Bipolar Medications — What Five Labels Say About the Drugs 6 Million Americans Combine

The Package Insert series

Fifty-seven percent. That is the somnolence rate for Seroquel in bipolar depression trials — not drowsiness as a side effect that some patients may experience, but a majority of trial participants sedated on the most prescribed antipsychotic in the United States. The FDA-approved label reports it plainly, in a table, next to the placebo rate of 15%. The drug that millions take daily for bipolar depression puts most of its users to sleep.

That number appears on page 23 of the Seroquel prescribing information. On nearby pages: Depakote’s label warns it causes neural tube defects, decreased IQ, and neurodevelopmental disorders in children exposed in utero. Abilify’s discloses postmarketing reports of pathological gambling, compulsive sexual urges, and binge eating — urges patients may not recognize as abnormal. Lithium, the oldest drug in this group, carries a 530-line label while the others run to thousands — because it was approved in 1970, before the FDA required the clinical trial data that modern labels demand. And every one of these five drugs admits, in its own mechanism of action section, that how it works “is unknown” or “unclear” or “has not been established.” These are statements from the FDA-approved labels, written by the manufacturers, reviewed by the regulators, and filed with the agency.

This installment covers five drugs from three pharmacological families: quetiapine (Seroquel), an atypical antipsychotic; lithium (Lithobid), a lithium salt; lamotrigine (Lamictal) and divalproex sodium (Depakote), both anticonvulsants; and aripiprazole (Abilify), another atypical antipsychotic. Together they represent the most commonly prescribed medications for bipolar disorder in the United States. What makes this class different from statins or ACE inhibitors — the subjects of previous installments — is that bipolar patients routinely take two or three of these drugs simultaneously. Each label was written as if the drug would be used alone or with one specified co-medication. The reality of the prescription pad is something else entirely.

These drugs appear to work because they sedate, flatten, and suppress. The labels document this in their own adverse reaction data: 57% somnolence on Seroquel for bipolar depression; 34% somnolence as adjunct therapy for mania; 19% somnolence on Depakote; sedation and akathisia on Abilify. Lithium’s discoverer, John Cade, found it by accident in 1949 when guinea pigs injected with lithium became “extremely lethargic and unresponsive to stimuli.” He then tried it on hospital inmates and reported a “nonspecific leveling effect.” Controlled studies of lithium in normal volunteers, led by NIMH director Lewis Judd, found “a general dulling and blunting of various personality functions.” The volunteers’ partners and roommates noticed the dulling. The trained mental health professionals evaluating the same volunteers “were unable to detect any behavioral changes.” The early neuroleptics were originally called tranquilizers and marketed for their “taming” properties. They were rebranded as “antipsychotics” and “mood stabilizers” — but the pharmacology did not change with the name. As psychiatrist Peter Gøtzsche has observed, these drugs “don’t stabilise anything.” Their primary effect is to suppress emotional responsiveness. A person who has been chemically subdued will score lower on a mania rating scale. That reduction in score is what the trials measured, and what the FDA accepted as evidence of efficacy.

Once started, these drugs are difficult to stop — and the difficulty is built into the pharmacology, not the patient. In 1999, Harvard’s Ross Baldessarini reported that half of all patients relapsed within five months of quitting lithium. Without any drug exposure, it took nearly three years for 50% of bipolar patients to relapse naturally. The time between episodes after lithium withdrawal was seven times shorter than the natural course of the illness. In the few studies where patients were gradually withdrawn, only 29% relapsed — a rate lower than the drug-maintained group. What the withdrawal studies actually demonstrated was not that lithium prevented relapse, but that stopping lithium triggered it. Patients who took the drug and then quit ended up, in the words of UK psychiatrist Joanna Moncrieff, “worse than if they had never had any drug treatment.” Scottish psychiatrist Guy Goodwin concluded in 1993 that lithium may be “harmful to bipolar patients” if they take it and then stop within the first two years. Seroquel’s label documents a discontinuation syndrome in 12.1% of patients — insomnia, nausea, vomiting, dizziness, irritability — and advises gradual withdrawal. Neither antipsychotic in this group provides a structured taper protocol for bipolar patients.

The Package Insert is a paid subscriber series. Each installment reads the FDA-approved prescribing information for a major drug class and reports what the labels say — including what they admit has not been studied. If you’re currently taking bipolar medications, considering one, or have a family member on one, this is the installment to start with.

What you’ll find behind the paywall:

• What the labels say the drugs actually do — which drugs proved efficacy for mania only, which for depression, and which admitted their long-term effectiveness “has not been demonstrated”

• The metabolic cost — weight gain rates of up to 23%, thyroid suppression, cataracts in dogs, and mammary tumors in rats at doses below or near human equivalents

• What Depakote does to a pregnancy — the only drug in this group whose own label warns of decreased IQ in children exposed in utero

• Five mechanisms of action, zero explanations — every label admits the drug’s mechanism is unknown, unclear, or unestablished

• The reproductive damage — testicular atrophy, abnormal sperm, disrupted ovulation, polycystic ovary disease, and what the labels admit hasn’t been studied in humans

• What happens when you combine them — the admission buried in the Seroquel label about what has and hasn’t been evaluated

• The compulsive behavior warning — pathological gambling, compulsive shopping, and sexual urges reported after Abilify’s approval

• Questions to take to your doctor — 12 questions drawn directly from what the labels disclose, formatted as a printable list