I think the takeaway is that the scientific method is great, but since science is necessarily conducted by humans, the economic incentives within context of "doing science" also need to be correct for good science to get done.
(Currently, it seems like, this is very often not the case. Hence, your trust in current-science-as-conducted should be correspondingly lower. Lots of variation by field, probably, too.)
(Though then again maybe you shouldn't even believe anything until you see it having actual real effects in your day to day life?)
