The most comprehensive ADHD review of the decade
In 2026 the BMJ published a State of the Art Review titled “Advances in the management of ADHD in children and adolescents,” authored by Philippe Robaey, Maria Rogers and Russell Schachar across institutions in Ottawa and Toronto. State of the Art Reviews are commissioned for their relevance to academics and specialists internationally — this is the BMJ telling clinicians worldwide what the current evidence says. The review synthesises work published between 2019 and 2025, drawn from PubMed, PsycINFO, Embase, Web of Science and the Cochrane Central Register. It is rigorous, current, and comprehensive. It is also a near-perfect specimen of the apparatus describing itself.
The review’s conceptual frame is a dimensional, pleiotropic model of ADHD — characterised, in the authors’ words, by heterogeneity, developmental fluctuation, and frequent comorbidity. ADHD is positioned as a chronic, fluctuating condition requiring sustained and individualised care. Diagnosis favours semi-structured interviews over rating scales. Neuroimaging and biological tests are acknowledged to offer limited diagnostic utility. The catalogue of interventions is exhaustive: behavioural parent training, school-focused interventions, cognitive behavioural therapy, lifestyle interventions targeting physical activity and sleep, cognitive training, neurofeedback, brain stimulation, and pharmacology.
The verdicts on each intervention are candid, and this candour is the review’s genuine strength. Psychosocial interventions produce “small but meaningful improvements in functioning.” Lifestyle interventions show “modest adjunctive benefits.” Cognitive training “does not generalise beyond trained tasks.” Neurofeedback effects are “predominantly non-specific and contextual.” Digital tools are “expanding, but few are rigorously validated.” Pharmacological treatment “remains central,” with stimulants showing robust short-term efficacy. The review does not oversell. It reports the evidence as the evidence is.
None of this is wrong on its own terms. The methodology is sound. The synthesis is fair. The clinical recommendations follow from the literature. A general clinician reading this review will come away with an accurate picture of what mainstream ADHD management in 2026 can and cannot do. The problem is not in what the review gets wrong. The problem is in what the review cannot see — and what it claims to see while not seeing it.
Neurodiversity gets two mentions in sixty pages
Across the entire review, the word “neurodiversity” appears exactly twice. Once in the closing discussion, where the authors note that clinical care “must remain both pragmatic and diversity affirming,” and reference Sonuga-Barke and colleagues’ proposal to shift “from a disorder centered paradigm toward one that emphasises environments that promote positive experiences and outcomes rather than focusing on sources of brain dysfunction.” Once more in the reference list, citing that same work. “Social model” appears zero times. A strengths-based framing of the ADHD substrate appears nowhere. The paradigm shift that has been transforming how the field understands neurodivergence over the past decade is mentioned, gestured at, and immediately set back down.
The Sonuga-Barke reference is the tell. The authors cite a paradigm that proposes redesigning environments rather than treating brain dysfunction — and then proceed, across sixty pages, to catalogue methods for treating brain dysfunction. The paradigm is acknowledged as a perspective that exists, slotted in beside the others, and then structurally ignored. The review’s actual content — its diagnostic pathways, its intervention hierarchies, its care models — flows entirely from the disorder-and-management paradigm the cited work was proposing to move beyond.
It reads almost as an obligation discharged. The citation has the quality of something the authors felt required to include — the neurodiversity paradigm has become too prominent to omit entirely — while having no structural use for it within the paradigm they actually operate. You cite it because its absence would be conspicuous. You do nothing with it because there is nothing the disorder-and-management framework can do with a proposal to stop treating the substrate as the problem. The acknowledgement is the cost of remaining current; the ignoring is the cost of remaining coherent; the review pays both.
This is the demand-structure problem rendered at the level of the literature itself. The entire apparatus of “management” presupposes that the thing requiring intervention is the child’s substrate — that the ADHD substrate is the variable to be adjusted, optimised, medicated, and trained toward compliance with an environment treated as fixed. The possibility the cited neurodiversity work raises — that the environment is the variable, and the inherently human substrate is meeting a demand structure calibrated against it — is named in a single clause and then abandoned. The review cannot pursue it, because pursuing it would unmake the review. Sixty pages of management advice presuppose there is something in the child to be managed. Question that presupposition and the apparatus has nothing left to catalogue.
The claim to be “diversity affirming” is what makes the erasure remarkable rather than merely conventional. A review that simply operated inside the medical model without comment would be unremarkable — that is the default. This review reaches for the language of the paradigm it is displacing, claims its progressive mantle, and proceeds as though claiming the language were the same as doing the work. It is not (progress and progressive are opposite sides of a coin after all). Affirming neurodivergence means questioning whether the demand structure that produces the “impairment” is the thing requiring redesign. The review never asks. It asks only how to manage the child more precisely.
The interventions work because raters say they do
The review contains an admission that deserves far more attention than its understated placement gives it. The psychosocial interventions — the behavioural parent training, the school-focused programmes — produce outcomes that are, in the authors’ own words, “strongly shaped by rater, expectancy, and context.” Elsewhere the review states that effects are “rater and context dependent,” with “limited evidence for core symptom change under rigorous designs,” and that the benefits are “strongest for proximal functional outcomes” like organisation and classroom behaviour. Restricted to well-blinded, high-quality randomised controlled trials, the effects shrink further.
Read plainly, this means the interventions appear to work largely because the adults rating the child’s behaviour — parents, teachers — rate it more favourably after the intervention. The child has not been objectively measured as changed. The raters’ perception has changed, or their expectations have, or the context has shifted. When you remove the rater bias by blinding the trial properly, much of the effect disappears. The review names this methodological collapse clearly and then recommends the interventions anyway, classifying behavioural parent training and school-focused interventions among the primary psychosocial approaches.
The structural reading is unavoidable once the admission is in view. If an intervention’s measured benefit is the adult rating the child more favourably, the intervention is not changing the substrate — it is changing the adult’s relationship to the substrate, or training the child to perform compliance the rater then rewards with a better score. Either way, what is being optimised is the fit between the child’s observable behaviour and the rater’s expectation. That is the demand structure adjusting its own measurement, not the substrate being “healed” or normalised. The apparatus measures compliance, calls the improved compliance-rating a treatment effect, and recommends the treatment.
What "managing" ADHD is actually for
The review tells us, in its own concluding words, what the psychosocial apparatus is for. These interventions “primarily target functional impairments, family burden, and contextual adaptation rather than symptom reduction alone.” Three phrases, each of which repays structural translation. “Functional impairment” is the gap between what the child’s substrate produces and what the demand structure requires — named as a property of the child rather than of the mismatch. “Family burden” is the cost to adults of living with a child whose substrate does not comply with the structure — the intervention exists partly to reduce that cost. “Contextual adaptation” is the substrate being adjusted to fit the context, with the direction of adaptation running one way only: the child adapts to the environment, never the environment to the child.
This is the entire logic of “management” laid bare in three phrases. The substrate is the thing to be managed; the environment is the fixed point (“the round hole”) around which management organises. The child is trained, medicated, and rated toward compliance with a demand structure that is never itself the object of intervention nor analysis. Drug treatment, the review concludes, “remains the most effective intervention for the core ADHD symptoms” — and on the apparatus’s own terms this is true, because pharmacology is the most direct route to making the substrate produce behaviour the demand structure rewards: stimulants raise the child’s baseline dopamine (the drive to pursue) and norepinephrine (the arousal to engage) levels, producing the engaged, compliant behaviour the structure was calibrated to expect.
The BMJ published the most comprehensive ADHD management review of the decade, and the paradigm shift transforming the field appears twice, in passing, in a document that claims to be diversity-affirming. This is not an oversight by careless authors — the review is the opposite of careless. It is the apparatus declining, with full rigour and complete citations, to see the one thing that would unmake it — the thing anyone who has spent time in the neurodiversity movement already knows: the substrate was never what required managing. The demand structure was. A review that could see that would not be a review of how to manage ADHD. It would be a review of why we keep trying to.
Citations
Robaey, P., Rogers, M. A. & Schachar, R. J. (2026) — Advances in the management of ADHD in children and adolescents — BMJ 393:e082507
Sonuga-Barke, E. & Thapar, A. (2021) — The neurodiversity concept: is it helpful for clinicians and scientists? — The Lancet Psychiatry 8:559–561
Sonuga-Barke, E. J. S. (2023) — Translational science: Outlining a neurodevelopmental science framework from a ‘neurodiversity’ perspective — Journal of Child Psychology and Psychiatry 64:1405–1408
