About One in Three Children Can Stop ADHD Medication Without Relapse. Almost Nobody Checks.

If your child has been on the same ADHD prescription for two years, has anyone actually tested whether he still needs it?

The short answer

A 2021 review in Frontiers in Psychiatry found that roughly 30 percent of young people on stimulant medication for ADHD tolerated stopping it without relapse or worsening symptoms. Clinical guidelines support periodically testing whether medication is still needed. In practice that test is rarely run, so a child can stay on a prescription for years without anyone checking.

What the researchers studied, and what they found

Lohr and colleagues reviewed the published evidence on intentionally stopping stimulants in children and teenagers. Not stopping because a family gave up, or because a prescription lapsed, but stopping on purpose to find out whether the medication was still doing something.

Their central number: about 30 percent tolerated discontinuation without relapse or symptom deterioration. In one observational study, 11 of 42 children showed no clinical deterioration at all when medication was stopped.

Roughly one in three or four medicated children may no longer need the medication, and the only way anyone finds out is if somebody deliberately checks.

The review also described how a proper test is run. A one week drug free trial, with the child assessed in more than one setting, because rebound at home and rebound at school do not look the same. If that week is inconclusive, a placebo controlled trial over three to four weeks. Most relapse shows up within two weeks, which means a family does not wait long to get an answer.

The review supports considering these trials annually, lasting several days to a week.

Who is most likely to succeed, and who is not

Answer first: the strongest predictor in the data was being older. Youth above a median age of 13.8 years showed no worsening when switched to placebo.

That matters, and it cuts against younger children. If you are the parent of a second grader, the evidence favouring successful discontinuation was largely generated in kids considerably older than yours. That is not a reason to skip the conversation. It is a reason to have realistic expectations going into it.

The other predictors were:

• No recent need for a dose increase
• No deterioration when doses were missed, or during previous drug holidays
• Symptom free for longer than a year

The one predictor a parent can observe without any help from a clinic is whether a missed dose or a skipped weekend actually made things worse.

If your child has gone Saturday and Sunday without medication and nothing fell apart, that is real information. Write it down. It is the single most useful thing you can bring to the prescribing physician.

The part about school, which is not what most parents are told

This is where the research gets uncomfortable, and where most articles stop short.

The Multimodal Treatment of ADHD study, funded by the National Institute of Mental Health, ran from 1994 to 2014 and is the largest trial of its kind. Children were randomly assigned to medication management, behavioural therapy, both, or routine community care.

By 36 months, none of the randomly assigned groups differed significantly on any of five clinical and functional outcomes. Reading achievement was one of them.

The early advantage did not hold. Once the study interventions ended and medication was used less intensively, the differences between the groups dissipated.

Parents are routinely told that medication will help with school. The largest long term trial found the between group difference in reading achievement was gone by three years.

Two things are worth saying carefully here. Treatment did improve long term outcomes compared with untreated ADHD, though generally not all the way to typical levels. And for the outcomes that sit alongside the core symptoms, including academics, medication and behavioural treatment came out equivalent, with combined treatment looking stronger than either one alone.

So the honest version is not "medication does not help him learn." It is that medication alone is not the thing that carries academic gains forward, and the trial that looked hardest for that effect did not find it at three years.

What the guidelines already say, and what usually gets skipped

The American Academy of Pediatrics clinical practice guideline for children aged six to twelve does not recommend medication on its own. It recommends FDA approved medication along with parent training in behaviour management, behavioural classroom interventions, or both. Both arms carry Grade A evidence and a strong recommendation.

The guideline also asks clinicians to reassess and document the response four to six weeks after starting a medication, using rating scales from more than one informant plus a side effect checklist, and to manage ADHD as a chronic condition rather than a solved problem.

The behavioural half of the guideline is the half that builds capability the child keeps. It is also the half most often left out.

That is the connection to everything above. Behaviour management and classroom supports are skills and structures that do not stop working when a pill wears off. They are also what makes a discontinuation trial more likely to hold rather than merely possible.

Why this is your prescriber's job and not a home experiment

Answer first: stopping is not risk free, the effects are specific, and they need someone watching who can prescribe.

Documented withdrawal and rebound effects in the review included vomiting, headaches, light sensitivity, muscle cramps, rapid weight gain, movement disorders, and psychiatric symptoms including depression and a psychotic like presentation. Acute dystonia was reported when stimulants were combined with antipsychotics.

There is also timing. A discontinuation trial run during a move, a school change, or any other period of upheaval will produce an answer you cannot trust, because you will not know what caused what.

None of this is an argument against asking. It is an argument for asking the physician who prescribes the medication, and letting them design the trial.

What this means for you

If you are wondering whether your child still needs his prescription, here is what the research supports you doing.

1. Write down what you have already observed. Missed doses, skipped weekends, school holidays. Did anything actually get worse? That is predictor data and you already have it.
2. Ask the prescribing physician directly about a planned trial off medication. Use those words. You are not asking to stop the medication, you are asking whether a reassessment is appropriate.
3. Ask whether the behavioural half of the guideline is in place. Parent training in behaviour management and classroom interventions are Grade A recommendations for this age group, not optional extras.
4. Expect to be assessed in more than one setting. Home and school separately. His teacher's observations are part of the data, not a courtesy.
5. Do not run it during a transition. Not during a move, not the first month of a new school year.

What we do not know

This is a review of existing studies, not a single large trial of discontinuation, and the studies it drew on differ in method and quality.

The 30 percent figure is a proportion across a body of research, not a prediction about any individual child. Nothing in this literature tells you which group your child is in. The strongest predictor available, older age, works against younger children specifically.

The MTA result is also frequently over read in both directions. It does not show that medication is useless, and it does not show that behavioural treatment is superior. It shows that by three years the randomly assigned groups had converged, in a study where treatment fidelity naturally dropped after the trial phase ended.

One more thing worth addressing, because parents encounter it constantly online. The claim that ADHD is caused by a damaged gut lining, and that healing it will resolve symptoms, is not supported. The evidence usually cited traces back to a single small case control study of zonulin, a marker of intestinal permeability, where the group difference was marginal and the variation within each group was nearly as large as the difference between them. A larger 2023 study specifically in children with ADHD found no difference in zonulin at all. Every study in this area is case control or cross sectional, which means none of them can establish cause. Intestinal permeability is an interesting research question. It is not a treatment target.

Reader questions

Can a child safely stop ADHD medication?
About 30 percent of medicated young people tolerated stopping without relapse in the reviewed evidence. Whether any individual child is in that group is not predictable in advance, which is why the decision and the monitoring belong to the prescribing physician.

How is a discontinuation trial actually run?
A one week drug free trial with the child assessed in more than one setting, home and school separately. If that is inconclusive, a placebo controlled trial over three to four weeks. Most relapse appears within two weeks.

Which children are most likely to succeed at stopping?
Older children, those with no recent dose increase, those who did not deteriorate during missed doses or previous drug holidays, and those symptom free for longer than a year.

Does ADHD medication improve school performance long term?
The MTA study found no significant difference between randomly assigned groups on five outcomes including reading achievement by 36 months. Medication and behavioural treatment came out equivalent on academic outcomes, with combined treatment stronger than either alone.

Is a damaged gut lining the cause of ADHD?
No. The zonulin evidence traces to a small case control study with a marginal difference, and a larger 2023 study in children with ADHD found no difference at all. All of this research is cross sectional, so none of it can establish cause.

Should I stop my child's medication on my own?
No. Documented withdrawal and rebound effects include vomiting, headaches, movement disorders, and psychiatric symptoms. A trial needs a prescriber designing and watching it.

Citations

1. Lohr WD, Wanta JW, Baker M, Grudnikoff E, Morgan W, Chhabra D, Lee T. Intentional Discontinuation of Psychostimulants Used to Treat ADHD in Youth: A Review and Analysis. Frontiers in Psychiatry. 2021 Apr 20;12:642798.
2. Wolraich ML, et al. Clinical Practice Guideline for the Diagnosis, Evaluation, and Treatment of Attention-Deficit/Hyperactivity Disorder in Children and Adolescents. Pediatrics. 2019;144(4):e20192528.
3. National Institute of Mental Health. Multimodal Treatment of Attention Deficit Hyperactivity Disorder (MTA) Study. 1994 to 2014.
4. Asbjornsdottir B, et al. Zonulin-Dependent Intestinal Permeability in Children Diagnosed with Mental Disorders: A Systematic Review and Meta-Analysis. Nutrients. 2020 Jul 3;12(7):1982.
5. Lee SY, Li SC, Yang CY, Kuo HC, Chou WJ, Wang LJ. Gut Leakage Markers and Cognitive Functions in Patients with Attention-Deficit/Hyperactivity Disorder. Children (Basel). 2023 Mar 5;10(3):513.

This article is educational and is not medical advice. Decisions about starting, changing, or stopping any prescription belong to you and the physician who prescribes it. I am a chiropractor. I do not prescribe or manage ADHD medication, and nothing here should be read as a recommendation to alter your child's treatment.