Looking at road accident data, it's likely that the original police reports would have given the names, as well as the ages and sexes and possibly other details like occupations, relationships, and so on of the victims and/or perpetrators. When data is aggregated, someone has to decide what information to lose. It may be that top-down decisions have been made to exclude certain details: most of us would like to know what makes and models of cars are involved in accidents, and to relate that to the numbers of each on the road, car and driver characteristics, etc. However, the public is not given information to allow us to make informed decisions as to which models to buy or shun. The insurance companies presumably have their own data on accidents and breakdowns, but aggregates of the publicly-owned data are kept from us to save the sensibilities of manufacturers. It must be the government that decided not to tell us all which models are the most dangerous to drive,. Vast quantities of health data are collected by the government. Traditionally they have looked at occupational risks of death and later at disease among factory workers: but maybe not at shops and offices?