The problem with that approach is that it is just about acute lethality and does not consider long-term effects from prolonged exposure that are hard to measure

The problem with that approach is that it is just about acute lethality and does not consider long-term effects from prolonged exposure that are hard to measure. Further it ignores interactions with other chemicals that are also in use, something that we know is pretty common from medications. Especially with endocrine disruptors there is also the effect that several chemicals might work in the same direction, where a safe dose of one chemical might be harmless but safe doses of 10 chemicals combined might not be safe anymore.