> They write that "an interventional method, such as a randomized controlled trial using a LAN exposure, is not possible because of ethical concerns".

They're wrong about that. The intervention group could be advising people to avoid light, and having someone come to their house and install better shades and put tape on LEDs.