The Beloved was staying with her BFF when they were both teenagers. One day they were being driven by the Mammy in an open-topped car with two dogs. Passing a rival dog the two hounds went in paroxysms of barking and looked likely to plunge out of the car. The girls put a restraining hand on a dog each and the ungrateful terrier turned round and bit TB in the arm. For the rest of the holiday this was a major topic of conversation and, as the tale unfolded visitors would edge away from the Alsatian and closer to the psychotic terrier. Such failures of risk assessment are clichéed out of the frying-pan and into the fire, and there are a lot of them about. Famously, there were 1000 excess deaths in the aftermath of 9/11 because travellers avoided intrinsically air-mile safe aeroplanes and died making journeys by car. I have also made an argument that the space shuttle was as safe/dangerous as travelling by car in Ireland. Nobody said risk-assessment was easy: not everyone agrees about the balance between immediate possible vaccine damage to teenage girls vs distant possible HPV-induced cervical cancer to those same girls in 40 years time.
Modern governments, even those that purport to be entirely free-market capitalist, allocate resources to ensuring that drugs, food, the roads, air-travel are as safe as possible. If adverse events occur, protocols are put in place to see that they don't happen again but Risk Assessment is carried out. You don't want to spend $1 billion to prevent seamstresses getting needle-stick injuries but you might allocate money to prevent the same accident in HIV clinics or Porton Down. When it was shown that large doses of cyclamate artificial sweeteners caused cancer in lab rats, they were taken off the market because saccharine was available as a substitute. When saccharine turned out to be similarly (and not very) hazardous, the food lobby bullied the FDA into allowing them to continue its use. I wrote three pieces on this dilemma last October.
Regrettable Substitution is what scientists and risk-managers call frying-pan vs fire issues. The piece in Wikipedia cites the chemicals used for brake-drum clearing: dichloromethane was phased out because it was destroying the ozone layer and replaced by the solvent n-hexane which turned out to be a neurotoxin. That's a hard one because the targets for damage were barely commensurate: the atmosphere and the planet being set against mechanics in garages. Dichloromethane is a rather potent CFC which, as we now know, are super-stable at ground level but eats ozone when UV is added to the mix up there. Pressurized CFCs were used for decades as propellants for aerosols: bug-killer, deodorants, WD-40. They were replaced by butane gas which is a lot less stable than CFCs <WHOOOMPH!> . . . a flame-suppressant is added too.