I look the ref out for one of the trials for an atypical neuroleptic,
The control was thorazine, for which no additional therapeutic effects occur at above approximately 350mg a day. Side effects continue increasing in a fairly linear manner with increasing dose.(discovery of that ceiling for therapeutic dose came several decades after thorazine became available, hence the mental hospital inmates of my childhood memories, had something very visibly wrong with them, with distorted faces, weird movements,twitches,and slurred speech... Because they'd been receiving ludicrous doses of thorazine for years).
The trial had the atypical at an optimum therapeutic dose, the thorazine was at some ludicrous dose, iirc, about 3.5g day.
Result, the atypical had fewer side effects...
Even if a quack spotted that particular skew in that particular trial. Later meta analysis assumes that all trials are created equal, and the skew can be inherited by an analysis claiming a larger sample size and hence greater statistical significance for its conclusions. Few, outside of the industry would have the time inclination or resources to go through all the included studies to find that kind of deliberate flaw.
To say that it is almost as bad as politics for lies and distortions, would be partly true
The lies and distortions are inherited from politics, via the state's grant of monopoly privileges in return for a payment to the state - in the form of a patent.
A clear example of the distortion is available in the histories of the two main drugs for psychoses, the neuroleptics and lithium.
The effects of both were discovered and published at around the same time, however, lithium was a common widely available and cheap chemical, thorazine was patented by Rhône Poulenc...
Without the special privileges available with a patent for one, we could probably have expected physicians looking for the best treatment for their patients to have adopted both with similar rapidity, that isn't what was seen.
Similarly the rush for atypical neuroleptics appears to have more to do with their patents than it has to do with their therapeutic advantages compared to the typicals.