[ed. note: Michael Jorrin, who I call Doc Gumshoe, is a longtime medical writer (not a doctor) who writes for us about medicine and health a couple times a month. He has agreed to our trading and disclosure restrictions, but does not generally write directly about investment ideas. His ideas, thoughts and words are his own, and you can see all his past pieces here. This is a follow-up to his earlier piece about depression, available here.]
Staring me in the face one morning last week is this New York Times headline: “Panel Urges Screening for Maternal Depression.” The U. S Preventive Services Task Force has recommended screening for depression during pregnancy and after giving birth. Just screening, mind you – these USPSTF guidelines don’t go on to recommend any specific type of treatment. The former USPSTF guidelines regarding depression recommended screening if the health-care provider – whether an individual physician or a larger practice – had the staff and facilities to provide some sort of treatment – unspecified – and those earlier guidelines, issued in 2009, did not specifically single out pregnant women and new mothers.
I have previously railed against the USPSTF with regard to some of their recommendations, such as that women below age 50 do not need to have mammograms, and that men of any age should not have PSA tests. My position on both counts is that omitting those procedures would lead to more deaths of breast and prostate cancer, and the data appears to support my viewpoint. The USPSTF’s position on both of those screening positions is that the harms might outweigh the benefits – yes, some deaths might be avoided, but there would be unnecessary biopsies and perhaps unnecessary surgeries, and that those harms were greater than the few deaths that might be avoided.
The depression screening recommendation, from the USPSTF’s perspective, entails no such risks. Depression screening would not in any circumstances lead to biopsies or surgery, and might help considerable numbers of women get treatment that could be beneficial and perhaps life-altering in a positive way.
Depression linked with pregnancy is a highly interesting subtype of depression, and looking at it in a bit more detail might be a good way to get into the details of treatment options.
The underlying causes of post-partum depression, as it is commonly referred to, are not entirely clear, although the major hormonal changes that take place during and after pregnancy are certainly factors. Another set of factors is the great dissonance between the love and joy and pride and sense of accomplishment that the new mother is expected to feel and manifest, and the stress and fatigue of coping with the new reality that has been visited on the new mother. The expectant / new mother cannot, cannot, cannot actually come out and say that she is anything but ecstatically happy at the birth of her child. If she is anything but ecstatically happy, if follows that she must be a bad mother. A bad mother and a bad person.
So, what’s going on, and what could possibly be done about it? It’s clear that pregnancy and childbirth are not, in and of themselves, bad-news external events, a normal response to which would be feelings of doom and gloom; therefore, something physiological must be going on to account for the bad feelings. But at the same time, bad feelings in response to what is supposed to be a joyful event inevitably produce conflicting inner feelings that somehow need to be sorted out and dealt with. Both the emotional dissonance and the brain chemistry need to be addressed.
A type of therapy that might help
We’re not talking here about a lengthy series of 50 minute sessions with a psychiatrist, in which your deeply-buried neuroses are painstakingly excavated, and the way these complicated emotional tangles mess up your life is examined, and the whole mess is eventually successfully exorcised. This takes forever, and your insurance won’t pay for it, and in most cases of depression, that’s not what is needed.
The term for a kind of therapy that has a pretty good track record of helping people with major depressive disorders is “cognitive behavioral therapy,” or CBT. This type of therapy focuses on concrete problems and tries to identify ways many people think and feel about those concrete problems, and the activities that they may engage in when they are feeling depressed about those concrete problems. These, all too frequently, are anything but helpful. Some people have a habit of engaging in highly polarized thought processes, e.g., if they experience a single setback in their jobs or personal lives, they conclude that they are total failures. Or, if the boss is short-tempered with them on occasion, it’s because the boss dislikes them, which, in turn, might be because they themselves are fundamentally not likeable. CBT may help to bring these persons around to the perspective that one setback does not constitute terminal failure, and that, just possibly, the boss is short-tempered with lots of people – just possibly, the boss is a nasty guy. The therapist may be able to help in dialing down the response from a focus on their own perceived shortcomings.
And if the individual with a depressive disorder engages in compensatory behaviors that are, to say the least, counterproductive – e.g., drug or alcohol abuse, acting in ways that erode valuable personal relationships, neglecting the essential activities of daily life such as eating poorly, ignoring medical issues, not getting out of bed, not paying bills – the therapist can also try to identify these aggravating behaviors and help the depressed individual understand the connections.
It’s no doubt possible, or even likely, that in some individuals the root of depression goes very deep, and that a few sessions of therapy focused on relatively superficial behaviors doesn’t do much good. Such individuals may benefit from the more extended (not to say nearly perennial) process of psychiatric therapy. It’s also certainly the case that major depression coexists with significant mental disorders, which are beyond the scope of short-term CBT and, in many cases, should definitely not be treated with conventional antidepressants.
But even in the case of the individual with a depressive disorder who can be helped to understand his or her own thought processes and emotional reactions and how they result in major and sometimes debilitating changes in mood, there is a brain mechanism that brings about these changes, and it is mediated by those brain chemicals that transmit impulses inside the brain, which is to say, neurotransmitters.
Let’s take a look at the drugs that have an effect on the brain chemistry of depression.
Antidepressants: what’s out there and what’s in the works?
First, a few words about serotonin, which everybody has heard about by now. I hope I can restrain myself from going into too much detail, but here’s a quick summary. Serotonin’s chemical name is 5-hydroxytriptamine, and you’ll sometimes see it referred to as 5-HT; this is especially useful when there’s discussion of the various serotonin receptors, of which there are more than a dozen. These are labeled as 5-HT1A, etc. We (and most other animals) make serotonin ourselves, and we store most of it in the digestive tract, where it promotes the constricting action that propels the food in our gut to its ultimate destination.
What we mostly make serotonin from is a protein called tryptophan, which is present in abundance in protein-rich food. Serotonin is active in regulating just about every form of physiologic activity in our bodies, particularly including digestion, emotion, cognition, motor, and autonomic actions. In the most general way, serotonin can be thought of as having an effect that is the opposite of norepinephrine. Norepinephrine stimulates the heart to beat faster, dilates the blood vessels so they can transport the greater cardiac blood flow, dilates the bronchi in our lungs so that we can take in more oxygen, and in general wakes us up and gets us ready for the daily fight with our antagonists in the great world. Serotonin mostly does the opposite. It slows the heart, constricts the bronchi, dilates the blood flow in the digestive system so that we can digest our dinner, and in general calms us down and prepares us for a good night’s sleep. The impulse to have a nap after our Thanksgiving dinner is said to be due, at least in part, to having consumed large helpings of turkey, which is rich in tryptophan.
Norepinephrine and serotonin release in our bodies varies with the Circadian rhythm. In the morning, the burst of norepinephrine gets us going, and then at night, serotonin, which is a precursor to melatonin, slows us down. It promotes the opposite of the notorious “fight-or-flight” response. It makes intuitive sense that dialing down that fight-or-flight response would improve our mood; however, explaining precisely how serotonin accomplishes this is best left to the neurologists who analyze brain scans, determine which regions of the brain are associated with which brain functions, and are able to detect the activities of various neurotransmitters in these brain regions. The relationship between an electrochemical reaction and a change in mood is still, from my perspective, a deep mystery. Fortunately, we don’t have to understand precisely how that link functions. It may be enough to accept that there is a link, and that, mostly, it works fairly well.
So it should be no surprise that most drugs used as antidepressants in some way interact with serotonin. We’ll look at the classes of antidepressants, describe their mechanisms of action (MOAs), and discuss some of the agents in each class. Before we get into this, I want to make the point that there is no good evidence that any single antidepressant works better in most patients than any other. In an individual with a major depressive disorder, one drug might hit the spot, causing meaningful improvement in mood and behavior, while another drug might miss by a mile. However, the differences observed between individuals have not been generalizable to the overall population of persons with MDD. Some classes of antidepressants are more associated with a range of side effects, and the preferences of clinicians, as well as the popularity of some agents, are frequently based on tolerability rather than pure efficacy.
Selective serotonin reuptake inhibitors (SSRIs)
Selective serotonin reuptake inhibitors (SSRIs) are the most commonly prescribed agents for the treatment of major depressive disorders. The site at which they function is the tiny space between neurons – the synaptic cleft. One neuron, the presynaptic neuron, dispatches molecules of serotonin via a dendrite (a long, thin fiber) to the synapse, where the serotonin is released, to be absorbed by the postsynaptic neuron. The serotonin molecules are admitted to the postsynaptic neuron by ion channels, which recognize the polarity of the serotonin and pass them through the gates. But a certain quantity of the serotonin remains in the synaptic cleft. There seems to be a sort of “waste-not-want-not” principle operating here, because in the ordinary course of events, the serotonin molecules are taken back into the pre-synaptic neuron, and do not pass through into the postsynaptic neuron. The SSRIs prevent the reuptake of the serotonin by the presynaptic neuron by occupying the reuptake transporter. This leaves more serotonin in the synaptic cleft, for absorption into the postsynaptic neuron.
Essentially, the SSRI permits the serotonin to do its immediate job rather than being saved for next time. SSRIs are the recommended first-line treatment for MDD, according to a number of influential guidelines, including the Adult and Adolescent Depression Screening, Diagnosis, and Treatment Guidelines issued by the Group Health Cooperative. Drugs in this class include citalopram (Celexa, from Forest Pharmaceuticals, FRX), escitalopram (the enantiomer of citalopram, named Lexapro, also from Forest), fluoxetine (Prozac, from Eli Lilly, LLY), sertraline (Zoloft, from Pfizer, PFE), and paroxetine (Paxil, from GlaxoSmithKline, GSK).
(I have mentioned enantiomers in previous pieces. Complex molecules can exist in conformations that are mirror images of each other. The two mirror images are chemically identical, but they may function in different ways because of their shape. In the case of SSRIs, one conformation may fit snugly into the reuptake receptor, blocking it from accepting any serotonin molecules, while the other conformation does not fit. Enantiomers consist of only one of the two possible mirror image conformations. In some cases, one enantiomer can be more effective because it is not accompanied by the ineffective version. In some cases, however, the so called racemic compound, including both enantiomers, can be just as effective; pharmaceutical companies sometimes market enantiomers as a way of prolonging patent life.)
The SSRIs were initially hailed as miracle drugs, especially Prozac, which was treated by some media mavens as the magic pill that would put an end, once and for all, to depression. All we had to do was get on Prozac, and our blues would fade into glowing shades of pink and gold. Well, it’s not quite that easy, but on the other hand, the effects of the SSRIs were evident, and, compared with the previous classes of antidepressants (which we’ll discuss later), they were relatively free from adverse effects. They helped many, many persons manage their depressed emotional state, especially when there was a little additional therapeutic help.
The SSRIs as a class are considered to be well-tolerated by most clinicians, which is why they have become among the most frequently-prescribed drugs. As Doc Gumshoe reported in the previous post about depression, about 7% of all visits to physicians of any kind, whether general practitioners or specialists, result in a prescription for an antidepressant. There are, however, some general effects associated with any drug that increases the effects of serotonin. There may be a tendency to weight gain, and this may be more pronounced with some agents (i.e., paroxetine or Paxil) than with others. This is a natural consequence of serotonin’s interaction with the GI tract – serotonin helps us do a better job of extracting nutrients from our food. And there’s a general tendency to induce sedation, which is common in all the agents that involve boosting the activity of serotonin.
A more troubling group of side effect is the effect on sexual function. More than half of patients taking SSRIs report some sexual dysfunction, including loss of sexual drive, failure to attain orgasm, and erectile dysfunction. In the Doc Gumshoe piece about “Viagra for Women,” I observed that part of the mechanism of action of that drug was to boost norepinephrine and inhibit serotonin activity as a way of increasing sexual drive. Increasing serotonin activity would seem to be a sure-fire way to diminish sexual function. However, being gloomy and depressed probably doesn’t do a whole lot to fire up sexual drive either, so maybe an effective approach is to deal with one problem at a time.
The damping effect on sexuality is specific to the classes of agents that raise serotonin activity. However, an adverse effect of much greater concern, which can occur with antidepressants of all classes, is triggering of episodes of mania. This happens when an individual with bipolar disorder is incorrectly diagnosed as being affected by depression alone – what is termed “unipolar” depression to distinguish it from bipolar disorder, or what was formerly called “manic depressive” disorder. When such bipolar patients are misdiagnosed and given antidepressants, as many as 40% have manic episodes, and some of them experience rapid cycling between manic and depressive phases of their disorder. Rapid cycling is officially defined as four or more changes from depression to mania or the reverse during one year, but the changes can occur as frequently as every few days or even during the course of one single day. Rapid cycling is an exceedingly difficult-to-treat condition and is managed as a psychosis. Simply withdrawing the antidepressant drug is not the easy answer, because sudden antidepressant discontinuation brings its own significant adverse effects. The risk of misdiagnosis of individuals with bipolar disorder is another reason to be wary of having physicians who are not familiar with the mental health territory prescribe antidepressants.
Finally, any drug that increases serotonin levels presents the risk of bringing about a condition called serotonin syndrome, which can cause a number of symptoms, including agitation, restlessness, insomnia, and confusion. It is not common, and mostly occurs only when the dose of the agent that raises serotonin levels is especially high. The symptoms usually resolve after dose reduction or gradual withdrawal of the serotonin-boosting drug.
Serotonin norepinephrine reuptake inhibitors (SNRIs)
The guiding principle here is that a small boost in norepinephrine is a good thing. While the SSRIs shift the overall neurotransmitter balance towards serotonin, the SNRIs shift it back, not enough to diminish the benefit of increased serotonin activity, but perhaps enough to offset some of the sedating effects of serotonin, and perhaps offset the negative effects on sexuality.
SNRIs are a newer class of antidepressants and, as such, have received a good deal of recent attention, after the initial bloom was off the SSRIs. The SNRIs include venlafaxine (Effexor, developed by Wyeth and marketed by Pfizer), desvenlafaxine (Pristiq, also Wyeth/Pfizer), which is the enantiomer of venlafaxine, and duloxetine (Cymbalta, from Eli Lilly).
To this list should be added vilazodone (Viibryd, currently from Allergan, AGN, perhaps about to be acquired by Pfizer?) Vilazodone is somewhat different in its MOA from the other agents in the SNRI class in that it also addresses one specific serotonin receptor, denominated 5HT1A. There is reasonable evidence that the 5HT1A receptor is involved with a number of responses in the brain, such that an agent that interacts with that receptor may demonstrate benefits in other disorders, including the motor symptoms in Parkinson’s disease and some symptoms in schizophrenia.
Which antidepressants lead the pack in terms of sales?
Together, the SSRIs and the SNRIs comprise the majority of the most-prescribed antidepressants. The top 10, in terms of total numbers of prescriptions written, are:
- Duloxetine (Cymbalta, Eli Lilly), an SNRI. In addition to being indicated for the treatment of MDD, duloxetine is also FDA indicated for treatment of neuropathic pain associated with diabetic peripheral neuropathy, generalized anxiety disorder, fibromyalgia and chronic musculoskeletal pain.
- Desvenlafaxine (Pristiq, Pfizer), an SNRI.
- Vilazodone (Viibryd, Alllergan), an SNRI.
- Citalopram (Celexa, Forest), an SSRI.
- Sertraline (Zoloft, Pfizer), an SSRI. Sertraline is also indicated for the treatment of obsessive-compulsive disorder (OCD), panic disorder, post-traumatic stress disorder, premenstrual dysphoria, and social anxiety disorder.
- Fluoxetine (Prozac, Eli Lilly), an SSRI. Fluoxetine is also indicated for the treatment of OCD, bulimia, and panic disorder.
- Escitalopram (Lexapro, Forest), an SSRI. Escitalopram is also indicated for the treatment of generalized anxiety disorder.
- Trazodone (formerly Desyrel, brand discontinued; generic manufactured by Keltman Pharmaceuticals). Trazodone is a serotonin agonist as well as a reuptake inhibitor.
- Paroxetine (Paxil, GlaxoSmithKline), an SSRI. Paroxetine is also indicated for panic disorder and OCD.
- Venlafaxine (Effexor, Pfizer), an SNRI. Venlafaxine is also indicated for generalized anxiety disorder and panic disorder.
To that list one more agent should be added. Bupropion (Wellbutrin, Zyban) was originally developed by Burroughs Wellcome, which eventually became GlaxoSmithKline. A number of generic formulations are available, manufactured by about a dozen different pharma outfits. The Zyban-labeled version can be used to help people who are trying to quit smoking. Bupropion is also indicated for the treatment of seasonal affective disorder. The MOA of bupropion is somewhat murky, but it is one antidepressant that appears to have little or no effect on serotonin activity. Nonetheless, it has been shown to possess equivalent efficacy in MDD. An important point of distinction between bupropion and most of the rest of the antidepressant field is that bupropion is not associated either with weight gain or with sexual dysfunction. The Group Health guidelines, which we cited earlier, place bupropion as a second-line antidepressant agent, after four established SSRIs. A cautionary note is that bupropion is associated with a low risk of delusions and hallucinations, and should not be used in patients with psychotic disorders.
That list of the most popular antidepressants should make it clear that the health-care community is by no means reluctant to jump on the new bandwagon. The SNRI category has vaulted to the top of the list, despite much higher prices and scant evidence of superior efficacy or safety. But in a treatment area where solid evidence supporting any single agent or even class of agents is absent, clinicians are eager to try something new.
Duloxetine’s huge sales figures – not only in terms of numbers of prescriptions (about 18 million per year), but in dollar volume ($5.2 billion) – is due largely to the fact that it is indicated for treatment not only of MDD, but of several other medical conditions. And that, in turn, is at least partly due to aggressive marketing by Eli Lilly. Lilly’s marketing strategy has included going after as many FDA indications as possible for their prized drug, and it has obviously paid off.
But there’s one more drug that demands attention. It is aripiprazole (Abilify, Bristol-Myers Squibb, BSM). It was approved in 2013 as add-on treatment for patients with MDD who do not respond satisfactorily to first-line antidepressants such as SSRIs or SNRIs. That, however, is not aripiprazole’s primary indication. Aripiprazole is used to treat adolescent schizophrenia and also autistic disorder in children. And, although aripiprazole is not specifically indicated for bipolar disorder, it can be used to treat manic episodes in individuals with MDD. This combination of indications, coupled with BSM’s marketing, has catapulted aripiprazole to the top of the nation’s drug sales figures, with $6.2 billion in sales.
A few more classes of antidepressants: monoamine oxydase inhibitors (MAOIs)
MAOIs are the oldest class of antidepressants, having been first used as treatment for depression in the 1950s. The basis for using this class of agents to treat MDD is that serotonin, norepinephrine, and dopamine are all monoamine neurotransmitters – monoamine meaning nothing more complex than that they have a single amino group. Monoamine oxydase is an enzyme that triggers the breakdown of these neurotransmitters, and inhibiting that enzyme leads to an increase in the levels of those neurotransmitters and an improvement, it is hoped, in mood and emotional tone. None of the current widely used antidepressants are in this class, which has a number of significant disadvantages, among them a delay of a week or more before any effectiveness is experienced. A more serious potential adverse effect occurs when MAOIs are taken with foods containing tyramines, which are present in many kinds of cheese, especially aged cheeses, and also many kinds of processed meats and sausages. The effect can be sudden and dangerous increases in blood pressure. Nonetheless, MAOIs are sometimes given to patients who do not respond to other antidepressants, and there is a phalanx of clinicians who swear by MAOIs.
Tricyclic antidepressants (TCAs)
TCAs have also been around for a good long while. The first drug in this class was chlorpromazine, developed in the 1950s, and found to be quite effective in managing psychotic symptoms. Chlorpromazine was one of the first psychoactive drugs, and was extremely important in the after-care of patients released from mental institutions; prior to the availability of these early psychoactive drugs, mental patients were often confined in mental hospitals for life, but chlorpromazine and other drugs permitted some mental patients to live “on the outside” in relative stability.
The use of TCAs to treat depression came a bit later, in the 1960s, and some TCAs continue to be used today. They are considered to be similar in efficacy to SSRIs or SNRIs, although these later classes of agents are preferred, mostly on the basis of more favorable side effects profiles. Amitryptiline (formerly Elavil, AstraZeneca, AZ), doxepin (Sinequan, Pfizer), and imipramine (Tofanil, Malinkrodt Chemical), among others, continue to be used. All are available generically for a modest price. TCAs do come with a panoply of side effects some of which are non-trivial, including seizures, both hypo- and hypertension, cardiac arrhythmias, and intra-ocular pressure.
Shifting our focus to recent developments
A defect in the current regiment of antidepressant warriors is that it takes awhile for them to kick in – sometimes almost a month. In contrast with those slow-pokes, it has been known for some time that ketamine, a drug used as an anaesthetic, also has antidepressant effects, which appear very rapidly. Unfortunately, ketamine is highly addictive, so, sure, you can use it once or maybe twice, but more than that and there is a definite risk of getting hooked. In fact, ketamine is used as a party drug – it’s known as “Special K.”
Ketamine is an antagonist of the n-methyl-d-aspartate (NMDA) receptor. Scientists at Johns Hopkins have announced that they have identified an agent that works on that same group of proteins, but takes effect at a different point in the chain reaction and has fewer side effects. In studies in mice, the drug demonstrated antidepressant effects in as little as half an hour. The drug, labeled at present as CGP3466B, had previously been found to prevent cocaine addiction in mice (who knew that mice could become addicted to cocaine?). It has also been studied, without success, in Parkinson’s disease and amyotrophic lateral sclerosis (Lou Gehrig’s disease), however, those studies demonstrated good safety data in humans.
Several pharmaceutical companies, including Johnson & Johnson and Allergan, have been looking into finding a drug that demonstrates some of the same antidepressant characteristics as ketamine. There’s something highly tempting about the prospect of a drug that can banish feelings of doom and gloom in half an hour. However, Doc Gumshoe’s skeptical side wonders whether simply taking a happy pill with a rapid onset of action would not somehow seem to make it unnecessary for the person experiencing the depressive episode to attempt to understand what was going on? The thoughts and behavior patterns related to depression wouldn’t have to be addressed – just over-ridden.
Meantime, several pharmaceutical companies have discontinued work on drugs that targeted the supposedly promising fatty acid amide hydrolase (FAAH), an enzyme that regulates cannabinoid receptors, which has been thought to be a promising avenue for the management of pain and anxiety. This comes after a clinical trial of a similar drug in France in which six volunteers were hospitalized and one died. The pharm outfits that have suspended work on drugs with similar mechanisms include Johnson & Johnson, Pfizer, and Sanofi. We’ll see if interest in that pathway resumes and what comes of it.
… and finally, St John’s wort!
Doc Gumshoe is only mentioning St John’s wort because I know some of you would surely ask. I will lead off by saying that there’s no question that St John’s Wort has some antidepressant efficacy, probably by elevating serotonin levels. However, like other herbs, it contains a lot of chemicals that can have physiologic effects. One of the chemicals, hyperforin, induces reuptake inhibition of monamines such as serotonin, norepinephrine, and dopamine. Which is to say, it functions very much like an MAOI. But it has a wide range of other potential physiologic effects, and the list of drugs that St John’s wort interacts with is long and frightening. It includes antiretrovirals used to treat HIV, anti-anxiety drugs, contraceptives, a number of cardiac drugs, drugs to treat migraines, and several others. Add to those warnings the usual warning that when you take an herbal supplement, you don’t really know exactly what you’re getting, and Doc Gumshoe’s “Beware” sticker is firmly affixed.
Where are we headed?
There’s no question that depression is a huge global problem, affecting not only the lives of billions, but the economies of nations large and small, and health-care systems everywhere. One would think that the prevalence of depression – or, to use the accepted term, episodes of major depressive disorders – would track living conditions in various nations, but this doesn’t seem to be the case. Nations and regions where standards of living are high have MDD prevalences that are comparable to nations and regions at the other end of the scale. The typical patient receiving intensive treatment for an episode of MDD is not a poverty-stricken individual living in a war-ravaged nation, but a relatively prosperous person living in a developed nation experiencing symptoms of depression that cannot be explained by exterior events. The large number of such patients presents opportunities both for the pharmaceutical industry, which seeks to address their conditions with drugs, and for the therapeutic community, that offers some form of “talking” therapy. And there is little question that, in the right combination, and pursued with intelligence and persistence, the combination of drugs and talking therapy offers relief, and even remission, for a considerable proportion of such patients – maybe, as we saw in the STAR*D study, as many as two-thirds.
And yet, it’s difficult to say exactly how this therapy works. The efficacies of the many classes of antidepressants are comparable. We increase serotonin levels – not too much! – by a variety of means, and it helps. We increase norepinephrine levels, also not too much, and that helps also. We tinker with other neurotransmitters, and that helps as well. The connection between the effects of neurotransmitters on the specific brain areas can be detected by imaging, but the change in mood and emotion remains a mystery.
Surely, because of the scope of the problem, research will continue, and it’s possible and even likely that new classes of antidepressants will emerge with genuinely superior efficacy. But in the meantime, we should not forget the advice of Robert Burton, 400 years ago, who suggested good food, music, enough sleep, meaningful work, and talk with friends.