Early signs of autism in toddlers: a follow-up study in the Danish National Birth Cohort.

Lemcke et al. (2013) asked moms in Denmark's national birth cohort about early red flags. They looked at 6- and 18-month-old babies who later got an ASD or ID diagnosis. Moms filled out a short checklist about language, play, and social habits.

The early signs were weak and fuzzy. Items like 'no babble' or 'poor eye contact' did not clearly split babies who would later be diagnosed from those who would not. Predictive accuracy stayed low at both ages.

Goodwin et al. (2019) extends this work. They showed that even kids diagnosed at school age had almost the same early red flags. Social gaps were the clearest hint, but still subtle. Pilgrim et al. (2000) seems to disagree at first glance. Their DAISI parent interviews found big social-engagement gaps in infants later diagnosed with ASD. The gap closes when you see the samples: C et al. studied clinic families already worried about autism, while Sanne looked at a general population where most kids are typical. Levinson et al. (2020) used the ITSEA at 18 months in high-risk siblings and also found only modest accuracy, matching Sanne's low numbers. Pandey et al. (2008) adds that M-CHAT scores at 18 months miss more cases than at 24 months, so timing matters as much as the tool.

Do not rely on a single parent checklist at 6 or 18 months to rule ASD in or out. Layer your screening: combine parent report with direct observation, grow the data at 24 months, and track social milestones over time. If a toddler shows soft social delays, schedule a follow-up rather than a full evaluation wait-list.