How do scientists, lacking the ability to make controlled genetic crosses in humans, study the inheritance of?

when there is a preponderance of dependent and independent evidence, control isn't necessary in experimentation.

If millions of type A moms and type B dads have type AB kids, you really don't need controlled genetic crosses to come to a conclusion.

The underlying hypothesis WERE tested using controlled genetic crosses... one the theories were formed, and applications were considered and developed, they could be "logically" extended to apply to humans where controlled genetics experiments really weren't feasible.

The logic says "if it works on the rats in the lab, it will work on humans too, with an acceptable margin of error"