Science1 publisher2 min readPublished
One-third fewer children received an oppositional defiant disorder diagnosis by 2021
A 2024 study of more than 13 million records measured the drop without explaining it. The account of what replaced the label comes from one pediatrician and a phone call about pathological demand avoidance.
The Scientist · Science desk
What happened
- Ramin Mojtabai and Mark Olfson examined more than 13 million records of children aged 4 to 17 from 2013 to 2021 for a 2024 paper in the Journal of the American Academy of Child and Adolescent Psychiatry.
- Diagnoses of oppositional defiant disorder and of conduct disorder each declined by one-third over the period, while rates of anxiety, depression and PTSD in children markedly increased.
- In 2013 the American Psychiatric Association added disruptive mood dysregulation disorder to the Diagnostic and Statistical Manual as an alternative to the pediatric bipolar diagnosis, keeping many of the same criteria.
- The account of what has replaced ODD begins with a mother telephoning a Bay Area pediatrician about two years ago to ask whether her son had pathological demand avoidance, an acronym unfamiliar to the pediatrician.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint No amount of extra records settles the question the column raises: a prevalence count lumps children who changed in with clinicians who changed their vocabulary.
- precedent The 2013 DMDD addition set the template for how a childhood diagnosis gets displaced: a label circulating between parents cannot absorb the cases, but a competing label with a place in the manual can.
- exposure Which label a child ends up with decides what gets tried first, since the bipolar diagnosis came with psychiatric medication and serious side effects while DMDD came with nondrug strategies.
A diagnosis rate counts what clinicians wrote down. It falls when fewer children meet the criteria, when clinicians stop reaching for a code, or when a different code becomes available for the same child, and prevalence figures of the kind Ramin Mojtabai and Mark Olfson published cannot tell those three apart [1].
The third route is visible in the bipolar numbers. Pediatric bipolar had climbed 40-fold in the 20 years before the study window [4], pushed as an explanation for acting-out behavior by the leading child psychiatrist in the United States at the time and by his colleagues at Harvard, who held it could be diagnosed in children as young as 2 [5]. Once the alternative label existed, the count fell eight-fold, the largest decline in the study [3].
Multiply the two figures as the column gives them: 40 divided by 8 leaves the rate about five times its pre-boom level [1]. The two numbers may not rest on the same population, so five is indicative.
ODD moved much less in the same terms. A one-third decline is a 1.5-fold decline, roughly a fifth as steep as bipolar's [2]. Mojtabai and Olfson recorded the drop and stopped there [10].
The interpretation comes from the author of a STAT opinion column [19], a behavioral and developmental pediatrician with 46 years in a Bay Area private practice who had noticed ODD going missing from children's neuropsychiatric evaluation reports over the past decade [12]. "It's easier to catch an uptick than an absence," the pediatrician wrote [13]. ODD was once among the most common behavioral diagnoses in child psychiatry [11].
The evidence offered for what took its place is a telephone call. The mother asked the pediatrician to check her son for PDA, an acronym he knew only as patent ductus arteriosus [14]. "My son won't do anything I ask," she said [15]. The column calls PDA "the latest and perhaps most extreme shift in how parents and professionals view children's noncompliance" [16].
That is one case, and the column does not report how often PDA is diagnosed or whether the children now described that way would have been coded ODD in 2013. On the simplest competing explanation, that children are complying more, the author wrote: "I'd like to believe that children are behaving better for their parents, but I doubt that's the case" [17].
What to watch
- Whether anyone publishes a prevalence figure for pathological demand avoidance in the same kind of administrative records where ODD was counted.
- Whether a study follows individual children across codes from 2013 to 2021. That design would test substitution directly.
- Whether a future DSM revision adds a category that absorbs the noncompliance ODD used to describe, as DMDD did for pediatric bipolar.