Often times, you can find cures just by discovering mechanism of action, without PICO/RCT: for example, Helicobacter pylori causing peptic ulcers.
Placebo effect is NOT magic, but we don't understand the mechanism. Good relationship between doctors and patients help recover patients. Patients prefer to see particular doctors, even if the latter prescribe same medications as others, yet the differential effects on patients (when treated by preferred doctors vs others) is non-trivial.
[0] https://en.wikipedia.org/wiki/Category:Drugs_with_unknown_me...
I’m not what 7% of reviews represents, but it seems like particularly dumb austerity measure.
But more importantly Cochrane allows for the pre-registration of such reviews, i.e. where the methodology and goals (hypotheses) are stated before the review is conducted. This is probably the most scientifically rigorous method of assessing evidence, and is not performed for most systematic reviews.
They do mention that Cochrane is only accounting for 7% of systematic reviews these days.
BTW, evidence-based medicine is not the same as science-based medicine. The Placebo Effect is evidence-based medicine.