Autism With and Without Regression: A Two-Year Prospective Longitudinal Study in Two Population-Derived Swedish Cohorts.

The two cohorts defined and classified word loss differently, so the combined findings may not be directly comparable. Stockholm children came from clinical diagnoses, while Gothenburg children came from population screening, which makes representativeness hard to judge. Some children were excluded or did not take part, including children with very severe co-occurring neurological conditions and profound intellectual disability, families who declined, and incomplete follow-up. Genetic testing was reported only for the Stockholm group, and the Gothenburg genetic results were still in process. The study describes groups and associations. It does not show that word loss causes worse outcomes or that any treatment works.

Do not conclude that loss of previously used words proves a child has a new neurological disease or that it causes worse language and thinking. The study only compared children who had and had not lost words. It did not test a treatment or establish cause. Epilepsy was not more common in the word-loss group.