Dr. Howland is Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania.
The author has disclosed no potential conflicts of interest, financial or otherwise.
Address correspondence to Robert H. Howland, MD, Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Pittsburgh, PA 15213; e-mail: HowlandRH@upmc.edu.
Within the past 6 decades, a large and diverse assortment of medications has become available for the treatment of various mental disorders. Given what is available, how should a clinician decide which medication(s) to try for any particular patient? In this article, I discuss eight questions that should be considered when selecting medication for a patient.
What is the Medication Being Used for?
The first step in selecting medication is to have a clear understanding about what the medication is to be used for in the treatment of a patient. All marketed medications have labeled indications approved by the U.S. Food and Drug Administration, but many psychotropic and some non-psychotropic drugs can also be considered appropriate for off-label use in the management of mental disorders.
Medications are prescribed for five main reasons. Medications typically are selected as the principal treatment for major diagnostic syndromes (i.e., schizophrenia, major depression, and other mental disorders). Medications can also be added to target residual symptoms when the principal treatment is only partially effective. They can be co-prescribed initially along with principal treatments to target particular symptoms associated with the main condition (e.g., anxiety, insomnia). Similarly, additional medications might be instituted during particular phases of a disorder. Examples of this would include a patient with bipolar disorder being treated with a mood-stabilizing agent who becomes depressed or a patient with recurrent major depression being treated with an antidepressant agent who develops a psychotic episode. Finally, medications can be used to treat the adverse effects caused by other drug treatments, such as tremors or parkinsonian side effects.
Although the importance of an accurate characterization of the patient’s diagnosis or diagnoses, as well as associated symptom patterns, is intuitively obvious, it is not always easily accomplished in clinical practice. Patients can have multiple diagnoses. Symptoms or symptom clusters are not necessarily unique to a particular disorder and can cut across diagnostic boundaries. Signs or symptoms attributed to the disorder could be medication side effects and vice versa. Often, medications can be used for more than one reason, and symptoms or syndromes can often be treated by different types of medications. These issues can be exemplified by three common examples. The selection of medication(s) for a patient with depression will depend in part on recognizing whether or not the patient has a history of manic symptoms. Selecting a medication for the treatment of insomnia might differ for patients with major depression, bipolar disorder, or a substance use disorder. Medications used to treat antipsychotic drug-induced tremors would not be effective for lithium-associated tremors or for tremors sometimes seen in patients with severe anxiety.
What Could Happen When the Medication is Taken?
As described in the preceding section, medications are generally selected for their potential therapeutic benefit, depending on the clinical scenario, but it is equally important to consider the potential adverse effects of medications. Among a group of medications used for the same clinical reason, the drugs might be somewhat or very different in their expected safety and side effect profiles, and this knowledge should be used during the medication selection process. For example, drugs that promote weight gain would not be ideal for overweight patients but might be preferred for patients with poor appetites or who are underweight. Similarly, drugs that tend to have sedating effects might be a good choice for sleepless patients but not necessarily for those who are sleeping excessively. In the treatment of bipolar depression, using antidepressant medication carries the adverse risk of inducing hypomania, mania, or psychosis, whereas certain other mood-stabilizing drugs that have antidepressant properties are less likely to be associated with this particular adverse risk. Dopamine agonist drugs might be the preferred treatment for antipsychotic drug-related parkinsonian side effects in older patients because they are less likely to cause the adverse cognitive, bowel, and bladder effects associated with anticholinergic drugs. These scenarios each demonstrate how the expected adverse effect profile of different medications should be considered when selecting medication.
Does the Patient Have Other Conditions That Might Affect Medication Therapy?
Many patients have comorbid medical or psychiatric conditions that are important to be aware of when selecting medication. Some medications might exacerbate comorbid conditions or may complicate their treatment. Some conditions might pose a risk when using certain medications (e.g., psychotropic drugs having cardiovascular effects in patients with heart disease, controlled drugs in patients with substance use disorders), so alternative medications should be considered. Likewise, certain conditions (e.g., liver or renal disease) could influence the metabolism or clearance of some drugs but not others. Hence, a good evaluation of a patient’s medical and psychiatric history is imperative when selecting medication.
What Other Drugs is the Patient Taking?
Even more important than the presence of comorbid conditions is the concurrent use of other pharmacological substances, including prescription and over-the-counter medications; nicotine; alcohol; illicit drugs; and vitamins, supplements, or herbal preparations. Drug interactions mainly occur in the liver (affecting drug metabolism), the kidneys (affecting drug clearance), or the central nervous system (influencing therapeutic or adverse effects). Depending on the drugs and the type of interaction, this can result in a less-than-optimal therapeutic benefit, can potentiate known side effects, or can cause unexpected toxic effects. Patients should be systematically queried about their use of all types of drugs. Knowledge about potential drug interactions should be used to avoid the selection of medications that might adversely interact with anything the patient is taking.
Has the Patient Been Treated in the Past with Medication?
One of the best (although imperfect) predictors of treatment outcomes is past treatment experiences. As a result, one of the most useful strategies for medication selection is carefully assessing and categorizing past medication trials according to efficacy and tolerability, although this obviously would not be helpful for treatment-naïve patients. It is not uncommon for patients to have been treated with various medications in the past. If so, they may have a relative perspective on what helped completely, partially, or not at all. In addition, they might remember how well they tolerated particular medications. Trying the same medication again (or using a pharmacologically similar medication) is rational if it was previously effective and well tolerated, but it is sensible to avoid any medication that was absolutely ineffective or intolerable.
A clinical axiom is that with each medication treatment trial failure, the odds of a satisfactory response to subsequent medications tend to drop, although this is more likely to be true among patients who have been treated previously with a large number of different medications. For such patients, I would consider selecting medications (or medication combinations) that are quite different pharmacologically than what has been tried before. For example, this might include turning to older classes of medication, such as tricyclic antidepressant or monoamine oxidase inhibitor drugs, for depression or the older, first-generation typical antipsychotic drugs for schizophrenia.
For patients with a history of medication intolerance, selecting a new medication is challenging. One approach, as above, would be to select an entirely different drug class. Another approach would be to select a previously tried medication but only after formulating a particular plan for anticipating and managing the expected side effects.
Is There a Family History of Medication Treatment?
The concept of personalized medicine includes the potential use of a patient’s genotype (genetic makeup) to predict how he or she will respond to or tolerate medication. Although there are ongoing research efforts to develop such genetic tests (and some tests, such as the AmpliChip CYP450 Test from Roche Diagnostics, are commercially available), none have been validated as being clinically useful for medication selection. However, a crude method of using genetic information for medication selection is the collection of family history information. Most mental disorders have a significant (although not absolute) genetic component. As such, having a family history of a mental disorder (especially among first- and second-degree relatives) is often associated with an increased risk of developing the disorder.
Extrapolated from this is the notion that treatment responsiveness might have a genetic component as well. It would be reasonable to ask patients whether they have any blood relatives who have been treated for the same condition they have and, if so, what type of treatment was used and what the treatment response was. If a patient reports a family member as having had a satisfactory response to a particular medication, then it would be appropriate to consider trying the same drug. On the other hand, if a family member has had a negative experience with a particular mediation, then trying something else is more sensible.
Does the Patient or Family Have a Treatment Preference?
Beliefs, opinions, and preferences are strongly associated with medication adherence and medication treatment outcomes. Whether or not patients’ beliefs are clinically or scientifically justified, their beliefs are likely to have a strong influence. If a patient has a favorable impression about a particular medication based on the experiences of friends or family, or based on information the patient has heard or read, he or she will be more likely to take it and may have a better outcome. Conversely, if patients have an unfavorable perception about a particular medication, they will be less likely to accept or respond to it. Conducting an “opinion poll” with patients about medication should be part of the medication selection process.
What Does the Medication Cost?
Medication costs affect treatment adherence and outcome. For uninsured patients, certain medications are prohibitively expensive. Even patients with insurance coverage may find that their out-of-pocket copay for some medications is too great to bear. Because many patients take multiple medications, they might decide not to take one or more of them based on cost. This could also occur if patients have to budget their medication resources together with other family members. Taking into account the relative efficacy, tolerability, safety, and track record of available medications, cost-effective drugs usually can be appropriately selected, without the patient (or the clinician) believing they are settling for a “second-rate” product.
In addition, generic drugs are increasingly available within each therapeutic class. That newer products (often more expensive) are “better” than older drugs is more myth than reality. Average efficacy rates across drugs are usually similar. Side effect profiles may differ, but this typically means only that there is a clinical trade-off between two different sets of side effects.
Medication selection should be individually tailored by taking into account factors such as diagnosis and associated symptoms, expected efficacy/tolerability profile, medical and psychiatric comorbidity, concurrent drug use, past treatment history, family history, patient/family preferences, and cost considerations. Nurses have an important role in the medication selection process, not only by helping collect information in these areas, but also for those advance practice nurses who have prescriptive privileges.