Trying to cut down on groupthink in medical research
Happy World Cancer Research Day to those who celebrate by receiving fat monthly checks from NIH and NSF. (Remember, though, that if cancer were cured tomorrow by a heroic Dr. Anthony Fauci we would all still die of diabetes not too long after our current expected death dates.)
“Launching a Second Scientific Revolution” is an interesting article by Jay Bhattacharya, the current director of NIH and former heretical Stanford Med School professor who argued, beginning in March 2020, that lockdowns were counterproductive in our war against SARS-CoV-2.
Some excerpts from the above, not in quote style for readability:
… for every dollar we spend on science today, we get far less scientific advancement than we did over the past five decades. Another way of looking at this is that for every additional research paper in biomedicine, there are fewer improvements in health per paper.
… most scientific papers, far from introducing new ideas, rely on old ideas. I once collaborated on a study that looked at every word and phrase published in biomedicine in 1940 and 1941. When we subtracted all the 1940 words and phrases that were published in 1941, we were left with the ideas introduced in 1941. Doing this for every subsequent year up to the present produces a kind of history of biomedicine. What we see in that history is that in the 1990s, about 55 percent of NIH funded research was based on new ideas—zero-year-old or one-year-old ideas—and that research in which the newest ideas were ten to 30 years old received much less funding. And that is as it should be. The NIH is supposed to fund ideas that are on the bleeding edge of science. But what has happened since then—especially since 2000—is a collapse in NIH support for new ideas.
… the average age at which scientists receive their first NIH grant has risen from the mid-30s in 1980 to the mid-40s today … younger scientists are the most likely to have new ideas.
The NIH receives about 100,000 funding applications each year, out of which only eight to ten percent are approved. In recent years, the winners have been chosen by scientific reviewers who score each application based on two factors: (1) the strength of the proposed methods and (2) the innovation involved in the proposal. Looking at the results of this process, the overall score has correlated very strongly with the methods score, and innovative ideas have been left by the wayside. To fix this, the NIH is replacing that process (called the payline method) with a new approach called the Unified Funding Strategy. The scientists who review applications will now be empowered to approve projects on the bleeding edge of science—projects that may very well not work but that will produce a fundamental advance in biomedicine if they do.
The third problem to be addressed is the fact that about one third of all NIH grant money goes to about 20 institutions, despite the fact that there are excellent scientists all across the country. … Our solution to this problem is to sever the link in the NIH grants between the direct money for research and the indirect money for facilities. Institutions outside of the top 20 will be able to compete for facilities support, leading to a dynamic market. The number of institutions that receive NIH money will increase, and funds will go to wherever the best scientists are.
Google AI:
The Howard Hughes Medical Institute (HHMI) funding model is widely considered more effective than traditional government grants because it funds “people, not projects,” bypassing the restrictive and expensive bureaucratic overhead typical of university research. Instead of requiring professors to apply for short-term, hyper-specific grants that divert up to 60% of funds to university administrative overhead, HHMI appoints scientists as employees, providing them with long-term, flexible capital to pivot their research freely. HHMI investigators produce high-impact, heavily cited papers at a significantly higher rate than their NIH peers.
Who wants to bet on the next significant advance in medical technology? I’m guessing that it will be primarily by an AI, not by the humans who’ve been mostly stalled for decades, so it needs to be in any where there are good data to feed to the AI, e.g., genomics, radiology, etc.
Full post, including comments
