Strangely, this wasn't mentioned in this or the other English-language articles about this that I saw: The setup was to during 23 months test if six-hour days

Strangely, this wasn't mentioned in this or the other English-language articles about this that I saw: The setup was to during 23 months test if six-hour days resulted in any meaningful improvement on the (rather high) amount of sick leave the employees used. The conclusion was that the amount of sick leave only went down with 0.6%.