Skip to Content

Deepika Mohan Receives NIH Director’s New Innovator Award

Mon, 10/05/2015

When Deepika Mohan cold-emailed all the professors at Carnegie Mellon University’s School of Social and Decision Sciences in February 2009, asking for help with her research into physician decision making, she was not confident in the prospect of getting a reply. But then, days later, she got just that. Suddenly, Mohan’s luck was about to take a propitious turn.

In July 2014, Mohan, Assistant Professor of Critical Care Medicine and Surgery, met with Jeremy Berg, Associate Vice Chancellor for Science Strategy and Planning in the Health Sciences, University of Pittsburgh School of Medicine, who encouraged her to submit an application to the National Institutes of Health for a grant designed primarily for early stage investigators whose research doesn't necessarily fit into the traditional pathways. Mohan, who studies how physicians make decisions and in particular the role that intuition, or heuristics, plays in this process, had been finding it difficult to get traction for her research, which could easily fall into the category of behavioral science as much as it could traditional clinical research. After receiving a K award in 2012, Mohan had run into problems securing additional funding to continue her research. When Berg had told Mohan that she “had nothing to lose,” Mohan couldn’t agree more.

On October 6, 2015, Deepika Mohan, MD, was announced as one of 41 recipients of the 2015 NIH Director’s New Innovator Award from the National Institutes of Health. The New Innovator Award, which is a part of the NIH Common Fund’s High-Risk, High-Reward Program, “supports unusually innovative research from early career investigators,” according to a press release from the NIH. The total funding given through the High-Risk, High-Reward Research program for 2015 awardees is $121 million. Previous recipients have been listed in "The 100 Most Influential People" in Time Magazine, and have been named Young Global Leaders by the World Economic Forum.

Mohan’s interest in how physicians, and people more generally, make decisions began while an undergraduate pursuing a bachelor’s degree in Religion and Political Theory at Princeton University. Mohan studied how pluralistic societies negotiated over resources when the competing parties lack a shared conception of what constitutes the “common good.” Later, when pursuing a Master of Public Health degree from Columbia University, she continued this line of inquiry, looking into how health care resources are allocated. But what was most interesting to her, beyond how the resources were eventually divvied up, was how people came to their decisions in the first place. Though Mohan enjoyed her research, she felt that the liberal arts track lacked sufficient opportunities to participate in initiatives with tangible real world implications, so she applied to medical school. Little did she know that she'd soon be applying the lessons that she learned from her more theoretical studies to try to solve some of those very practical problems.

But what was most interesting to her was how people came to their decisions in the first place. 

While in medical school, Mohan began to notice that what defined an expert physician from an average physician was almost intangible. Good doctors “almost never can explain how they are able to tell that someone is going to become sick,” says Mohan. “There is frequently nothing obvious about any of the patient’s signs or symptoms.” Mohan’s gut was telling her that the key was just that: the gut—that doctors use their intuition to make decisions, rather than any particularly rational weighing of patient information and various external factors, as some would have explained it. This intuition, which sets some physicians apart, is both a defining characteristic as well as an indefinable instinct that is nearly impossible to replicate.

But this cuts both ways: doctors can also act in entirely irrational ways. Mohan began, with mentorship from Derek Angus, Chair of the Department of Critical Care Medicine, Amber Barnato, Associate Professor of Medicine, and Matthew Rosengart, Associate Professor of Surgery and Critical Care Medicine, to hone her inquiries, eventually asking, Why do doctors behave in a particular way?, Can we learn anything about how to make good decisions from the effective doctors?, and Can the ones with poorly calibrated heuristics be taught how to make better decisions? In other words, can good intuition be taught—and learned?

Prior research on the topic wasn’t particularly helpful. Most of it had been done in experimental contexts, and therefore lacked external validity. There is nothing to suggest that a college student’s mental process of estimating how many pennies are in a jar can be used to explain why pilots or doctors operate the way they do in cockpits and operating rooms.

One presiding theory on the matter posits that people are inherently flawed, and that the best that can be done, then, is to try to reduce the opportunities to make mistakes. As the U.S. Institute of Medicine titled their 1999 report on the U.S. Health System, “To Err is Human.” The limitation with that strategy, Mohan perceived, is that it eliminates a potentially powerful tool: the human capacity to parse complexity in ways that even computers cannot. That which makes us human is both curse and blessing.

Mohan was convinced that doctors shouldn’t just be removed from the algorithm, but rather needed help in honing the way they interpret complexity. She posited, employing theories put forth by Baruch Fischhoff, Daniel Kahneman, and Amos Tversky, that better heuristics could be developed using simulated environments and immediate feedback. For a trauma surgeon in a Level I Trauma Center like UPMC Presbyterian, which has more than 130 beds in its trauma units, there is no shortage of feedback about whether the decisions she makes are good or bad. But for an intensivist in a community hospital, who might see one critically ill or injured patient for every 1000 patients, opportunities for feedback are rare. Is it possible, then, to simulate an environment and provide immediate evaluative information in a way that helps develop better heuristics?

The human capacity to parse complexity in ways that even computers cannot is a potentially powerful tool.

The pursuit of an answer to this question is what led Mohan, upon Barnato’s advice, to send emails out of the blue to all those CMU professors, and how she found herself in communication with Fischhoff, the Howard Heinz University Professor in the department of Social and Decision Sciences at Carnegie Mellon University. Fischhoff, who had done research as a post doc under Kahneman, a recipient of the Nobel Memorial Prize in Economics, has made his name in heuristics, particularly in developing the concept of the hindsight bias, or the “knew-it-all-along effect.” With Fischhoff’s help, Mohan turned towards serious games, which could simulate an environment and provide real-time feedback. With this New Innovator Award, Mohan will have the resources she needs to develop and test a new intervention to recalibrate physician heuristics by using serious games.

Why might all of this be useful? Because almost 50% of all severely injured patients receive care at non-trauma centers rather than at high-level academic centers, and because this occurs in part because of the decisions that physicians make, helping to crack such a pesky “real-world” problem could have huge implications. As NIH Director Francis S. Collins, MD, PhD said, “This program has consistently produced research that revolutionized scientific fields by giving investigators the freedom to take risks and explore potentially groundbreaking concepts.” Sometimes, when you’ve got nothing to lose, winning can mean so much more.

--Ethan Lennox