I don't have time for a full rebuttal, and it may well be true that medication is a worse approach to ADHD treatment than behavioral therapy. However, I want to point out by example that science is not true that "science clearly says" this.
---
1. "At the end of 14 months, core ADHD symptoms were reduced more in the children treated with stimulants than with behavioral therapy. However, at the end of three years, 'medication use was a significant marker not of beneficial outcome, but of deterioration. That is, participants using medication in the 24-to-36 month period actually showed increased symptomatology during that interval relative to those not taking medication.'"
If you actually look at the data, it shows that all participants (behavioral, combined therapy, medication only) showed marked improvement at the end of 14 months, then deterioration from the 14 month improvement at 36 months.
In conclusion, all participants showed improvement at 36 months versus their scores at the start of the study.
Additionally, they found that a number of participants who begun in the the medication group stopped using their medication after 14 months (down from 90% at 14 months to 70% at 24-36 months). Whilst, on the converse members from even the behavioral and community care group begun to use medication through out the study (up to 40% in the behavioral group. 60% in the CC group).
Their conclusion says it best: "It would be incorrect to conclude from these results that treatment makes no difference or is not worth pursuing.... medication management-may only make a persistent long term difference if it is continued with the same intensity as during the MTA's initial 14 month period".
----
2. "At the end of six years, medication use was 'associated with worse hyperactivity-impulsivity and oppositional defiant disorder symptoms,' and with greater 'overall functional impairment.'"
As compared to whom?
--------
3. "Within 21 months, 11 percent of children treated with a stimulant for ADHD developed 'psychotic symptoms.'"
ADHD is usually treated at the same age when psychosis is first diagnosed. It could be that ADHD symptoms are positively correlated with psychosis.
A close reading of the paper might actually provide that data.
Even if the side-effect does exist to some degree that does not mean that it is "clear" that the drug isn't beneficial especially considering the 89% of children that didn't experience this side effect.
Traditionally, what is done is more work into identifying those who could suffer from damaging side effects early and shunting them out of a medication program.
----
That's about all I can do now, but essentially there is no indication that stimulant medication is wholly poor performing and that early studies indicating efficacy were flawed.
What there is is evidence that side-effects exist, and preliminary investigation into the degree and distribution of those side-effects.
Additionally there is strong evidence that medication use is variable amongst participants versus other drugs for chronic conditions where drug use is sustained.