In the most recent issue of the MIT Sloan Management Review, Professors Eva Guinan (Harvard Medical School), Kevin J. Boudreau (London Business School), and Karim Lakhani (Harvard Business School) detail the work Harvard Catalyst is engaged in to open up and reimagine how a university conducts research.
Harvard Catalyst is a “pan university clinical translational science center” situated at Harvard Medical School exploring how to rethink “all aspects of the scientific research process.” To the end of solving “high-risk, high impact problems in human health-related research,” Catalyst takes each aspect of the research process and designs a strategy for opening it up to more information and input from a wider audience. The goal is to bring in fresh ideas and novel perspectives.
In 2010, Catalyst undertook an experiment to open up how universities generate research questions. Typically, an academic decides on the direction for his or her lab. In an effort to generate new ideas from unlikely sources for promising approaches to fighting Type I Diabetes prior to investing research funding, Catalyst sponsored a $30,000 prize-backed challenge to come up with research topics that might be promising. (The White House copied this model when the Sloan Foundation crowdsourced suggestions for research topics in connection with Smart Disclosure in 2012). The Catalyst challenge was unique because it didn’t ask people to come up with answers as was typically the case in crowdsourcing projects. Rather, contributors supplied the questions. This enabled people to suggest ideas whether or not they had the resources to solve the problem they proposed. After six weeks, 150 solid research hypotheses were submitted encompassing a broad range of approaches from different disciplines. The authors analyzed the subject matter of the submissions and found that they were “quite different from what existed in the literature and from the existing body of ideas under investigation within the Type Diabetes research community.”
Rather than ask a handful of experts, Harvard Catalyst conducted an “open peer review process” with 142 judges from a variety of backgrounds. The authors then compared the impact of having a diverse and large group of reviewers to having traditional peer review by any three faculty members. They found that a triumvirate would have been far less likely to award grants to those that were the top picks of the larger group. Catalyst provided awards to the 12 best submissions based on average score of the large group. Winners included the unlikely suspects of a college senior, a human resources professional suffering from diabetes, a retired dentist with a relative with diabetes and a biostatics professor.
Catalyst developed a plan to implement and fund further work on the good ideas proposed. The fact that submitting something could lead to real results, the authors, contend, was part of the recipe for success. With these hypotheses in hand, Harvard Catalyst mined the profiles of all Harvard medical school faculties and advertised the opportunity to submit funding proposals to implement one or more of these dozen winning hypotheses to those most likely to be in a position to apply. In the end, seven proposals were funded. In the end, seven proposals were funded with five of the lead investigators having not done any prior work in T1 Diabetes.
It’s too soon to know the impact for sufferers of the disease of this experiment. Typical timelines for medical innovations translating from bench to bedside are 15-30 years. And we don’t know anything about the impact on those who participated in the challenge. Did the senior go on to do a PhD in biology or go to medical school? [Update: Lakhani reports that the senior (Megan Blewet) is doing a PhD at Scripps! Here’s an inspiring video of her presenting her proposal. Courtesy of Harvard Medical School)
It’s also not clear from the article whether the “crowd” converged on ideas that were more or less radical in approach. Where it succeeded, in the authors’ view, however, was that the experiment attracted diverse participation. In democratic terms, this engagement would be viewed as beneficial in and of itself.
[Lakhani replies: “It’s clear from our analysis that the ideas were different – i.e. the combination of approaches proposed in the various proposals were not previously in the entire medical and T1D literature - what we don’t know precisely is if they were left or right tail ideas. Radical is often an ex-post analysis.”
In the end, the experiment took off, they tell us, because Catalyst didn’t oversell the potential of open innovation work, especially at an elite institution like Harvard where people might be more likely to be skeptical about whether the open model could improve upon Harvard’s traditional model. “This strategic layering of open dimension on traditional processes positions open innovation as a tweak to currently accepted practice rather than a radical break with the past.” When it comes to reimaging how we work, sometimes it is important to take small steps and bring everyone along for the ride.