In a new JAMA Pediatrics viewpoint, Lester Liao of McGill and Eric Fombonne of Oregon Health & Science University argue that the rise in autism prevalence is not solely a detection success story. Overdiagnosis, they suggest, is part of the picture — whether through diagnostic substitution, clinicians stretching boundaries to secure services, mechanical use of standardized tools without adequate attention to confounders, or broader shifts in diagnostic concepts and thresholds, including changes in diagnostic criteria such as the shift to a spectrum model and the inclusion of co-occurring conditions.

One of the more striking datapoints they cite: close to half of children who received autism diagnoses in the community did not meet criteria when reevaluated by a research team, many of whom had higher rates of co-occurring psychiatric conditions. They also note that concurrent intellectual disability in autism has fallen from roughly 70% to 30% over four decades, which they interpret not as the disappearance of profound autism, but as a substantial shift in who now receives the diagnosis.

Their central concern is less diagnostic purity than resource allocation. If autism’s boundaries widen too far, they argue, scarce evaluation and treatment capacity can be pulled toward children with milder or more ambiguous presentations — often those with families better able to navigate the system — while children with the most significant challenges wait longer or receive less. The authors also emphasize that better-resourced families may be more able to secure diagnoses and services, further amplifying disparities. As they put it, “Overdiagnosis diverts resources from children with the most significant challenges.”

The paper is careful about incentives. Clinicians who stretch a diagnosis to unlock support are not necessarily acting in bad faith; they may be responding to a system that ties services to labels. The authors’ argument is that the diagnostic process needs to do a better job distinguishing autism from co-occurring psychiatric conditions, contextual constraints, and nonspecific behavioral features before the label is applied.

Disclosures: Dr Fombonne reported honoraria from the Association for Child and Adolescent Mental Health for editorial work, personal fees from Beech-Nut for expert testimony, and consulting fees from King & Spalding; no other disclosures were reported.

Source: JAMA Pediatrics