Stimulants beat behaviour therapy for ADHD at 14 months. By 8 years the gap was gone.

For the core symptoms of childhood ADHD, stimulant medication, carefully managed, clearly outperformed behaviour therapy and ordinary community care in the largest trial ever run — over its first 14 months. But when the same trial followed those children for years, the advantage faded: by 6 to 8 years, the groups that had been randomly assigned to different treatments no longer differed on symptoms or on real-world outcomes like grades and arrests [s1] [s2]. Both findings are true, and holding them together is the honest answer to what stimulants do for children.

The trial that set the standard

The Multimodal Treatment Study of Children with ADHD (MTA) randomly assigned 579 children with ADHD Combined Type, aged 7 to 9.9 years, to one of four strategies for 14 months: medication management (careful titration followed by monthly visits); intensive behavioural treatment involving parents, school, and child; the two combined; or standard community care, in which families were referred to whatever local providers they chose [s1].

All four groups improved substantially over the 14 months [s1]. But the differences between them were clear. For most ADHD symptoms, the children in the medication-management and combined groups improved significantly more than those in intensive behavioural treatment or community care [s1]. Medication management and combined treatment did not differ significantly from each other on core symptoms — meaning adding intensive behaviour therapy on top of well-managed medication produced no significant additional gain for the ADHD symptoms themselves, though combined treatment showed modest advantages for some non-ADHD outcomes such as oppositional and aggressive behaviour, social skills, and reading [s1].

The most quietly important result concerned community care. The trial's carefully managed medication was superior to routine community treatment even though two-thirds of the community-care children were also taking medication during the study [s1]. The lesson was not simply "medication versus none" but that how medication is managed — the titration and monitoring — mattered as much as the drug. This trial is the empirical backbone of the guideline position that stimulants are an effective short-term treatment for childhood ADHD.

What happened when the children grew up

The MTA's investigators kept following the children, and the long-term results complicate the short-term triumph. In the 8-year follow-up, covering 436 of the original participants, the researchers reported that "in nearly every analysis, the originally randomized treatment groups did not differ significantly" — not on symptom measures, and not on newly examined real-world outcomes including grades earned in school, arrests, and psychiatric hospitalisations [s2].

Two details sharpen the point. First, medication use fell by 62% after the controlled 14-month phase ended, as families made their own choices, and statistically adjusting for that did not change the finding [s2]. Second, what did predict a child's later functioning was not which treatment they had been assigned but their ADHD symptom trajectory over the first three years, which predicted 55% of the outcomes examined [s2]. In the authors' words, "type or intensity of 14 months of treatment for ADHD in childhood ... does not predict functioning 6 to 8 years later" [s2].

The finding that gets lost

Alongside the treatment comparison, the follow-up carried a sobering observation about ADHD itself. Compared with a local group of children without ADHD, the MTA participants fared worse on 91% of the variables tested by adolescence [s2]. Despite the real symptom improvement all groups showed during treatment, children with combined-type ADHD still exhibited significant impairment years later as a group [s2].

That is a statement about the disorder's persistence, not a verdict against treatment. It says that 14 months of any intervention in childhood, however effective at the time, was not enough to alter the longer arc — which is why the authors called for treatment approaches aimed specifically at adolescent impairment [s2].

How to read this without overreaching

The evidence supports a layered conclusion that resists a simple headline. Stimulant medication, well managed, produces clear short-term reductions in ADHD symptoms, larger than behaviour therapy or routine care, and that is the strongest, most replicated finding [s1]. What the MTA does not establish is that any one strategy delivers superior functioning years later; by 6 to 8 years the randomised advantage had disappeared, and early symptom course, not treatment assignment, tracked with long-term outcome [s2].

Several caveats bound both claims. After the first 14 months the trial was no longer randomised — families chose their own care — so the long-term comparison is weaker than the short-term one, and confounding creeps back in. And the study was of school-aged children with combined-type ADHD, not of every age or presentation.

This article is informational and is not medical advice, and it does not address medication choices, doses, or the diagnosis of any individual child. Decisions about diagnosing or treating ADHD belong with a child, their family, and a qualified clinician.