This could help people with diabetes — who have high blood sugars at the beginning of the month and low sugars near the end of the month — avoid visits to the emergency department.
“On the other hand, a reasonable alternative theory is that it would just make shopping a pain for everybody,” said Basu. “So we’re doing the first trial that compares those.”
Big Data for Single Problems
Working with national data sets, Basu’s team is planning several projects that would use large data sets to create tools for physicians to customize care. Using deep learning — a strategy that creates “miniature-brains” with logical reasoning skills — the team will develop virtual doctors to review millions of charts. The digital doctors will identify under-treated or improperly treated patients so that a human doctor can review their cases.
But first the team has to figure out how to analyze the data. These data sets are too big for traditional methods, so Basu’s team is working on new ways to process them.
“Actually the best ways we’ve seen are to use chips and processors normally used to play video games,” said Basu. “They’re much more robust at running very large-scale models on these really massive data sets.”
“It’s a fun and interesting little challenge.”
The goal is to use the data to improve access to and quality of primary care.
“Solano County is giving us all of their information from every clinic, every qualified health center, every pharmacy, every lab for the whole county that covers 95 percent of their residents,” said Basu. “It’s the first time I know of in the U.S. that we can get an all-capture data sets.”
Having an extensive data set for a single area will help Basu’s team determine what areas of health most need improvement and how to reach people who are falling through the cracks. Using big data, this model can then be applied to larger areas of the United States and perhaps the world.
Applying Research to Policy
For Basu, watching recommendations turn into actual policy is most rewarding. A recent study looking at ways to reduce diabetes in India was particularly satisfying, he said.
Funded by the Rosenkranz Prize, the study examined whether taxing sugar-sweetened beverages would decrease rates of obesity and Type 2 diabetes. Basu and his co-authors found that it did. Using their findings, Indian officials have implemented national programs to tax sugary drinks.
“It was a bit unexpected,” said Basu. “As academics, we hope to have a policy impact but don’t fully expect that we will — so this has been really nice.”
Sharing the Wealth of Knowledge
To create change with a real impact, Basu has found there is much to gain from sharing the means of discovery.
“Within our group, we’ve always tried to be really open about data and code,” said Basu. “People have been using our approach to do crazy, interesting things that we wouldn’t have done.”
Basu’s team posted their work on a chronic disease simulation online which inspired Emmanuel Drabo, a student from Burkina Faso. Drabo ended up joining Basu's team as a post-doctoral scholar.
“Openness about data, codes and research methodologies is critical to maintaining the integrity of scientific research,” Drabo said. “Dr. Basu is a pioneer in this movement. This approach to research was an important factor in my decision to join his team, and I am forever indebted to him for his excellent mentorship.”