Treating Adults With ADHD Requires Special Considerations

Many people commonly think of attention-deficit/hyperactivity disorder (ADHD) as a pediatric disorder, but symptoms often persist into adulthood. Estimates suggest that about 4 percent of adults have ADHD.

(Image: iStock/shironosov)

Fortunately, good medication options are available, but there are several considerations specific to adults with ADHD that you should keep in mind before prescribing medications.

To help you navigate the potential risks of medications commonly used to treat ADHD, my colleagues and I have developed a step-by-step algorithm on adult ADHD pharmacology. However, it is a work in progress and has not yet gone through peer review. Indeed, important new research published just in the last month has led to one change in recommendations. We are looking for feedback.

As is the case when treating patients with pediatric ADHD, stimulants such as methylphenidate or amphetamines are considered front-line treatments for adults with ADHD. These medications are well tolerated, effective, and can easily be used alongside most other psychotropic medications. Atomoxetine—a norepinephrine reuptake inhibitor—can also be used, should a clinician or patient have concerns about stimulant use. Currently, there are not sufficient data to make firm recommendations for prescribing any other medications commonly used to treat patients with pediatric ADHD (clonidine, guanfacine, modafinil, or bupropion) for adults with the disorder.

Before prescribing a stimulant or atomoxetine to an adult with ADHD, make sure the patient does not have angle closure glaucoma. If they have another type of glaucoma or pheochromocytoma (a tumor of the adrenal glands), treatment with stimulants and atomoxetine is possible but only after effective treatment and clearance by the specialists treating those disorders.

Clinicians are also advised to hold off on initiating stimulants or atomoxetine in patients with hypertension, angina, or a history of cardiovascular defects until receiving medical clearance. Clinicians treating women with ADHD who are pregnant, thinking about becoming pregnant, or breastfeeding should make medication decisions based upon the severity and impairment of the patient’s symptoms.

You should next assess for certain comorbidities that could require deviation from the standard algorithm. If the patient has a history of psychosis, clinicians should typically avoid stimulants or atomoxetine because of the risk of exacerbating the psychosis. The use of stimulants in patients with uncontrolled (or undiagnosed) bipolar disorder poses significant risks of emergent mania if the patient is not already stabilized on an appropriate mood stabilizer such as lithium; the risk of mania could be increased by 6-fold or 7-fold compared with bipolar patients not given stimulants (1).

For patients who do not have comorbid psychosis or bipolar disorder, clinicians can choose between a methylphenidate product or an amphetamine product. A recent meta-analysis considering efficacy and tolerability of stimulants in adults found (based on a small number of studies) that amphetamine products appear to be preferred (2). Whatever stimulant is chosen should be initially prescribed in an immediate-release formulation, as effectiveness and side effects can be quickly identified, and dosage adjusted most efficiently, as needed. Titrate the medication for up to several weeks, if needed, to find an optimal profile of symptom improvements versus side effects, and then switch to an extended-release formulation if necessary, depending on the patient’s needs for the optimal timing of the benefits.

Adults can often require doses as high as 1 mg/kg to 1.3 mg/kg per day for methylphenidate or 0.5 mg/kg to 0.65 mg/kg per day for amphetamine before optimal symptom improvements are seen. If after a reasonable trial, the results are unsatisfactory, switch patients to the other class of stimulant and titrate again in the same manner.

If neither stimulant class results in symptom improvement, one can consider trying atomoxetine, typically starting with 40 mg per day and raising to 40 mg twice daily in a week. A clinically meaningful difference from placebo with atomoxetine is very slow to develop compared with stimulants, and can take four to six weeks, and the maximal response (which appears smaller than the response to stimulants) may take several more weeks after that (2, 3). This makes atomoxetine impractical as a first-line agent compared with stimulants, which work within hours when you have an effective dose. In considering it as a third-line agent, as we do, it is important to note that we have no evidence that atomoxetine can be effective after one or two stimulants have failed.

If misuse or diversion occurs during either stimulant trial, consider trying extended-release formulations which are much less prone to abuse or diversion. If it occurs again even with these formulations, consider atomoxetine.

The presence of comorbid substance use disorders (SUDs) may require modification of the algorithm. In order to avoid the high risks of abuse and diversion in these patients, clinicians should initiate treatment with an extended-release stimulant formulation even though it takes longer to observe effects. Clinicians should also try to treat patients simultaneously for both their ADHD and substance use. If the substance use is severe, treat the SUD first and achieve a period of sobriety before considering medications for the ADHD.

Patients with cocaine use disorder have shown both an improvement in ADHD symptoms and a reduction in cocaine use with extended-release amphetamine treatment, so this specific stimulant class is a recommended choice (4). For people with methamphetamine use disorder, amphetamines should be avoided, but methylphenidate has been shown to be effective at reducing symptoms of ADHD and reducing drug use as well (5). For patients with ADHD and co-occurring alcohol use disorder, atomoxetine can be given first, as there are some data to suggest atomoxetine can reduce heavy drinking days gradually over time (6). For opioid use disorder, more study is needed, but you could consider an extended-release stimulant first.

Much more research is needed to understand adult ADHD and its treatment. There are much more data relevant to what works when treating children with ADHD, and we need more direct evidence in adults.

A flowchart of the latest version of the algorithm for adults with ADHD may be seen on the website of the Psychopharmacology Algorithm Project of the Harvard South Shore Psychiatry Program. ■