How much NIH funding is waste?

It starts to add up

There's no shortage of people complaining about cuts to science funding. Of course, they have a bias. If you're enjoying an academic career living off public funds, you're not going to like cuts to those funds.

Cuts to science funding can be unwarranted, or they can be justified. Recently I saw New York times article complaining about a randomized control trial of the flu vaccine that was cut. It was a study across many hospital systems where pediatric patients received a vaccine before discharge versus usual care. On the face of it, it seems like a useful study.

Then I looked up the primary endpoint. I expected to see something like all cause mortality. Maybe influenza related mortality or influenza related ICU visits. Instead, I found the endpoint was:

Receipt of flu vaccine.

You mean to tell me we're spending millions of dollars on a randomized study comparing giving a flu vaccine before discharge versus usual care and the primary endpoint, the thing gets powered for, is receiving a flu vaccine?

What if I randomized people to getting a Big Mac, versus usual care, and the primary endpoint was how many people ate McDonald's?

Now the authors might say something like our secondary endpoints were to look at… but that's already misleading. If it's a secondary endpoint, the study is likely not designed or powered to evaluate it.

How much of science funding is wasted?

How much of our precious taxpayer money is thrown in the toilet?

Well obviously any basic science research that cannot be replicated is a waste of money.

Many empirical studies suggest this is the majority of basic science research.

You take a post doc or a PhD student, you tell them that they will never have any career unless they generate results, then you leave them have access to equipment where the result can hinge on how the equipment is set or managed, and you wait. There's no question that in an environment like this fraudulent and reproducible results will run rampant.

Indeed that was the case with the former Stanford president. Wherever he went, questionable results would follow. Obviously it probably was the fault of the junior researchers doing the science, but also very likely he created a pressure cooker where they felt like they had no alternative.

One obvious thing is that the national institutes of health should sample experiments from major universities and labs and check if they can be replicated. People in the bottom quartile of replication scores should lose their funding. Ironically, some of the biggest labs may crumble.

The next sort of waste is asking questions for which there's a right answer. Who has better surgical outcomes, men or women? If you do this study, only one answer can be published. If you find the answer that pleases the political left, you can publish that, but if you find the opposite answer, you cannot.

There's so much of this sort of research. Does banning bump stocks reduce gun related death? They're only two acceptable answers: yes, and it does, but we were unable to demonstrate that in this study, because we didn't capture the necessary information.

Did the Dobbs decision hurt women? The only acceptable answers are yes, or it did and we are unable to measure it. Had we measured it correctly we would have found it did. Never mind that national evidence shows an increase in abortions since the decision.

What sense does it make to fund science that directly applies to ongoing political debates, when that science is preferentially done by liberal researchers, at liberal universities, with liberal peer review, with a clear agenda?

Instead, it makes complete sense to defund this research.

Racial disparities. Billions of dollars have been spent on racial disparities research and the net effect has largely been to enrich the wealthy, upper class, researchers who will conduct this work. The least effective way to help a racial minority is to give money to a university to study the problem.

Then there is just bad science. Each year millions of people who have never done science before start doing it. They make loads of mistakes. Methodological mistakes. Ignorance of history mistakes. Numerical mistakes. But the system is particularly ill-equipped to criticize these studies.

How much medical science is false or useless? Different empirical estimates exist but it's entirely plausible it's the majority. In fact, this was the conclusion of two famous papers by John Ioannidis.

So what does science funding mean in a world where so much of it is wasted?

It's a tax on all citizens of America to divert large amounts of money to wealthy well-educated people, in liberal universities, who are intolerant to individuals who disagree, to further weaponize science in political disputes. Given this, it is no surprise that it is being fiercely defunded.

I'm surprised that people are surprised. The cycles of boom and bust are just going to get worse. Universities have made no effort to actually cater to the median American or to actually have common sense, or follow science properly. As such, very likely, future Democratic administrations will give lots of funding to science, leading to massive hires, which will follow by future Republican administrations cutting it severely, leaving to massive layoffs.

What's the way to escape this cycle? Perhaps universities could consider doing what they're meant to do, which is actually have voices across the political spectrum, across ideological beliefs, including heterodox voices and actually enforcing the rules of methodology in science. I won't hold my breath. In the meantime, I'll read the next New England journal article telling pediatricians to advise their patients not to play American football. ;) a science-based position obviously.