A study covering the 20 years between 1997 and 2016 shows that substantial changes have occurred in the way doctors are treating outpatients diagnosed with bipolar disorder.

Among a number of important changes, one in particular stands out: a much smaller fraction of outpatients today are being treated with a mood stabilizer and a greater fraction are being treated with a second-generation antipsychotic medication.

Other trends include less frequent psychotherapy, more frequent prescription of antidepressants without mood stabilizers, and overall, a near doubling of the total patients with bipolar disorder annually being treated (rising to over 1 million per year from about 470,000 in 1997). The results appeared in the American Journal of Psychiatry. Taeho Greg Rhee, Ph.D., was the paper’s first author.

The team, led by 2016 BBRF Young Investigator Samuel Wilkinson, M.D., of the Yale University School of Medicine, drew upon data from the National Ambulatory Medical Care Survey (NAMCS), a national registry of people with bipolar disorder diagnoses being treated as outpatients. The survey contains details of 4,419 outpatient visits, selected mathematically so as to be representative of 4.2 million such visits made over the 20-year period. For purposes of comparison, treatment decisions were assessed in four-year segments, the most distant being 1997-2000 and the most recent being 2013-16.

Between 1997 and 2000, outpatients with bipolar disorder were prescribed mood stabilizers—lithium being the most important drug in this class—62% of the time; by 2013-16, mood stabilizers were prescribed only 26% of the time.

In the same intervals, the prescription rate among bipolar disorder outpatients for second-generation antipsychotic medications jumped from 12% to 51%. While prescriptions of antidepressant medications rose moderately from 47% to 58% over the 2 decades, prescription of an antidepressant without a mood stabilizer rose substantially from 18% to 41%. After holding steady at about 50%, the fraction of outpatients receiving psychotherapy declined to 35% in the most recent period.

Mood stabilizers, in addition to lithium, prescribed over the years of the study included carbamazepine, lamotrigine and valproic acid. Second-generation antipsychotics included aripiprazole, asenapine, cariprazine, lurasidone, risperidone, quetiapine, olanzapine and clozapine, among others. Antidepressants of many kinds were prescribed over the period, with SSRI (selective serotonin reuptake inhibitor) medicines such as Prozac becoming more frequently prescribed and other types, such as MAO inhibitors, less frequently prescribed.

The team, which also included Andrew Nierenberg, M.D., a BBRF Scientific Council member, 2013 Colvin Prize winner, and BBRF 2013 Distinguished Investigator and 2003 and 2000 Independent Investigator; and Mark Olfson, M.D., 2005 BBRF Distinguished Investigator, noted in their analysis that “second-generation antipsychotics have in large measure supplanted lithium and other mood stabilizers in the absence of any comparative effectiveness data” that would indicate superior outcomes for patients. The researchers also noted “a consistent lack of evidence for the efficacy” of antidepressants in outpatients with bipolar disorder.

In view of the trends they identified in how patients are being treated, the team strongly recommended the urgency of conducting comparative effectiveness studies for mood stabilizers vs. second-generation antipsychotics, as well as efficacy studies for antidepressants in bipolar disorder patients.

In noting the substantial changes in treatment practices, the team offered a number of possible explanations. During the study period, they noted, second-generation antipsychotics received regulatory approval for treatment of bipolar disorder. This, combined with vigorous marketing campaigns, including direct-to-consumer campaigns, by makers of these medicines, likely accounts for their great increase in use in bipolar outpatients.

The researchers noted that lithium, the most commonly prescribed mood stabilizer, does have side effects in some patients, ranging from hyperthyroidism to diabetes and lithium toxicity. But they also noted that lithuium has been demonstrated in clinical trials to reduce suicidality in bipolar disorder patients. Lithium’s side-effects alone cannot explain the rise of second-generation antipsychotics, they noted, since those medicines, too, have potential side effects, including tardive dyskinesia (a motor disorder) and diabetes.

In trying to account for the continued popularity of antidepressant prescriptions for bipolar disorder outpatients, the team noted the fact that most patients spend a much larger portion of the time in the depressive phase of the illness than in the manic phase. Still, “antidepressant use has been shown in large samples to heighten the risk of mania” in bipolar disorder patients when they are not paired with mood-stabilizing medicines, they observed.

While acknowledging that the NAMCS data upon which their findings are based cannot be used to measure the effectiveness of outpatient treatments for bipolar disorder over the 20-year period, the team stresses the importance that such studies be performed in light of the changes in treatment patterns that their study revealed. They also noted the need to distinguish prescribing patterns and responses of patients with bipolar I disorder from those with bipolar II disorder—something the current study was not able to do.