We run into a few interesting possibilities here. Start with the assumption that more children are being diagnosed as being on the autism spectrum. That gives us a few possibilities.
Because there's more and better screening autism spectrum disorder (ASD) is being caught more often. Okay, maybe. But.
1.a) If more children are being appropriately diagnosed with ASD, then perhaps the criteria needs to be tightened up; at a certain point, behavior/feelings/thoughts are just normal.
Because there's more screening--but not necessarily better screening--children are being pathologized as having ASD when they do not, because too many clinicians don't have the necessary expertise. This is a distinct possibility, in much the same way that kids are being labelled as having ADD/ADHD--and then getting drugs--when they're more frequently just being kids.
More children are actually on the autism spectrum now than there were 30 years ago. E.g., it's not that more kids slipped through the cracks 30 years ago, but there is actually a higher rate of ASD than there was 30 years ago. This is the one that should cause the most concern; if this is actually the case, and can be demonstrated to be the case, then what factor is causing this maladaption?