On Antidepressants

Macbeth: Cure her of that.

Canst thou not minister to a mind diseased,

Pluck from the memory a rooted sorrow,

Raze out the written troubles of the brain,

And with some sweet oblivious antidote

Cleanse the stuff'd bosom of that perilous stuff

Which weighs upon the heart?

Doctor: Therein the patient

Must minister to himself.

Macbeth: Throw physic to the dogs, I'll have none of it.

The other day I was reading an account of Mozart's final hours, in which he lay wracked by fever and pain, still idiopathic to this day. When the doctor arrived, he prescribed cold compresses. If that good medical man had any notion of what he was dealing with, what might it have been like to attend to one of the greatest composers in the history of the world and to be able to do nothing better than cold compresses? (Of course, it is thought that his life may have been iatrogenically shortened by previous blood-letting as well).

I've been reading Thomas Hines's Architecture of the Sun, a history of modernism in southern California, in which various people moved to that region upon medical recommendations of a more salubrious climate. Those were the days, when a prescription for sunlight, dry air, and palm trees was respectfully viewed as deep medical wisdom! A doctor's word could relocate people across continents.

The history of medicine is a weird amalgam of ignorance and authority, incapacity and power. Even through the millenia in which the practical (positive) effects of medicine were minimal, doctors nonetheless occupied a crucial social locus of judgment and prestige.

Physicians now fight for space in a more crowded arena, but some of their outsized influence remains. For better or worse, a diagnosis from a psychiatrist often carries more weight than one from a therapist. When patients need temporary time off from work or apply for disability, the paperwork seeks the opinion of the physician, not the therapist, nurse practitioner, or physician's assistant. The physician still wields the gavel of the sick role most vigorously.

But if, as the commenter to the previous post suggested, the primary function of the doctor must be the relief of suffering, what happens when the doctor's tools are in fact too weak to accomplish this, or what is more complicated, what happens when the effect of those tools is owing to their wielders' social power rather than to any inherent properties (i.e. the placebo effect)?

Much has been heard of late about the dubious effects of antidepressant drugs, an issue that can only give any psychiatrist serious pause. This issue has been raised many places, and this post is not intended as a literature review, but Sharon Begley's Newsweek article may be as good a summary as any. If antidepressants are truly no better than placebo, then a scientific fraud on an unprecedented scale would have been perpetrated over the past half-century, and there would be something seriously rotten in the state of psychiatry.

I believe, of course, that antidepressants have real effects, otherwise I could not do my job in good faith. This belief could be deemed meaningless inasmuch as I have self-interested reasons of both professional standing and financial stability for holding it, and the human capacity for self-deception is unfortunately vast. But there are of course scientific reasons to doubt ambiguous drug trials, chief of which is the fact that many patients (or "patients") enrolled in drug studies are not representative of real clinical settings. Their disorders tend to be milder and more pure (i.e. uncomplicated by other diagnoses), and the very fact of their willingness to participate in a drug trial may heighten their response to placebo.

So, what do I believe about antidepressants based on 15 years of prescribing them to many hundreds if not a few thousand individuals in various settings? I believe that they are imperfect drugs that too often fail to work--my ECT experience alone would tell me that. I believe that antidepressants treat symptoms of still mysterious illnesses; they do not target the illnesses themselves. They are non-specific, affecting a broad spectrum of emotional response and anxiety level; in that sense they are more like what David Healy, in The Antidepressant Era, referred to as tonics than like magic bullets (think aspirin, not penicillin).

I believe that antidepressant effects are stronger, relative to placebo, the more severe and sustained depression or anxiety is. For mild and transient conditions they are often useless (thus the irony of the moral panic over Prozac-based emotional enhancement). I believe that people (prodded by drug advertising and cultural momentum) rely on them too much on average. Social and psychological interventions should be tried first.

But I refuse to believe, broadly and scientifically, that antidepressants are indistinguishable from placebos. For one thing, within the pharmacopeia there are a few drugs that I think of as internal placebos (buspirone or hydroxyzine, anyone?) having little effect beyond a hope and a prayer. But the mainstream antidepressants aren't like that, and while I know that the placebo effect is powerful and still not well understood, I have seen too many unimpressionable people with dramatic and sustained responses, and too many impressionable types who fail to improve, to believe that there is no physiological effect.

On what may seem like a trivial note, I haven't researched the literature, but many pet owners and veterinarians attest to the effects of Prozac, etc. on neurotic cats and dogs; can that merely be a mass phenomenon of placebo effect by proxy?

Mad Scientists at Work

I have a new hero after reading a profile of neuroscientist and writer David Eagleman:

Eagleman rejects not only conventional religion but also the labels of agnostic and atheist. In their place, he has coined the term possibilian: a word to describe those who "celebrate the vastness of our ignorance, are unwilling to commit to any particular made-up story, and take pleasure in entertaining multiple hypotheses."

Sign me up--I want to be a possibilian.

The "guinea pig" complaint is far and away the most common one I hear about previous psychiatrists (and I'm sure it is said about me by those patients who move on to other prescribers). A new medication is tried every month, seemingly willy-nilly, without a sense of an overall framework or plan. A psychiatrist, plainly, is no auto mechanic. Psychotherapy, truth be told, really is a neverending experiment, but medication somehow is supposed to be different.

We know a vast amount about the effect of medications over large populations, but idiosyncratic variation in drug response remains too great to predict outcomes for individuals. In that sense medication reactions are almost like an extension of the assessment process. Treatment, diagnosis, and prognosis become one. Medication trials obviously don't give the same kind or precision of information that a brain MRI will give a neurologist, but they are very informative of a patient's dynamics and likely outcome.

The problem of prognosis is fundamental to psychiatry. Neurologists, even though they often can do relatively little about sometimes appalling diseases (MS, ALS, Huntington's Disease), nonetheless enjoy a greater stature than psychiatrists because, even if they can't do more, they know more. A patient would obviously prefer to get well, but if he can't get well, he wants to know what the future holds so that he can wrap his mind around it and plan accordingly.

There are of course crude measures of prognosis: general intelligence, education, financial and social support, and the presence or absence of past hospitalizations, suicide attempts, personality disorder, and substance abuse. But on a more subtle level, in psychiatry prognosis declares itself only over time, as the myriad variables involved in a mind interact with unique life circumstances. The physiological systems generating identity, behavior, and other aspects crucial to psychiatry are far more complex and unpredictable than those giving rise to, say, motor or sensory control.

The milder or more subtle a condition is, the harder prognosis can be to pin down. I can no more predict how a low-grade dysthymia will behave over decades than I can predict when a person might get married, or how much money they'll be making ten years from now. This isn't to say that I can't predict at all, but such prognostications are based as much on common sense (the past tends to predict the future, etc.) than on any grand professional expertise. Since I'm not allowed to keep a crystal ball in the office, I'm limited to indirect measures of understanding.

Maybe more psychiatrists should aspire to be possibilians, to "celebrate the vastness of our ignorance" rather than pretending to more knowledge than we actually have. Prescribing a medication isn't like performing an oil change--it is accompanying a patient in an experience of self-discovery. Physicians differ from drug dealers in that the substances we purvey, by social contract, must have minimal standards of safety, uniformity, and usefulness. In addition, we are expected to be wise and discerning students of human nature. Beyond that, things get interesting.

The Unnameable

After almost twenty years in my mind, the syllabic cornucopia of psychotropic drug names has made its case for a celebratory post. What has taken so long? And would I want the job of coming up with a moniker for Lilly's next miracle?

I considered a top ten list, but let us consider them by class:

1. Antidepressants: The meat and potatoes of psychiatry, these names are sure to be written with hand-numbing repetition, so it is a good thing that for the most part these drugs are happily named. A linguistic and pharmaceutical titan, Prozac is an arresting amalgam of the soothing, almost soporific Proz- followed by the hard "ack" that provokes comparison with another very popular drug of the 1980's. As the drug itself is meant to do, the name both calms and enlivens.
Zoloft and Paxil, completing the original SSRI trifecta, are also remarkably mellifluous names, although the former's buoyancy wafts dangerously close to corniness. Paxil follows flattery of Latinate pedantry (Latin pax=peace) with the relaxing -il evoking dutiful memories of Elavil, grandaddy of them all.
After these, antidepressant inspiration was spent in both chemistry and name. Celexa, its knock-off Lexapro, and Luvox? Not memorable. Effexor showed a lack of subtlety (Get it, "affects her," the majority of depressed patients being women?), regrettably dubbed "Ineffexor" when failing to work or causing dismaying withdrawal symptoms. Wellbutrin is a name simply rebarbative and without redeeming qualities, and it has been painful to hear a few concretely-minded patients over the years refer to it dismissively as "Badbutrin," which is a crime against both wit and alliteration.

2. Anxiolytics: It is a shame that Librium, lifted whole from a pleasing state of balance, did not prevail as a popular benzodiazepine. Similarly, Valium, connoting valiant equanimity, has largely fallen by the wayside. Instead we have Klonopin, which brings to my mind some kind of blunt instrument; Ativan, which seems to me a very gray sort of word, summoning nothing whatsoever; and the always suspect Xanax, whose pernicious influence may draw somewhat from its palindromic potency. A drug used for panic attacks should not end in -ax. But then again, I have heard that the color red, which tends to make people feel agitated and uncomfortable, sells best in grocery stores. Appropriately an afterthought, Buspar is a drug that eminently deserves its lame designation.

3. Mood-stabilizers: With the exception of lithium, which enjoys the elegant purity of being plucked right from the periodic table, the third lightest element in the universe, this group is composed of referential failures. Depakote. Tegretol. Lamictal. Whatever their pharmacological effects (which may be considerable), these names do not inspire confidence, and can be the insult on top of the injury of a diagnosis of bipolar disorder.

4. Antipsychotics: The hoary first generation of "major tranquilizers" could not be topped in dignity. Thorazine summons a compound of Nordic power with neuropsychiatric precision, a hammer brought down with pinpoint accuracy. Haldol conveys both majesty and trustworthiness, as of a respected elder. Navane is a name both implacable and imperious, sounding as if it should have been given by injection only. The contemporary offspring suffer from a failure of ingenuity. While Seroquel suggests a certain sophistication, Risperdal and Geodon are vaguely boorish in tone, while Abilify is simply an embarrassment. The unfortunate progeny of ability and fortify, it is a name that can't be taken seriously, which is a shame, because it isn't at all a bad drug. "Have you ever taken Abilify?" is a bit like asking, "Have you ever drunk Kool-Aid?" Please.

5. Stimulants: Provigil is sort of cool, evoking the steadily tenacious all-nighter, but maybe a bit too ominously. Compared to Ritalin, which is vaguely reassuring but forgettable, Adderall was a triumph of shameless audacity. The name is a naked directive: add this drug to all the patients you can, period. Why simplify or streamline your life when you can, in fact, add more? Add what? Adderall! You can have it all. At this point we are beyond subtlety, the closest possible thing to the drug name "Takethis."

A note on generics: Chemical drug names, with very few outliers, are infelicitous and afford little pleasure, except to the most self-righteous who refuse to pay their respects to brand names. Fluoxetine, carbamazepine, and thiothixene are flashbacks from Organic Chemistry. The only exceptions would be haloperidol and valproic acid, which are stimulatingly, sinisterly(?) decadent, sounding akin to something like absinthe.

In the world of medication management, we take our gratification where we can. (Note: this spoof concerns names and not the medical value of any of the medications mentioned. Ask your doctor).

On Med-Seeking

There cannot be a pinch in death

More sharp than this is.

Imogen (Cymbeline)

I was thinking yesterday about the benzodiazepines (Valium/diazepam, Klonopin/clonazepam, Ativan/lorazepam, Xanax/alprazolam), those somewhat disreputable but often unavoidable therapeutic reinforcements (sort of like a mercenary army, the Blackwater of anxiety treatments). Every doctor seems to have his/her unique philosophy about these drugs, ranging from benzos-as-Tylenol to benzos-as-Four-Horsemen-of-the-Apocalypse. Indeed, the sheer variety of benzo perspectives one encounters pretty much guarantees that ideology more than science or pragmatism is at work. And like much prescribing and diagnosing that goes on in psychiatry, benzos are both overdone and underdone depending on the situation.

My understanding of a substance use disorder is continued use of a substance despite ongoing detrimental effects, whether to health, finances, legal status, work, or relationships. Depending on severity, it may involve involuntary cravings for the substance, desperate attempts to obtain it, and the development of tolerance and potential withdrawal. As is the case for opioids and pain treatment and for stimulants and ADHD, though, the handling of benzos can involve some ambiguous situations.

The central conceptual problem is trying to distinguish recreational euphoria from real relief from ongoing distress, a relief that in itself can seem like a kind of euphoria. After all, a frequent complain among doctors about benzos involves those patients who come in specifically asking for them. To many clinicians this "med-seeking" behavior in itself reflects a likely abuse problem, and surely in some cases it does, but does it necessarily?

It is noteworthy that these three classes of drugs--opioids, stimulants, and benzos--are not only susceptible to abuse but are also among the most reliably effective treatments for pain, ADHD, and anxiety, respectively, especially but not only in the short term. This is no coincidence of course--they clearly have potent and pronounced effects on neurotransmitters that can, depending on the patient, induce therapeutic relief or iatrogenic problems. People in pain may "seek" Percocet, and people with ADHD who have benefited from Adderall in the past may "seek" it again.

To return to benzos, it is worth pointing out that the overall pharmacologic options for anxiety are quite limited. Sure, all the antidepressants, from MAOI's to SSRI's, have effects on anxiety, but they are neither as reliable nor as rapid as those of benzos, and antidepressant side effects are generally worse than benzo side effects (setting aside abuse liability). Beyond antidepressants, one has, for anxiety, such imperfect options as antihistamines, buspirone, atypical antipsychotics, and mood stabilizers, all of which may be plagued with weak efficacy or major side effects. And no, I'm not forgetting psychotherapy, but most of the patients I see have previously or currently tried that. (Primary care and ER docs prescribing benzos may be another matter).

So when someone comes in specifically asking for a benzo, it is at least possible that they do so because benzos have in fact been head-and-shoulders above other drugs in treating their symptoms. Why do patients not come in specifically asking for drugs like Prozac or lithium? (Well, occasionally they do, but not often). Because for depression or bipolar disorder Prozac and lithium are not as clearly superior to competing options.

Another thing doctors tend to hate is patients availing themselves of another person's medication, even if at reasonable doses (this is often from a close friend or family member and therefore more like "in the house" rather than "off the street"). This certainly has its hazards and I don't condone it, but in itself, in the absence of other red flags, could imply a person straightforwardly seeking relief rather than trying to get high. Maybe I'm being psychotically naive here, but I don't think so. Sometimes I think doctors object to such behaviors more because they tend to bypass medical decision-making than because they guarantee a drug abuse problem.

I certainly don't view benzos as panaceas and do not hand them out in a cavalier fashion. And like most psychiatrists I think, I view alprazolam as representing a significantly higher risk of addiction and potentially heinous withdrawal. But unlike some doctors apparently, I don't view patients seeking benzos as prima facie wrong. Innocent until proven guilty.

It occurs to me that just as the prescribing of opioids is somewhat concentrated in specialized pain clinics, which helps to focus attention on potential abuse problems (by random drug screens, the tracking of prescriptions, etc.), clinics devoted to the treatment of anxiety could help to defuse much of the, well, anxiety, related to benzos. Obviously some psychiatrists, particularly in academic settings, specialize in treating anxiety, but the practice is not as widespread as that of pain clinics. Maybe I'll start an anxiety clinic--in my next life.

Old Wine, New Bottles

Human nature, essentially changeable, unstable as the dust, can endure no restraint; if it binds itself it soon beings to tear madly at its bonds, until it rends everything asunder, the wall, the bonds and its very self.

Franz Kafka, "The Great Wall of China" (trans. Edwin and Willa Muir)

"I have a new antidepressant for you...Seroquel XR." These encouraging drug rep's words were directed at a colleague in the office next door (happily I escaped). How did medications initially indicated for schizophrenia successfully migrate to bipolar disorder and now depression? The answer involves both the very porous psychiatric diagnostic system and the slack standards for FDA approval.

Last year, when Abilify was approved for "adjunctive" treatment of depression, patients who saw some commercials started asking for it. The advent of direct-to-consumer advertising on television a few years back allowed pharmaceutical companies to go around physicians at will, and this affected psychiatry more than other specialties due to diagnostic ambiguities. Now that Seroquel XR is indicated for acute treatment of bipolar depression, a number of patients will overlook the bipolar part and will start asking for Seroquel for depression, and a certain fraction of doctors will comply. In fact, Seroquel is widely used "off-label" for both anxiety and insomnia; the maker could probably obtain those indications too if it were worth the company's while.

I think the atypical-antipsychotics-for-bipolar-disorder-and-depression phenomenon represents the confluence of no fewer than five factors. The least interesting is the perennial and understandable drive of pharmaceutical companies to make as much money as possible by accumulating indications. Second, the drug companies are not required to conduct head-to-head trials of prospective drugs with drugs already known to be effective; they need only to demonstrate superiority over placebo. The result is that "new" treatments are not necessarily advances over already available treatments; they are merely different (the "me too" drug).

Third, in recent years the profession has undergone a crisis of confidence in both the safety and the effectiveness of standard antidepressants, particularly for bipolar depression, which is often resistant even to standard mood stabilizers like lithium and valproic acid. Fourth, and related to this, treatment-resistant depression remains a huge problem in general; depression is increasingly recognized as a major source of disability worldwide, and the bad news is that a significant number of patients do not get better even with aggressive treatment.

Why else has a treatment like electroconvulsive therapy, marred as it by stigma and side effects, remained a quite viable treatment for seventy years now? Severe depression can be grim indeed, and ECT still treats it better than anything else. So in the treatment of depression there is very much an effectiveness vacuum; patients urgently want help, and psychiatrists urgently want to help them, so anything at hand that can be shown to be better than sugar pills for depression is likely to be used, even if side effects--such as the high cost, weight gain, and other metabolic issues of atypical antipsychotics--can be alarming. Neither patient nor psychiatrist wants to hear, or to say, "There is nothing more I can do."

Fifth, atypical antipsychotics are both more convenient and, in a purely subjective sense, better tolerated than many alternatives. The standard mood stabilizers--lithium and the anticonvulsants--require monitoring of blood levels other laboratory tests, whereas atypical antipsychotics (unless one counts the lipid profiles that ought to be done to monitor metabolic side effects) do not. This can be an issue of compliance as well as convenience. And while the objective side effects can be dismaying, many patients are subjectively surprisingly tolerant of substantial weight gain (which only adds to the problem of course).

So while I'm thoroughly tired of the phrase "perfect storm," we have one here: a high potential for profit, malleable diagnostic boundaries, and a widespread eagerness to help and be helped with respect to a prevalent illness scourge. Helpful interventions might include ending televised direct-to-consumer drug advertising, requiring higher standards for FDA approval, devising antidepressants that are actually better than the ones we have now (I'm told people are working on that, but breath-holding is not advised), and when it comes to prescribing decisions, knowing when enough is enough.