If their data is solid enough to justify it, they should publish it.
Until they do so, I remain deeply skeptical. It’s all too easy to engineer measurements or interpret data to support the outcomes you already want.
If the data is truly so revealing, show it to the world.