Time dilation is a well tested result from special relativity, this is probably about frame dragging which is a prediction based on general relativity that hasn't been so well tested.
Right. I didn't realize, however, that satellites are moving at a significant enough speed relative to earth frame that this was necessary. If I recall, it was the Lorentz transforms (almost sadi Helmholtz, haha) we'd use to adjust from one frame to another.
it goes like t'=gam*(t-v*x/c^2) and gam=1/(sqrt(1-v^2/c^2) where v is the velocity of the prime frame relative to the other. You can see that gam gets pretty small if v isn't a significant fraction of c - as does v*x/c^2.
I guess I'm recalibrating my idea of the relative-to-earth speed of those satellites.