ADHD Medication and Academic Achievement: What Do We Really Know?

A German team of researchers performed a comprehensive search of the medical literature and identified 35randomized controlled trials (RCTs) published in English that explored this question. Participating children were between three and six years old. Children with intellectual disabilities, sensory disabilities, or specific neurological disorders such as epilepsy were excluded.

The total number of participating preschoolers was over three thousand, drawn almost exclusively from the general population, meaning these studies were not specifically evaluating effects on children with ADHD. But given that ADHD results in poorer executive functioning, evidence of the effectiveness of cognitive training would suggest it could help partially reverse such deficits.

RCTs assign participants randomly to a treatment group and a group not receiving treatment but often receiving a placebo. But RCTs themselves vary in risk of bias, depending on:

• whether the control condition was passive (i.e. waiting list or no treatment) or active/sham (an activity of similar duration and intensity to the treatment condition)
• whether the outcome was measured by subjective rating (e.g. by questionnaires, susceptible to reporting biases) or more objective neuropsychological testing;
• whether the assessment of outcome was by blinded assessors unaware of participants' treatment conditions;
• whether there was a risk of bias from participants dropping out of the trial.

After evaluating the RCTs by these criteria, the team performed a series of meta-analyses.

Combining the 23 RCTs with over 2,000 children that measured working memory, they found that cognitive training led to robust moderate improvements. Looking only at the eleven most rigorously controlled studies strengthened the effect, with moderate-to-large gains.

Twenty-six RCTs with over 2,200 children assessed inhibitory control. When pooled, they indicated a small-to-moderate improvement from cognitive training. Including only the seven most rigorously controlled studies again strengthened the effect, boosting it into the moderate effect zone.

Twelve RCTs with over 1,500 participants tested the effects of cognitive training on flexibility. When combined, they pointed to moderate gains. Looking at only the four well-controlled studies boosted the effect to strong gains. Yet here there was evidence of publication bias, so no firm conclusion can be drawn.

Only four studies with a combined total of 119 preschoolers tested the effects on ADHD ratings. The meta-analysis found a small but non-significant improvement, very likely due to insufficient sampling. As the authors noted, "some findings of the meta-analysis are limited by the insufficient number of eligible studies. Specifically, more studies are needed which use blinded assessments of subjective ratings of ADHD ... symptoms ..."

The authors concluded that their meta-analyses revealed significant, mostly medium-sized effects of the preschool interventions on core EFs [executive functions] in studies showing the low risk of bias."

Monitoring the Future is a multicohort U.S. national longitudinal study of adolescents followed up into young adulthood. 

The U.S. research team used data from this study to follow 5,034 twelfth graders over a period of six years, until they were 23 and 24 years of age.

Prescription stimulant misuse was assessed at baseline and each follow-up survey year by asking how often they used prescription stimulants without a physician’s orders. They were similarly asked about cocaine and methamphetamine use.

The study team adjusted for the following confounding variables: sex, race and ethnicity, parents’ level of education, urbanicity, U.S. region, cohort year, grade point average during high school, past-30-day cigarette use (at 18 years of age), past-2-week binge drinking (at 18), past-year marijuana use (at 18), past-year prescription opioid misuse (at 18), past-year prescription stimulant misuse (at 18), lifetime cocaine use (at 18), lifetime methamphetamine use (at 18), lifetime use of nonstimulant therapy for ADHD (at 18), and discontinued use of stimulant therapy for ADHD (at 18).

With these adjustments, they found that stimulant use for ADHD was in no way associated with subsequent cocaine use. In fact, it was associated with lesser odds of subsequent cocaine use, though the association was not statistically significant.

Likewise, they reported that stimulant use for ADHD was in no way associated with subsequent methamphetamine use.

On the other hand, those who used prescription stimulants without a physician’s orders were 2.6 times more likely to subsequently use either cocaine or methamphetamine.

The team concluded, “In this multicohort study of adolescents exposed to prescription stimulants, adolescents who used stimulant therapy for ADHD did not differ from population controls in initiation of illicit stimulant (cocaine or methamphetamine) use, which suggested a potential protective effect, given evidence of elevated illicit stimulant use among those with ADHD. In contrast, monitoring adolescents for PSM is warranted because this behavior offered a strong signal for transitioning to later cocaine or methamphetamine initiation and use during young adulthood.”

Noting that “little is known about whether school-level stimulant therapy for ADHD is associated with NUPS [nonmedical use of prescription stimulants] among US secondary school students,” a team of American researchers searched for answers in a nationally representative sample of 3,284 U.S. secondary schools with well over 150,000 high school students.

“Previous studies,” the authors continued, “have largely neglected school-level factors associated with NUPS among US secondary school students, including school size, school geographical location, school-level racial composition, school-level rates of substance use (eg, binge drinking), and school-level stimulant therapy for ADHD.”

In surveys, students were asked if they had ever taken stimulant medications for ADHD under a physician’s or health professional’s supervision, with three possible answers: no, yes but only in the past, and yes, currently. Responses for use in the past, and separately for current use, were combined and aggregated to the school level to reflect the percentage of the study body who used prescription stimulants for ADHD. 

The surveys explored NUPS by asking, “On how many occasions (if any) have you taken amphetamines or other prescription stimulant drugs on your own—that is, without a doctor telling you to take them... in your lifetime?...during the last 12 months?...during the last 30 days?” 

The study team controlled for sex, race and ethnicity, parental education, GPA, binge drinking, cigarette smoking, cannabis use, cohort year, school type, grade level, urbanicity, school size, US Census region, % of student body with low grades, % female, % with at least one parent with a college degree, % White, % binge drinking during past 2 weeks, % cigarette smoking in past 30 days, and % cannabis use during the past 30 days. The analysis also included individual-level medical use of stimulant therapy for ADHD history to estimate individual-level past-year NUPS. Finally, it included both individual-level and school-level risk factors to assess individual-level past-year NUPS.

With all these adjustments, at the individual level, both high school students presently on prescribed stimulant therapy for ADHD and those who had previously been on such prescribed therapy were more than twice as likely to engage in past-year NUPS as those who were never on prescribed stimulant medication.

Turning to the school level, in schools where 12% or more of students were on prescribed stimulant therapy for ADHD, students in general were 36% more likely to engage in past-year NUPS than in schools where none of the students were on prescribed stimulant therapy for ADHD.

This is not surprising, as it confirms that students who use prescription drugs for nonmedical often get their supply from fellow students who are prescribed those drugs.

While at the individual level, binge drinking, cigarette smoking, and cannabis use were strong predictors of NUPS, at the whole-school level they had no significant effect. A poor grade point average mildly increased risk in the individual, but high percentages of students with low grades had no effect on peer NUPS. Race and ethnicity made a difference at the individual level (NUPS significantly more likely among White students than Blacks and Hispanics), but made no difference at the school level.

The team concluded, “These findings suggest that school-level stimulant therapy for ADHD and other school-level risk factors were significantly associated with NUPS and should be accounted for in risk-reduction strategies and prevention efforts.”