Objectives: We estimate an age-period-cohort effect model for autism incidence and additionally examine variation in effects by autism severity.
Methods: Data were drawn from California birth records; autism diagnoses were ascertained via records from the California DDS Client Development and Evaluation Report. Generalized estimating equations were used to model age, period, and cohort effects in a repeated measures dataset that included person-years for each child from age two through age 12.
Results: Compared to those born in 1992, each successively younger cohort has significantly higher odds of an autism diagnosis than the previous cohort, controlling for age and period effects. The cohort effect observed in these data was primarily limited to high functioning children with autism; trends by cohort are less evident among low functioning children with autism. Age is strongly related to autism diagnosis, with those aged 3 at 37 times the odds of autism compared to those aged 2 (95% C.I. 31.7-43.9). We also document a significant effect of period; regardless of age and birth cohort, those observed in 1995 have 3.75 times the odds of an autism diagnosis compared to those observed in 1994 (95% C.I. 2.4-5.9).
Conclusions: Autism incidence in California exhibits a robust and linear positive cohort effect that is primarily confined to high functioning autism cases. This finding supports the hypothesis that a proportion of the autism trend is due to diagnostic drift as the signs and symptoms of autism spectrum disorders become more well-known. Period effects in these data possibly reflect changing diagnostic systems, and contribute to observed increases in autism incidence.
See more of: Epidemiology
See more of: Prevalence, Risk factors & Intervention