Swedish researches say they have found a tentative connection between autism and maternal use of either SSRI antidepressants or tricyclic ones while a mother is pregnant.
First, it's good that the study was cautiously worded. That's good research, not headline hunting.
Second, it's not the first study to find some sort of link.
Third, it doesn't claim this is the only link.
Fourth, it makes sense.
Throughout the developed world, antidepressant use has risen in the past couple of decades, as have births by older mothers, along with impregnation by older fathers, and age of both mothers and fathers has also been linked to autism.
Note: "Connection" or "association" are better words than "correlation" on this, as the degree of increase is quite small. Given that autism is a spectrum of disorders, somewhat of a grab bag, that's to be expected.