The DSM Wasn't Built for Children. We've Been Using the Wrong Map.
The DSM Wasn't Built for Children. We've Been Using the Wrong Map.
This is the kind of argument that gets nods in one hallway conversation and raised eyebrows in the next. I think it deserves both.
I've said versions of this sentence in supervision, in case conferences, and, more than once, at a conference bar with colleagues who pushed back hard enough that I went back and reread the sources before saying it again in public: the DSM was never built for children. It was built for adults, and children were added later, mostly by adjusting adult criteria downward rather than by starting from how children actually develop.
I expect this piece to land differently depending on who's reading it. Some clinicians will recognize this immediately — it's the thing they've felt in the room but rarely said out loud. Others, including colleagues I deeply respect, will push back, and reasonably so. I want to make the case anyway, and let both reactions sit side by side.
A Map Drawn for a Different Territory
The first DSM, published in 1952, was assembled largely from adult psychiatric hospital classification systems. Childhood-specific diagnostic categories weren't meaningfully built out until DSM-III in 1980 — nearly three decades later — and even then, much of the framework was adult psychopathology with age brackets attached, rather than a system built from the ground up around how a developing brain and personality actually present distress.
This isn't a fringe complaint. It shows up directly in the clinical literature. Reviews of DSM-5's approach to childhood and adolescent disorders have documented specific, substantive controversies — including how consolidating categories like autism spectrum diagnoses affected clinicians' and policymakers' ability to use meaningful, specific distinctions that mattered in practice.
Four Places the Adult-First Model Strains
-
01
Symptoms That Look Different at Different Ages
ADHD is the clearest example: hyperactivity and impulsivity dominate in early childhood, attention difficulties become more visible as academic demands increase later, and by adulthood the presentation shifts again toward internal restlessness and executive dysfunction. A single, largely static criteria set has to stretch to cover presentations that genuinely change shape across development.
-
02
Diagnoses Built From Adult Trauma Responses
Complex trauma presentations in children — the dysregulation that follows chronic early maltreatment — don't map cleanly onto PTSD criteria that were substantially shaped by adult trauma research. A specific childhood framework, Developmental Trauma Disorder, was proposed precisely because clinicians working with traumatized children found the adult-derived category didn't fit what they were seeing, though it was ultimately not adopted into the manual.
-
03
Categories Instead of Continuums
A growing body of research argues that psychopathology in children doesn't sort neatly into separate categorical boxes the way the DSM's structure assumes — it behaves more like overlapping dimensions of distress. Large-scale studies mapping these dimensions in children as young as nine and ten have found meaningful patterns that a category-first system tends to flatten.
-
04
An Age-of-Onset Line That Cuts Against Real Presentation
Criteria that require symptoms to be present before a specific childhood age — as with ADHD's before-age-12 requirement — assume reliable retrospective memory and available corroborating adults, assumptions that don't always hold and that researchers have specifically flagged as a source of diagnostic strain.
A diagnostic manual built primarily around what pathology looks like in a finished brain will always be translating, not describing, when it's applied to one that's still being built.
— Dr. Farrah Laviolette, MD
The Strongest Version of the Counter-Argument
I don't want to present this as a settled case. Colleagues I trust have real, substantive reasons to push back on this framing, and the disagreement is worth showing plainly rather than smoothing over.
The Manual Strains at the Seams
Childhood categories were retrofitted onto an adult-built structure, key presentations shift across developmental stages in ways a largely static criteria set struggles to capture, and researchers have proposed entirely separate frameworks — like Developmental Trauma Disorder and dimensional models — precisely because the fit felt wrong to clinicians in the room.
The Manual Has Evolved, Imperfectly but Genuinely
DSM-5 did add developmentally-specific language, age-based symptom thresholds, and a dedicated framework for early childhood exists separately in the DC:0-5 manual. A flawed, evolving tool used by an experienced clinician who understands its limits is not the same as a tool that's simply wrong — and "the map is imperfect" is a different claim than "the map is useless."
Both of these things can be true at once, and I think that's the more honest place to land than picking a side for the sake of a clean argument. The manual has genuinely improved. It has also never been rebuilt from a developmental foundation up — and those two facts sit uncomfortably next to each other.
If dimensional models of childhood psychopathology are showing real promise in research settings, why has the primary clinical tool most of us use every day not moved further in that direction? Is it inertia, insurance and billing infrastructure built around categorical codes, genuine clinical utility, or some combination none of us want to fully own?
Why the Distinction Isn't Just Academic
This isn't an abstract debate for the people it affects most. A diagnostic system that treats childhood symptoms as a scaled-down version of adult pathology can miss presentations that are genuinely different at different developmental stages — not milder or earlier versions of an adult picture, but distinct expressions shaped by a brain and a life still very much in progress.
That has real consequences: children who don't fit the adult-derived mold cleanly, misapplied age-of-onset criteria that exclude kids who present differently than the manual expects, and treatment plans built around a category that was never quite describing what's actually in front of the clinician.
A Better Map Isn't a Rejection of the Old One
I'm not arguing to throw out the DSM. It remains useful — for communication between clinicians, for research, for the practical realities of insurance and school accommodations that depend on a recognized diagnosis. What I am arguing is that we should hold it the way any clinician should hold a tool built for a different purpose than the one it's being asked to serve: with real usefulness, and real skepticism, at the same time.
I expect some of you reading this will feel like I've finally said something you've been circling for years. I expect others will feel I've overstated a genuine, gradual evolution as a flat failure. Both reactions are reasonable. I'd rather put the disagreement on the table than pretend it doesn't exist.
Curious How This Shapes the Way We Evaluate Kids?
At Cranbrook Psychiatric Group, every evaluation holds diagnostic categories alongside a genuinely developmental lens — because a child is never just a smaller version of an adult case. We see families via telehealth, from anywhere in Michigan.
- Thomas, C.R. (2014). DSM-5 and Neurodevelopmental and Other Disorders of Childhood and Adolescence. Journal of the American Academy of Psychiatry and the Law, 42(2), 165–172.
- van der Kolk, B.A. et al. (2009). Proposal for Developmental Trauma Disorder. National Child Traumatic Stress Network.
- Toward Dimensional Psychiatry in Youth: A Data-Driven Analysis of Transdiagnostic Internalizing Symptoms in Childhood and Adolescence. Healthy Brain Network study.
- Delineating and Validating Higher-Order Dimensions of Psychopathology in the Adolescent Brain Cognitive Development (ABCD) Study.
- Pickersgill, M.D. (2013). Debating DSM-5: Diagnosis and the Sociology of Critique. Journal of Medical Ethics.
- Zero to Three. Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood (DC:0-5).
