It's not causation since no experiment was conducted to measure this. It's a weak correlation at best, since again, no correlation study was conducted to measure this outcome. It was a meta analysis not an experiment. That's the limitation of the study. It should be touted as proving anything. What it does point to is an interesting of study, something that should be looked at.
I think a reasonable conclusion to be drawn is that prolonged sitting is associated with a number of bad health outcomes and that prolonged sitting should be broken up with breaks during the day.