We DO know the effects of radiation. Both lower dose, longer term exposures and high dose short term.
And the same issues with determining low dose effects occur with other agents than radiation. You need to follow a LOT of people, a LONG time and you have to have access to an equal number with no exposure. And then have to make sure that you actually have causality, not just correlation.
There was a study published about the effects of extremely low frequency RF radiation from power lines. Based on the voltage and number of phases on the distribution through neighborhoods. After it was published, another person showed that the data could have been used to compare the same health effects to the number of cars going by per day. This was since the power lines follow roads and the further you get into neighborhoods, the lower the transmission line voltage and also less traffic as there are less houses served by that road.
In general, EPA uses a linear dose response curve model. That is, they consider the only safe dose to be 0. But a lot of things are actually beneficial or required in low doses. Such as Zinc. A lot of zinc is bad. If you had a diet of 0 zinc, it would also be bad.
Lead has been implicated in developmental issues in children. In adults, there are less issues, especially in low doses. And lead that does get into an adult, is secreted over time. In the old days, lead miners would be exposed to high levels and have health effects. The "cure" was to have to not work in the mines, but they would spend their shift walking around in a very hot room. This would increase the rate the lead was excreted, and get them back to work in the mines faster. Not exactly acceptable practice today.