Yes, this is something I've been tracking for a while now. My Roland R-07 picks up a very distinct 2-3 second burst of white noise type static just before a Class A or B response comes through, and I mean consistently, not just occasionally. I started logging the gap between the static onset and the vocal component and it averages around 1.8 seconds across about forty sessions up in the Cairngorms area.
My working theory is that whatever is producing the EVP is drawing energy from the local electromagnetic environment first, and the static is essentially the acoustic signature of that draw happening. Like a capacitor charging before discharge, if that makes sense.
What recorder are you using and what gain settings? I wonder if cheaper condensers are actually more sensitive to this precursor effect because they pick up more ambient noise floor. Would be useful to compare notes with people running different kit. Has anyone tried running a gaussmeter simultaneously to see if the static correlates with any EM spike?