Amphetamines, of which Adderall is a commonly used agent, are harmful to the brain in animal studies, repeatedly and consistently. They lead to neuronal death, especially in the hippocampus, which is an important part of the brain associated with short-term memory, but also related to depression. These findings are repeated in many animal studies, mostly in rats, which are summarized here. These studies include the most commonly used agent methylphenidate (Ritalin; this agent is also the main ingredient in Adderall and many variants on the market).
Proponents of amphetamine treatment often state that these animal studies may apply to rats, but have not been studied in humans. This is true, but at least parents and patients should be informed about the animal studies and their strength and consistency. Human studies of the effects of amphetamines in children followed into adulthood have never been done. The longest follow-up is three years into adolescence. In those studies, benefits in the amygdala and basal ganglia were reported with amphetamines, but there also was some evidence of cortical atrophy with amphetamines in those children (a review of those studies is provided here).
Thus, any discussion of amphetamines should begin with a clear awareness of this biological neurotoxicity in animal studies. The claim that these agents are simply “safe” in humans, without any further elaboration, is not scientifically sound.
Symptom-oriented treatment, although very popular, breaks the Hippocratic tradition in medicine, and causes more harm than good. It is certainly the case that amphetamines “work” in the sense of symptomatically improving attention. They do so in all persons, sick or well, ADHD or non-ADHD, as proven in many studies. That is why they are abused in the general population. (3% of college students are prescribed amphetamines; 10% take them). Hence improvement in attention with amphetamines is not at all diagnostic of ADHD. This would be like saying that since benzodiazepines reduce normal anxiety in all persons, then everyone has an anxiety “disorder”. Anxiety occurs normally, as does inattention; we cannot be completely focused on everything all the time; it is normal to be inattentive. Hence there is an enhancement of attention in normal persons with amphetamines (probably a reason why about 15% of baseball players use these agents).
To prove that amphetamines are effective in ADHD per se, in a specific way, one must go beyond the symptomatic improvement in attention seen in all persons to showing an improvement in function that goes beyond other medication or non-medication interventions. This question is best assessed in childhood ADHD, where the diagnosis has been most validated, and where the most valid treatment studies have been conducted.
The longest, largest randomized study of amphetamines in childhood ADHD was the year-long Multimodal Treatment of ADHD (MTA) study. It is commonly cited as supporting the use of amphetamines as the most effective treatment for this condition. In our analysis of that study, we found that this common claim is not supportable. It is true that amphetamines produce more improvement of subjective symptoms of ADHD, meaning the subjective sense of inattention in children; but the comparison arm of behavior therapy was just as effective, or more so, for functional impairment: the children did just as well in academic improvement, in peer and parent and teacher relationships. In other words, for those functional outcomes, which are the most common reasons given for diagnosing and treating ADHD in children, amphetamines were not more effective than behavioral therapy. Further, in three year follow-up, almost one-half of the behavior therapy group remained well without needing to add any amphetamines at all. In other words, the MTA study proves the opposite of what many claim: behavioral therapy is as effective, or more so, than amphetamines for improvement of functional outcomes in ADHD; and with very long-term follow-up, about one-half of ADHD children can be treated with behavioral therapy, without ever needing any amphetamines at all.
It should be noted that the amphetamine treated children in the MTA study had almost an inch of impairment of physical height in the one-year randomized duration of the study. The three year outcome was not randomized and height did not remain impaired but the mistaken conclusion is drawn that height is not impaired long-term; given that the groups were receiving mixed treatments at three years (e.g., half of the behavioral therapy group was taking medications, and about 1/3 of the medication group went off medications), causal conclusions cannot be drawn on this medication effect at the 3 year outcome, but only at the randomized one year outcome. It is most scientifically valid to conclude that amphetamines impair height by about one inch per year in the important years of physical growth in childhood (preadolescence). The impact of being short and small on the self-esteem of young boys should not be ignored.
Besides neurotoxicity, which is a major ignored risk, and physical height, which is important psychologically (but also often ignored), the medical risk of cardiac arrhythmias has long been studied. A large literature suggests that risks exist but are small in children. In middle-aged adults, however, a large clinical analysis recently found about a doubling of risk of ventricular arrhythmia with methylphenidate. This potential risk should be described to individuals in their fourth decade of life or later who wish to take these agents. Whether this is a real risk (ie., causal) was downplayed by the researchers, but there is reason to take the finding seriously; at the very least, patients should be informed of this potential risk, rather than ignoring it altogether.
It should be noted that buproprion (Wellbutrin) is an amphetamine in its pharmacological structure (as is methylphenidate, Ritalin), and has been found to be equivalent to methylphenidate in some randomized studies of childhood ADHD. There are no randomized studies of amphetamines in adult ADHD showing that they are more effective than bupropion. Since bupropion is milder than other amphetamines, we recommend it if an amphetamine is going to be used, in children or adults. Other options include other stimulants, like atomoxetine (Strattera) or even Provigil. We generally do not recommend the use of methylphenidate or its derivatives (like Concerta) or dextroamphetamine or its derivatives (like Adderall). Usually they are not needed, or their harms exceed their benefits, or similar benefits can be obtained with bupropion or other non-amphetamine stimulants. In the minority of cases where clinicians may choose to use amphetamines, we strongly recommend the lowest possible dose for the shortest possible duration (i.e, not to exceed 6 months to a year in children, at a time). We do not recommend using these agents at all in adult ADHD due to the questionable scientific validity of that diagnosis, as explained here, along with the neurotoxicity and sudden cardiac death risks of those agents in adults.