Diagnostic trends in autistic spectrum disorders in the South Wales valleys.

Fullana et al. (2007) counted every child with an autism diagnosis in the South Wales valleys. They looked at clinic records from 1992 to 2005. They split the kids into three groups: classic Kanner autism, Asperger syndrome, and other ASD labels.

The overall rate was the kids per 10,000. Classic autism stayed flat. Asperger diagnoses rose fast and drove the whole increase. 'Other ASD' labels also crept up. The rise came from finding kids who were missed before, not from new cases appearing.

Hé lè ne et al. (2007) saw the same thing in U.S. school codes the same year. Both groups say better detection, not a new epidemic, explains the numbers. Pillay et al. (2021) extend the story to South African schools, but there the rate is only 8 per 10,000—one-tenth of Wales. The gap flags massive under-identification in poorer settings, not a real difference in kids. Root et al. (2017) add a twist: prospectively found preschoolers are milder and more often girls than clinic referrals. Together these papers show that who you look for, and how, shapes the count you get.

When you see a sudden jump in mild ASD on your caseload, ask who was missed last year, not what caused a new wave. Use the Welsh lesson: screen social skills early, especially in girls and kids first tagged with ADHD or language delay. That move shortens the two-year lag Tanidir et al. (2014) found and keeps your clinic count honest.