VAERS data are just unverified submissions - possible events that need to be investigate, by themselves they are indicative of nothing.
That would be like reading the amazon product reviews and assuming they have all been proven accurate - they are not.
Besides that, 10-15 percent of all pregnancies end in miscarriage, so if you want to prove something has to do with the vaccine, you are going to have to compare it to a control group - 3000+/- possible miscarriages, while sad, says absolutely nothing by itself; last I saw more than 125K pregnant women have had the vaccine - if the 10-15 percent miscarriage rate holds than you would expect 12-18K miscarriages from that group at a minimum; your data says 3k+/-, so in fact that is a much lower number - so maybe the covid vaccine is actually decreasing miscarriages? (see how easy it is to twist statistics to prove anything you want).