Update: If trials by manufacturers are biased - then how about some framework that would push trials made by competition. There are two things: One is bad scie

Update: If trials by manufacturers are biased - then how about some framework that would push trials made by competition. There are two things: One is bad science, and we have a lot of it (https://en.wikipedia.org/wiki/Replication_crisis https://en.wikipedia.org/wiki/Replication_crisis, https://library.mpib-berlin.mpg.de/ft/gg/GG_Null_2004.pdf https://library.mpib-berlin.mpg.de/ft/gg/GG_Null_2004.pdf) - but it should be improving and most of it is probably harmless (I have some hopes on better statistics education and https://www.google.com/search?q=causality+revolution https://www.google.com/search?q=causality+revolution). The other is bad application - and this is more complex. When I go to doctors (with my and my family issues) I, more often than not, have the feeling that they have no chance to give any good diagnosis as it would require too much time to test etc. (one simple case is when to use antibiotics and when the disease is virus based - but I have noticed that this is probably improving with strep and CPR tests) and they mostly just shoot first thing plausible, proscribe some medication and then in most cases the organism just cures itself - so the treatment is never tested. There is probably also 'medicine as signaling' (the theory by Richard Hanson: http://www.overcomingbias.com/tag/medicine http://www.overcomingbias.com/tag/medicine), but my intuition is that it is not verly important, and other stuff.