I used to read descriptions of deaths by electrocution in Finland as published by authorities for amusement and learning purposes, and I did notice the trend that while in 1980's 20 people died every year (out of less than 5 000 000, it was not a lot compared to 2000 in traffic accidents of the time, but still something), over half of them in households, that dropped to 0.5 per year in 2000's, 1:40 reduction. Meanwhile, traffic accident deaths saw 1:10 reduction, and a lot more work has been required to achieve that, mostly in car design.
20 a year out of 5,000,000 is negligible IMO, forget traffic accidents and look at less common stuff like falling down stairs, drowning in bathtubs and choking on food, I would bet that electrocution is not even on the top 10 causes of untimely death. It would be interesting to further break down the electrocution deaths and look at additional details, if you wanted to compare plug safety for example it would make sense to look only at cases involving that.
Taking ratios with the population is pointless, as is comparing to other accident types. If there are 20 deaths per year, that is 20 deaths per year that can be prevented.
I have a dock, and my neighbors have docs with electricity. There is a fault mode where the electric line develops a loss of containment and electricity leaks into the water. The current flows out from the source and flows through the body of people in the water, paralyzing them, so that they drown.
Since this is not a category previously defined, and it is not well known to the people who investigate such matters, it seldom is reported as the type of problem it is. So there is little data showing it is a problem which needs to be addressed. Catch-22. Until there are more reports of deaths, they won't even create a category for it. Until there is an appropriate category for it, there won't be more reports of deaths due to this problem.