How do I know if my child's ADHD medication is actually helping?

Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the child ADHD research overview.

Short answer. You cannot reliably evaluate stimulant medication from memory. The MTA Cooperative Group (1999) established the strongest evidence base in child psychiatry for stimulant benefit at the population level, but at the household level the question "is this dose helping this child at this stage" is an observation question that requires structured weekly data. Track 3–5 specific markers — on-task minutes during homework, number of morning prompts, meltdowns per week, sleep onset, appetite — for at least six weeks before any medication decision. Bring the data to the prescriber. The decision then moves from "I think it's helping" to "here is what changed and when."

What the research says about stimulant evaluation

Stimulant medication has the largest single evidence base in child psychiatry. The MTA Cooperative Group's 1999 trial established that medication management — when carefully titrated — outperformed community-care medication and combined well with behavioural treatment for functional outcomes. The AAP Clinical Practice Guideline (Wolraich et al., 2019) endorses stimulants as first-line pharmacological treatment for ADHD in children 6 and older, alongside behavioural parent training.

What the same literature is also explicit about: titration is a data-driven process. The MTA medication algorithm involved structured weekly ratings from parents and teachers, dose adjustments based on those ratings, and side-effect monitoring against a checklist. The "carefully titrated" arm outperformed community care in part because community care was titrated by impression. Pelham & Fabiano (2008) reach the same conclusion from the behavioural side: parent and teacher rating scales detect medication response that parental impression misses.

The implication is uncomfortable but important: the evidence base for stimulant benefit assumes structured measurement. Without it, you are running an intervention whose efficacy has been demonstrated under conditions you are not reproducing.

What parents are actually noticing

When parents ask "is the medication helping?" they almost always mean one of three things:

1. A specific behaviour has not changed. "He still melts down at homework" — but they have no data on whether the meltdowns are shorter, less frequent, or less intense. 2. A side effect is visible and the benefit is not. Reduced appetite, sleep onset delay, or emotional flatness are all observable in real time. Improvements in attention are not. 3. The first week was dramatic and then things plateaued. This is the predicted pattern, not evidence of failure — but it is widely misread.

In each case the underlying issue is the same: the cost of medication is salient day to day; the benefit shows up as a frequency shift that is invisible without measurement.

The research-backed markers to track

Pelham & Fabiano (2008) and the MTA medication algorithm both rely on a small set of ratings that detect stimulant response. A practical version for parents:

Marker 1: On-task minutes during homework

Time from sitting down to first off-task break. Track to the minute, weekly. This is the single marker most sensitive to stimulant dose response in school-aged children.

Marker 2: Number of prompts to leave the house in the morning

How many separate verbal prompts before the child is dressed, fed, has shoes on, and is at the door. Count, weekly average.

Marker 3: Meltdowns per week

A meltdown is operationalised as an emotional escalation lasting more than 5 minutes that requires adult intervention. Count, weekly.

Marker 4: Sleep onset time

Minutes from lights-out to actually asleep. Stimulants commonly delay sleep onset; this is the most reliable side-effect marker.

Marker 5: Appetite (lunch eaten / not eaten)

Binary or three-point rating from school lunchbox returns. Reduced appetite is the most common stimulant side effect.

These five markers together cover both the targeted benefit (markers 1–3) and the most common side effects (markers 4–5). Six weeks of weekly data on these is enough to support a medication decision; less than that is impression dressed up as data.

What the research suggests about timing

  • Initial titration window: 4–6 weeks to find a working dose with acceptable side effects (AAP, 2019). Within-week variability is high; weekly aggregates are more interpretable than day-to-day impressions.
  • Stable evaluation: another 6–8 weeks at the chosen dose to evaluate sustained benefit. The MTA trial detected medication-versus-community-care differences across a 14-month window with monthly structured ratings.
  • Re-evaluation: at minimum every 6 months during growth, more often during rapid growth phases. Doses that were correct at 7 are frequently subtherapeutic at 10. Bring weekly data to every review.

What does not reliably indicate medication response

  • The first three days. Initial effect is often dramatic but does not predict sustained response.
  • Today's homework session. Single-event evaluation is the dominant trap; ADHD behaviour is genuinely wildly variable day-to-day even on a working dose.
  • The teacher's "he seems better." Useful as a corroboration but not as primary data; teacher ratings are far more reliable when collected on a structured form (e.g. SNAP-IV, Vanderbilt) than as impressions.
  • Whether the child says it's helping. Children with ADHD are not reliable narrators of their own attention.
  • Whether you, the parent, feel less exhausted. Parent fatigue is driven by many things; it is a poor proxy for child response.

What to bring to the prescriber

The AAP and AACAP guidelines both treat the medication review as a structured, data-driven encounter. A useful packet:

1. The five weekly markers in a simple grid covering the last 6 weeks. 2. A side-effect note — sleep onset, appetite, mood, tics, headaches, cardiovascular symptoms. 3. One concrete question. "Should we adjust the dose?" "Should we add an afternoon booster?" "Should we switch from methylphenidate to amphetamine class?" Specific questions get specific answers; "is it working" gets reassurance.

Most prescribing decisions improve substantially when the parent arrives with structured data instead of impressions. This is the single most useful preparation any parent can do for a medication review, and it is the missing instrumentation that separates community-care titration from MTA-grade titration.

Related questions

References

  • MTA Cooperative Group. (1999). A 14-month randomized clinical trial of treatment strategies for attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 56(12), 1073–1086.
  • Wolraich, M. L., et al. (2019). Clinical Practice Guideline for the Diagnosis, Evaluation, and Treatment of ADHD in Children and Adolescents. American Academy of Pediatrics. Pediatrics, 144(4).
  • Pelham, W. E., & Fabiano, G. A. (2008). Evidence-based psychosocial treatments for ADHD. Journal of Clinical Child & Adolescent Psychology, 37(1), 184–214.
  • AACAP. Practice Parameter for the Assessment and Treatment of Children and Adolescents With Attention-Deficit/Hyperactivity Disorder.

---

Unseen Progress publishes long-form caregiver research. See the full child ADHD research overview for the complete framework.