And given the comparatively modest energy output per unit, it wouldn't take many rooftop falls to put a substantial dent into the deaths per trillion kilowatt h

And given the comparatively modest energy output per unit, it wouldn't take many rooftop falls to put a substantial dent into the deaths per trillion kilowatt hours number. Put it another way, if California produced 0.0004 trillion kilowatt hours of solar power every year, that would put the death rate of solar installers at one every five to six years. Sounds plausible.