As of this writing, large swathes of the scientific world are shocked (shocked!) by an unexpected move from the Trump Administration.

Writes STAT News: "A flurry of scientific gatherings and panels across federal science agencies were canceled on Wednesday, at a time of heightened sensitivity about how the Trump administration will shift the agencies’ policies and day-to-day affairs.

"Several meetings of National Institutes of Health study sections, which review applications for fellowships and grants, were canceled without being rescheduled, according to agency notices reviewed by STAT. A Feb. 20-21 meeting of the National Vaccine Advisory Committee, a panel that advises the leadership of the Department of Health and Human Services on vaccine policy, was also canceled. So was a meeting of the Presidential Advisory Council for Combating Antibiotic-Resistant Bacteria that was scheduled for Jan. 28 and 29.

"The scope of the cancellations was unclear."

The story goes on to quote an HHS spokesman: "This is a short pause to allow the new team to set up a process for review and prioritization. There are exceptions for announcements that HHS divisions believe are mission critical, but they will be made on a case-by-case basis."

So basically, the Trump Administration is pausing NIH activities pending review.

This is a very good thing.

The politicization of science, particularly the health sciences, has been a disaster, not only for public health, and for the public's trust into public health, but for science itself. And nobody who knows this field doubts that it has been extremely politicized these past few years. We can think of high profile cases like Covid and so-called transgender issues, you can also look at the implementation of DEI across the NIH under the Biden-Harris Administration, or you can pick your favorite example of a crazy-sounding Federally-funded study on depression among the LGBTQ+ indigenous populations of Peru or whatever.

But this is only the tip of the iceberg. When you speak to most conservatives, they are under the impression that, oh, yes, there's some crazy political stuff going on, but there's also real scientists working on "real science."

But this is increasingly doubtful. One of the most important issues today is the so-called "replication crisis" in science, particularly biological science and medicine. The reason science works is because it's replicable. The reason why we know Newton's laws of gravity are true (at least within specified parameters) is that anyone can replicate Newton's experiments and see for themselves that, under the same conditions, the same results obtain every time. This is how we know a scientific law is actually a law of the universe and not somebody's pet theory: not just when one experiment is done, but when that same experiment is replicated many times under the same conditions by many independent teams. On some level this is basic common sense, but it is also what caused the scientific revolution.

The replication crisis, then, is the fact that, over the past decades, it has been proven harder and harder to replicate findings from studies, particularly in biology and medicine. This issue gained widespread attention in the early 2010s when several high-profile attempts to replicate important studies failed, raising concerns about the reliability of published research and the robustness of scientific findings.

In the biological and medical sciences, this crisis has particularly serious implications due to its potential impact on patient care and drug development. For instance, a landmark study by Amgen researchers attempted to reproduce 53 "landmark" preclinical cancer studies and could only confirm the original results in 6 cases. Similarly, pharmaceutical companies have reported difficulties in reproducing basic research findings, leading to wasted resources in drug development programs.

There are many innocent reasons why a study might fail to replicate. But there are also non-innocent ones. Perhaps the most infamous case involves Dr Sylvain Lesné's influential 2006 research paper in Nature, which identified a protein called Aβ*56 as a potential key culprit in Alzheimer's disease. This research significantly influenced the field's direction and received substantial funding. However, in 2022, a Science investigation revealed evidence that images in Lesné's paper and subsequent publications appeared to have been deliberately manipulated and falsified. The paper had been cited over 2,300 times and had helped reinforce the dominant amyloid hypothesis of Alzheimer's disease. In other words, this fraudulent paper had been the basis for entire swathes of Alzheimer's research, which we only discovered over 15 years later is worthless. Many people died because of Lesné's fraud.

The replication crisis is a disaster. And one reason why the replication crisis is ongoing is because of incentives. There are very strong incentives in the academic world to publish new results, especially (seemingly) groundbreaking results. There are few incentives to try to reproduce existing results, which is expensive. There are also incentives against it, since a scientist who makes it a habit to essentially try to disprove his colleagues' research is not likely to make friends or gain promotions.

Which is why this new administration is a great opportunity to do a seemingly mild, but actually earth-shattering reform: mandate that 25% of NIH funds spent every year are spent on attempting to reproduce existing research.

The second order consequences would be massive, as it would completely change the incentive structure of academia. It would certainly do a lot to deter fraud. It would also go a very long way towards fixing political bias, as scientists who wish to produce politicized results would know they would be less likely to successfully reproduce. It would be one of the most positive things for science done in decades.