The physics behind modern VLSI was known to something like 8 sigmas of certainty before it was designed into devices. A lot of medical science, on the other hand, comes from one study with a p-value of 0.05 that didn't replicate when another lab tried it.
Science is having confidence in the former. Scientism is having confidence in the latter.
It's worth pointing out that the standards for evidence in particle physics are so high because they have to be: particle physics experiments create so much data that you can find a huge number of correlations with high confidence without much difficulty. Biology and medical science are completely unlike that: there is comparitively very little data to draw any conclusions from so standards of evidence are necessarily low, and yet still useful (that said, the lower standards also allow a lot of junk to appear which is difficult to filter from the rest if you are not very familiar with the subject).
I think we should point out that part of that 8-sigma confidence I am quoting isn't just the initial experiment (which may have had as little as 3-4 sigma confidence intervals for quantum tunneling), but also that there are further experiments that re-confirm quantum tunneling by relying on it to observe some other effect.
That is something that is also lacking in the social sciences, medical science, and even some parts of modern physics: experiments that confirm other results by taking those results as fact and finding new effects.
Science is having confidence in the former. Scientism is having confidence in the latter.