Comparison of Adulthood Outcomes in Autism Spectrum Disorder With and Without Regression: A Population-Based Birth Cohort Study.

This Japanese study followed 168 people with autism from childhood to adulthood, comparing those who experienced early skill loss (regression) with those who didn't. They looked at IQ scores, social abilities, and epilepsy rates. Surprisingly, they found no significant differences between the two groups in any area measured. This suggests that experiencing early regression may not predict worse outcomes in adulthood as previously thought.

This population-based cohort study from Japan examined long-term outcomes in 168 individuals with autism spectrum disorder, comparing those with and without early developmental regression. Participants were born between 1988-1996 and diagnosed with ASD by age seven. The study measured four key outcomes: IQ at age five, IQ in adulthood, epilepsy incidence, and composite social ability scores. Results showed no significant differences between groups across any outcome measures.

Effect sizes were small, with confidence intervals suggesting that if regression does impact long-term outcomes, the effect is likely small to intermediate. This challenges previous assumptions that early regression necessarily predicts poorer adult outcomes in autism.

Results suggest that experiencing early developmental regression in autism may not necessarily predict poorer long-term outcomes. This information may help reduce parental anxiety and inform more optimistic prognosis discussions. Clinicians should avoid making assumptions about adult outcomes based solely on early regression history.

Single geographic location (northern Yokohama) may limit generalizability. Study design not clearly specified in abstract. Long follow-up period may involve participant attrition. Classification of regression based on community screening records rather than standardized clinical assessment.

The long-term outcomes of regression in autism spectrum disorder (ASD) remain unclear. Previous evidence suggests that autistic individuals with regression have poorer adulthood outcomes across various indices than those without regression. We compared two groups-those with and without regression in ASD-among 168 participants from a population-based birth cohort study. These individuals were born in northern Yokohama, Japan, between 1988 and 1996 and were diagnosed with ASD by age seven.

Participants were classified into groups based on real-time records from a community-oriented mass screening system. This study focused on four outcomes: IQ at age five, IQ in adulthood, incidence of epilepsy, and composite social ability score. None of the outcomes showed significant differences between the two groups. The standardized effect size (where a positive value favors the non-regression group) was 0.06 (95% CI: -0.08 to 0.21) for the composite social ability score, -0.16 (95% CI: -0.55 to 0.22) for IQ at age five, and -0.15 (95% CI: -0.62 to 0.33) for IQ in adulthood.

The risk ratio of epilepsy in the regression group compared to the non-regression group was 0.59 (95% CI: 0.22 to 1.5). No significant differences were observed in any of the four outcomes between autistic individuals with or without regression. The confidence intervals of the effect sizes indicate that the impact of regression on IQ and psychosocial adaptation in adulthood, if present, is likely to be small to intermediate.