Books like “Why We Sleep” and podcasts like Huberman Lab are kind of like a junk food version of science: They deliver some bits of information that can be helpful, but they’re so coated in stretched truth, overconfidence, and stories crafted more to sound good than to be accurate that it’s not really accurate science material. In a case like yours (difficulty sleeping) that might actually be fine, to be honest. A lo…
While I have long agreed with this sentiment, I'd like to gently push back against the idea that "science" is itself coherent, reliable, monolithic, or able to be digested by a single human at all. I suspect what you're actually comparing to is in-depth reading of published papers from which you draw your own conclusions—reliable or not—drawn arbitrarily from a pool where there's basically zero way to actually verify the quality of the papers you're reading. In this light I don't think it's a bad thing that people are given a glimpse of scientific understanding through pop-literature because the vast majority of people—even those that read papers—grossly overestimate the certainty of knowledge that they have not specifically specialized in.
This is a long-winded way of saying basically nobody interacts with "science" directly. It all comes down to trust, and I strongly suspect that the trust in a give pop-literature writer isn't that much worse-placed than the average person writing the academic papers in the first place—outside the topic of the paper, of course. It's incredibly difficult to aggregate knowledge without serious evidence about how trustworthy the publishers and writers are, and sadly credentials and brand prestige tend to be piss-poor evidence.