The effect of daytime napping on the athletic performance of team ball sports athletes: systematic review and meta-analysis.

Over the past two decades, the reported prevalence of autism spectrum disorder (ASD) has risen dramatically, from approximately 0.4–2 per 1,000 in the 1960s and 1970s to current estimates exceeding 1 in 32 children in the United States (–). This trajectory has been met with both justified concern and considerable methodological debate. There is broad agreement that improved awareness, expanded diagnostic criteria, and better access to services have corrected genuine historical under-ascertainment, particularly among females, verbally fluent individuals, and adults presenting for the first time (, ). Yet these welcome developments do not fully explain what is now being counted as autism, nor do they guarantee that what is being counted corresponds to the same underlying construct. The central argument of this Perspective is that the very success of diagnostic expansion now carries a risk of diagnostic dilution: the inclusion of individuals whose phenotypic profiles, developmental histories, and functional impairments may not align with the core neurodevelopmental construct that autism has historically described (). This risk is especially acute in the assessment of adolescents and adults, where differential diagnosis is most challenging and where standardized instruments show limited specificity (). Concerns about diagnostic dilution are not tantamount to denying the validity of adult-diagnosed autism or of milder presentations, nor do they dismiss the real needs of late-identified individuals. Rather, they point to the scientific imperative of diagnostic validity, ensuring that prevalence estimates, biomarker studies, genetic research, and clinical trials rest on diagnostic valid, developmentally anchored, and well-characterized samples. The discussion that follows outlines the evidence for diagnostic expansion, the specific challenges of adult diagnosis, and a constructive research agenda centered on phenotypic stratification. Empirical work has demonstrated that changes in diagnostic practices account for a substantial fraction of rising prevalence. In a landmark retrospective record review of 7,003 clients enrolled with the California Department of Developmental Services, King and Bearman estimated that 26.4% of the increased autism caseload between 1992 and 2005 was uniquely attributable to diagnostic change, specifically, to diagnostic accretion (the addition of an autism diagnosis to a pre-existing diagnosis of intellectual disability) and diagnostic substitution (the replacement of an earlier diagnosis with autism) (). The odds of an individual acquiring an autism diagnosis were significantly elevated in years coinciding with changes to DSM criteria and California-specific diagnostic guidelines (). These findings have been corroborated by other studies: as Simonoff noted, broadening diagnostic boundaries, improved identification, diagnostic substitution, and service access all contribute to rising rates, although no evidence suggests that people without significant impairment are currently being diagnosed (). Blenner and Augustyn emphasized that the ADDM network’s methodology, while useful for surveillance, is not designed to determine whether autism is truly increasing in the population, only that we are counting differently ().