Do drug companies hold too much sway over doctors? A host of studies concerning antipsychotic drugs suggests that they may, although as in many debates, the devil is in the details.
The latest research - it appears in advance-of-publication form in the American Journal of Psychiatry - concerns the treatment of children and adolescents with early-onset schizophrenia. 
Linmarie Sikich, of the University of North Carolina, and colleagues across the country worked with 116 boys and girls, ages 8 through 19, who had substantial mental illness but were not acutely suicidal. The patients received either a second-generation antipsychotic medication, Zyprexa or Risperdal, or a medication that has been available since 1975, Moban.
The older, much cheaper medication worked as well as the new ones; if anything, it was more effective. And Moban did not cause weight gain or pre-diabetic symptoms. Kids on Zyprexa gained over 13 pounds, on average - a startling amount - in eight weeks; in that short time children on Zyprexa also showed worrisome increases in insulin, cholesterol, and lipid levels as well as changes in liver function tests, all suggestive of a risk for diabetes down the road. Patients on Risperdal gained a pound a week. The downside of Moban was moderate or severe akasthesia - restless agitation on a neurological basis - in 18 per cent of subjects, but children in every group had neurological symptoms, including an especially worrisome side effect, involuntary movements, or dyskinesia.
The new study extends the findings of other recent trials; those studies show that for adults the additional benefits of the newer, much more expensive antipsychotic medications are modest at best. (In a adults, too, Zyprexa caused the worrisome metabolic changes, and an older medication - in the largest trial, it was Trilafon - caused neurological problems.) Given the near equivalence of the drugs, you would think that doctors would be prescribing the older medications. In the public sector especially, where the taxpayer is footing the bill, the less expensive drugs should be popular. I recall a visit to Tennessee four years ago at a time when the Governor was trying to restrict the use of on-patent antipsychotics, arguing that the extra cost of the new medicines exceeded the amount he was being forced to cut school budgets. But state hospital and community mental health center patients generally put patients on the newer medications.
So, yes, it looks as drug marketing has been all too effective. But there are also other reasons for the prescribing pattern. The newer antipsychotics may be less likely to make people feel drugged. A review in yesterday's New York Times quotes Michel Greenberg in Hurry Down Sunshine (discussed here three weeks ago) about a first-generation antipsychotic: "On chlorpromazine, the poet Robert Lowell was unable to build a three-letter word on a Scrabble board or follow the count of balls and strikes in a televised baseball game."
Many patients were put on the newer drugs before the metabolic side effects were well documented; and most of the older antipsychotics - molindone tends to be an exception - also cause weight gain. But the decisive issue for many doctors is tardive, or late-appearing, dyskinesia, a terrible movement disorder - it can involve helpless writhing - that sometimes appears years after a patient has stopped taking a medication. Doctors hate and fear the syndrome; rightly or wrongly, they believe it is associated with early neurological symptoms, like akasthesia, associated with the older drugs.
When it comes to antipsychotic medications, side effects are enormously important. In an 18-month study testing five drugs, two-thirds to three-quarters of patients went off their medication. For Geodon and Seroquel, the discontinuation rates were around 80 per cent. When only one in five subjects takes a medicine, it's hard to say much about efficacy.
The picture is clear but not perfectly so. Patients should be on the older drugs - except that the (relatively) lucky patients who happen not to get metabolic side effects might prefer the newer ones. If doctors read the outcome literature, they will be prescribing from a wider pool of medications and monitoring side effects assiduously. What is called for is more attentive care, a difficult requirement in a medical system where patients tend to be treated in the overburdened public and primary care sectors.
It turns out that the drugs we have to choose from are terribly flawed. The more we learn about schizophrenia, the more it looks like a collection of related forms of developmental brain damage. The medications pick off some symptoms and cause others. So yes, doctors ought to move some patients, probably many patients, to older, cheaper drugs. But finally, there are no attractive choices. Any particular clinical situation is likely to be a matter of what I call "hard cases make bad law."
