The German In-Depth Accident Study (GIDAS) collects since mid 1999 accidents in Dresden and surrounding area as well as Hannover and surrounding area following a statistical sampling method that makes the data set representative for the German accident situation.

After a case by case review of all child restraint system (CRS) items of the data base the data quality is improved significantly and now allows for example analysis of the protection level of different CRS architectures (e.g., rear facing compared to forward facing or forward facing integral compared to booster seats) what was hardly possible before because of lack of information. As far as possible from the accident information also use pattern (no misuse detected, misused CRS inappropriate use of CRS will be included in the paper as well.

For the analysis the German national accident statistics and GIDAS data from the entire collection period involving children up to including 11 years as car occupants in accidents not including accidents with vulnerable road users.

The data set was recoded for CRS and CRS use variables based on a case-by-case review using child safety expert knowledge.

In frontal impact accidents without further consideration of age children in rearfacing CRS are protected best while forward facing integral CRS and booster seats are slightly less protective. In lateral impact accidents forward facing integral CRS are protecting the occupants best.

Although the case-by-case review allowed for a large number of cases a recoding of the variables there are still a number of unknowns in the data base that limit the final quality of the final conclusions.