I think we have to accept that the complexity of hormonal interactions, that can change minute by minute, is beyond the realms of what scientists can manage being able to measure and perform studies on.
This is when trial and error can help as you may just find something that gives an element of control to reduce the volatility that causes symptoms. Then those who have had success with a strategy can pass that on for others to try.
If you stick rigidly to what science can measure and ignore everything else you take the risk of throwing the baby out with the bath water.
I love science but the pandemic showed us how the risk aversive nature of how scientists tried to contain Covid-19 in most countries, versus Sweden's anti-scientific, common sense approach, put those countries at even more risk. Science can only do extremes of black and white, yes and no, positive or negative correlation etc. By ignoring the grey areas it may, for example, cut off too soon at the point where longer use might start to prove effectiveness or ignore infinite factors that may have hampered results.
Absolutely, try for the scientific proven methods first but don't be afraid to try alternatives, as not having gone through expensive trials is not the same as having gone through them and failed or caused harm.