That is pure science - accept something until better data comes along and changes the way we think about the topic.
We lose people there though, unfortunately. People like to "know" something absolute, and the media does a horrible job of communicating disagreements or opposing data.
So, while we now know that Newtonian physics breaks down in some circumstances, where relativity and quantum mechanics are needed Newtonian is still good enough for most everyday and a lot of not so everyday applications (like sending probes to Mars).
It doesn't flip / flop like the "eating eggs was good yesterday and is now bad".
In general an element of both is present, but on one end tou have, say Newton, and on the other you have some study that says electromagnetic waves are bad for you because of some statistical health outcome but proposes no causal mechanism.
The problem seems to be that the further towards the latter part of the spectrum someone's research is, the more likely they are to be adamant that they're "doing science" and promote all the BS about consensus and peer review that gives a pretend version of how science happens.
Presumably due to insecurity. When your work rests on a foundation of p-hacking, you're more likely to be pounding "the science" as justification, vs if you're a physistics that's traced a result back to basic axioms of your field, your work speaks for itself and you don't need to yell how the science is settled
This way, we ensure that our meta-analysis firepower (whatever/whoever that actually is), can be directed at studies that actually have a potential to uncover some real knowledge.