First defund bad science; then invest in good science

If you believe in science funding, you should sacrifice silly ideas

If I were in charge of science funding, I would increase annual spending by several orders of magnitude. I would take 5% of the trillion plus dollars that CMS spends yearly and use it to conduct pragmatic randomized trials that test things CMS is paying for that lack evidence. If, as our work on medical reversal would predict, 40% of those practices are in error, the savings will be hundreds of billions.

At the same time, I would not increase spending at this moment. The current system is so broken that merely increasing science funding would merely mean more waste. The media talks about science like every dollar we spend is valuable. It is not. Let me give two examples.

First, the media is critical of RFK for cancelling hundreds of millions in federal funding for mRNA vaccination and in this case an oral COVID vaccine against KP2.

If any company wants to make an oral COVID vaccine they should raise private funds from investors and do so. The US FDA should make them power their trial for hard endpoints (hospitalization and death) and not accept immunologic surrogates. With these rules, very likely the company would not be able to raise money. Investors would ask: wait a second— how many people are actually hospitalized any more with COVID? What is the chance this pill (a hitherto novel vaccine technology) would work? Will Americans want this product and use it? Likely, it would fail.

Speaking of investors, become a paid or free subscriber to invest in these ideas

So why should the government continue to fund the company if investors don’t believe in it. In some situations you could argue broader societal good, but in this case, you can’t. Because COVID19 is no longer a novel pathogen, but one of the now 5 circulating cold strains.

Second, I saw a researcher at Columbia complaining that this grant was cancelled.

Again, climate change is, as of 2025, inevitable, as is the growth of Air Conditioning globally. This researcher is asking:

Does climate change —> change the temp—> make it harder to sleep —> bad outcomes

Alternatively he could ask

Does burning energy—> make ACs cheaper—> make it easier to sleep —> improve outcomes

But either way, the question is unanswerable. From a methodological standpoint, I can’t imagine an agreed upon way to investigate it. I bet if you gave it to 25 research teams across the political spectrum, you would get 25 different answers. But if you give it to 25 climate activist researchers, you would get 25 answers that are all bad but very different point estimates.

For that reason, I strongly suspect the researcher will conclude climate change makes sleep worse. I already know the outcome!

Finally, what difference does it make? It’s not like there is some politician on the fence about signing a climate bill, but then “I never knew it will affect my sleep!” The grant has poor value of information.

Science can be scientific but science funding is currently low tier sociology. It is faddish and untested.

Funding low credibility science on politically disputed topics is a cauldron for public distrust.

If scientists want more funding they have to be honest and admit that we are funding some silly ideas right now. Cut bad spending before asking for more money.