Pharma's models are a joke compared to "the thing itself". It's more like a sitcom formula: p-hack it through trials, scam it through the FDA, harvest the prof

Pharma's models are a joke compared to "the thing itself". It's more like a sitcom formula: p-hack it through trials, scam it through the FDA, harvest the profits, set some aside for the lawsuits. If the liability proves too much, start over. If not, fiddle with the formula just enough to extend the patent, but not enough to require a new round of FDA approvals (see: evergreening). Some legitimate cures may emerge from this process, but only stochastically.