There are several versions of multiverse theory. The version I am going to discuss here is derived from string theory and the notion that there are many extra dimensions that we don't have access to (just like Matthew McConaughey in the fifth dimension in "Interstellar"). Some physicists think parallel universes are present in those extra dimensions. This multiverse notion is testable too. Physicists will be looking for mini black holes when the Large Hadron Collider restarts this month. It's impossible for the LHC to generate any sort of black hole that would be remotely unsafe, but this theory suggests that microscopic black holes that vanish almost instantly could be produced from the high-power particle collisions in the LHC.
The existence of black holes would specify that gravity from our universe is leaking into extra dimensions. Physicist Mir Faizal told Phys.org "As gravity can flow out of our universe into the extra dimensions, such a model can be tested by the detection of mini black holes at the LHC. We have calculated the energy at which we expect to detect these mini black holes in gravity's rainbow [a new theory]. If we do detect mini black holes at this energy, then we will know that both gravity's rainbow and extra dimensions are correct." That would be exciting confirmation for both string theory and parallel universes, and it would help clarify why gravity appears to be so much weaker than the other fundamental forces.
Still, there's no solid confirmation yet. And several physicist still doubt that these universes exist.
Brian Greene, a theoretical physicist at Columbia University, said while arguing about the multiverse theory "I only believe in things with concrete, verifiable experimental evidence supporting them, and that's not the case right now with the concept of parallel universes,”
The key, however, is that scientists are getting away from just philosophical considerations of the multiverse theory. They're truly putting the notion of multiverse to the test. Some are still betting on the more fundamental and so far untestable sorts of multiverse theory.