[ed note: Michael Jorrin, who I like to call Doc Gumshoe, is a longtime medical writer (not a doctor) who shares his non-investment thoughts with us a couple times a month. I suppose it’s a coincidence that he’s looking at depression at a time when we’re all a bit depressed about the market, but maybe it’s a good time for us all to learn more about this disorder. Michael’s past columns are here if you’d like to see more.]
W. C. Fields famously pronounced the following cure for insomnia: “Try to get a lot of sleep.” Following his eminent lead, we could say that the recipe for curing depression should be “Cheer up!” Oh, sure, easy to say.
And yet, that’s precisely the goal of treatment. The question is, and always has been, how to attain that goal, or even come close to it.
I want to start out by asserting that the mere fact that we now commonly discuss ways of treating depression is a sign of genuine progress. Within the memory of many of us, admitting to being depressed and seeking some kind of treatment for depression was seen as a sign of weakness of character. People were supposed to accept adversity. As Longfellow put it,
“Be still, sad heart! and cease repining;
Behind the clouds is the sun still shining;
Thy fate is the common fate of all,
Into each life some rain must fall,
Some days must be dark and dreary.”
Depression, as a condition affecting a considerable number of our fellow humans, has been recognized since the days of the ancients. It was known as “melancholia,” based on the Greel words for black (μέλας, melas) and bile (χολή, khole), and was thought to be caused by black bile, one of the four humors that essentially ruled our lives. (The other three humors were blood, which made us sanguine or hopeful, phlegm, which made us phlegmatic or slow and thoughtful, and yellow bile, which made us bilious or disagreeable.) Robert Burton, in his Anatomy of Melancholy, an eloquent and poetic 17th century work, suggested good food, music, enough sleep, meaningful work, and talk with friends. Not too bad a regimen to induce cheering up.
Right down to the middle of the 20th century, and persisting even into the present, there has been a tendency to try to distinguish real genuine bona fide depression from unhappiness over external events. The German psychiatrist Kurt Schneider distinguished between what he termed “endogenous depression,” meaning that it arose from an internal mental state, from “reactive depression.” The fourth edition of the Diagnostic and Statistical Manual of Mental Disorders, the DSM-IV, for example, excluded bereavement in its definition of major depressive disorder – if a person had lost a relative or a close friend in the immediately preceding period, then no matter how else that person felt, he or she was not to be diagnosed as being depressed. The death of a person that one loves should presumably be considered in the same category as the rain that must fall into each life. You were supposed to get over it.
The new edition, DSM-5, excludes that exclusion. Some people have a hard time getting over such events, and other “real world” events and situations as well; something is going on inside them that exacerbates or prolongs the feelings. The distinction between what happens in the outside world and what’s going on in our brains is not so clear. Reactions to real world events can result in endogenous depression; neurotransmitters do their thing more or less regardless of the cause.
Looking at depression from the outside
Let’s defer thinking about depression from the perspective of the individual affected by this condition, and take a look at the impact of depression on our society. In a word, it is enormous. Before throwing a few quick facts at you, let me first specify that these data are based on a condition defined as “major depressive disorder,” according to the criteria in the previous edition of the DSM, except that unlike those criteria, no exclusions were made for medical illness, bereavement, or substance abuse disorders. A major depressive episode is defined in DSM-IV as a period of two weeks or longer during which there is either depressed mood or loss of interest or pleasure, and at least four other symptoms that reflect a change in functioning, such as problems with sleep, eating, energy, concentration, or self-image.
You may be thinking that the DSM-IV definition is rather broad – after all, at some point nearly everyone has a problem with one of those categories. But on closer examination, how often do we have problems with four of those all at the same time, coupled with a loss of pleasure? That’s when the little routine problems that plague us every day morph together and become a pathologic condition.
So here are those few quick facts, from the National Institute of Mental Health:
For 2014, the latest year for which we have data, 6.6%, or about 16 million adults in the US had at least one episode of a major depressive disorder (MDD). MDD affected nearly twice as many women, 8.2%, as men, 4.8%. The 18 to 25 year-old cohort has the highest prevalence, 9.3%, and this drops to 7.2% for 26 to 49 year-olds and to 5.2% for people over 50 years old. Whites have a slightly higher prevalence (7.1%) than blacks (5.4%) and Hispanics (5.2%). And Asians have the lowest prevalence, 4.2%.
Rates of MDD are higher in persons whose sexual orientation is other than heterosexual, and the lifetime prevalence is nearly double in lesbian, gay, and bisexual persons, 71.4% versus 38.2% in heterosexuals.
The estimated total annual cost of depression in the US is upward of $80 billion. Unlike the costs associated with other illnesses, which are largely related to the cost of treatment, about 70% of the costs of depression are not treatment costs, but related to lost productivity. According to a study by the Epidemiologic Catchment Area, persons with an episode of MDD were 27 times more likely to miss work than those without this disorder, and when they were at work, they were much less likely to be productive, a phenomenon termed “presenteeism,” meaning, more or less, that the person with MDD is physically present, but not able to function normally.
Most of those lost productivity costs are borne by employers, without, of course, any reimbursement and not much recourse. A big chunk of the costs are, however, absorbed by the health-care system. Persons with major depressive episodes are about seven times more likely to present to hospital emergency departments than those without MDD, and many of those ED visits are not covered by health insurance.
If everyone with an episode of MDD sought medical intervention, the costs would be far higher. However, based on estimates by the National Suicide Prevention Lifeline, fully 50% of persons who experience major depression do not seek treatment of any kind.
Before we go on to the real meat of this piece, which is how do we treat or manage depression, no matter how it is defined, I need to inject a dose of Doc Gumshoe skepticism regarding some of those statistics.
It should be evident that the great majority of diagnoses of depression are based on the patient’s self-report, and that, regardless of the effort that has been made to create accurate questionnaires for the assessment of depression, individual patients will answer those questions according to their own disposition and experience. Are women really twice as likely to experience major depression as men? Or is it possible that men are about half as likely to acknowledge the feelings of depression as women? Or do men have more opportunities to overcome their depression than women? Or do women face greater real-world problems than do men? Similarly, why do Asians in the US have lower rates of MDD than any other ethnic group? Are they really less depressed, or are there cultural factors that inhibit Asians from admitting that they are depressed? (This, by the way, does not seem to be the case in China, where very high rates of behavioral disorders and mental illness seem to prevail, according to Western mental health professionals who have spent time there.)
Higher rates of MDD are intuitively more understandable in individuals with other than heterosexual orientation, since they are more likely to have encountered bullying, abuse, and condemnation, sometimes since childhood, and sometimes from close family members.
A factor that affects most of the statistics noted above is fear of being stigmatized. A depressed individual might prefer to present him/herself as feeling low from time to time rather than accepting the diagnosis of having a major depressive disorder. The MDD diagnosis might affect the way he or she is regarded in the workplace. Better to soldier on and hope that you feel better than to be labeled as sick.
In turn, the fear of stigmatization is likely to be a factor in the low rate of persons with major depressive disorder who actively seek treatment. That is likely to be the single major factor in how MDD is treated, or not treated. But there are certainly a number of other factors that have important consequences.
Factors that affect how major depressive disorder is treated
In spite of the reduced emphasis on the difference between “endogenous” and “reactive” depression, as demonstrated in the DSM-5 changes, the perception persists that some manifestations of depression are physiologic while others are predictable responses to adverse life situations. The former are more apt to be considered appropriate for medical treatment, while the latter are thought to be more effectively managed through some form of psychotherapy. We’ll return to that particular subject later – enough to say for right now that dividing the patient stream on those grounds is not always supported by the evidence.
However, in purely practical terms, patients do some of that dividing of their own accord. Many, many patients discuss their overall mental state with their physicians, and good doctors encourage this discussion. E.g., “Doc, I’ve been feeling kind of down lately. Do you think you might be able to give me something that would help?” This has been greatly magnified by the perception, fostered by pharmaceutical advertising and publicity, that all you need to banish your depression is that magic pill, whether Prozac or one of its many successors. And many, many primary care and family physicians write prescriptions for those magic pills. In fact, about 7% of all PCP visits result in an antidepressant prescription. About 53 million antidepressant prescriptions were written in 2014, and the great majority of those were written by non-psychiatrists.
Of course, an obvious reason for this is that there simply are not enough psychiatrists in the US for every person with possible symptoms of MDD to see a psychiatrist, or a psychologist who is licensed to prescribe drugs, without a long, long wait. The number of psychiatrists in the US is about 49,000, and they are by no means evenly distributed across the nation. In contrast, there are about half a million MDs in some form of primary practice. And it’s not just primary care and family physicians who prescribe antidepressants. The patient with fibromyalgia might get it from a rheumatologist; the COPD patient from a pulmonologist; the woman going through menopause from a gynecologist; the teenager with really bad acne from a dermatologist. Any licensed physician can and will write a prescription for an antidepressant.
That, in itself, is by no means a terrible thing. Many, maybe most of these patients, get some relief from the prescription. But there are obviously problems. The non-psychiatric physicians are not likely to be as aware of the range of adverse effects that may ensue from taking any and all of these antidepressants, and they are far less likely to monitor the patients for whom they prescribe these drugs than the physicians with specific training in the mental health field – after all, their main focus is the presenting symptom – fibromyalgia, or COPD, or acne – and not the depressive symptoms.
The biggest problem may be that the fundamental condition, the major depressive disorder itself, is not carefully examined. A complicating factor, beyond the ken of many physicians, is that depressive symptoms frequently occur in individuals with a number of other disorders or conditions that, if identified and evaluated by a competent practitioner, would make it unlikely that the patient would receive a diagnosis of MDD. In other words, feeling down or in the dumps is not, by itself, a sufficient medical reason to get a prescription for an antidepressant. Some of the conditions that specifically exclude a diagnosis of MDD (according to DSM-5) include alcohol or substance abuse, or what is termed a “Mood Disorder Due to a General Medical Condition,” although any of these may contribute to the onset of MDD.
Yes, it’s murky. And to add to the murk, there is often an overlap between MDD and other mental disorders, such as bipolar disorder, attention-deficit/hyperactivity disorder (ADHD), and what they call schizoaffective disorder. Bipolar disorder in particular, where depressed moods can alternate with manic moods (people with this condition used to be called “manic-depressive”), is a challenge, because treatment with antidepressants can aggravate the symptoms of this disorder, triggering a state called “rapid cycling,” where the depressive and manic phases alternate much more rapidly than normal. The clear advice when a patient is given an antidepressant, and the mood quickly switches from depressed to manic, is to withdraw the antidepressant. But this also can have adverse consequences.
One would think that a competent physician, regardless of specialty, would seek to rule out medical causes. Thyroid disease, systemic infections, hormonal abnormalities, as well as medication side effects, all can have potent effects on mood, resulting in symptoms that can be mistaken for MDD.
A mental health professional, whether psychiatrist, psychologist, psychotherapist, or any other, is also more likely to look at the numerous associated life-style conditions that are frequently linked with depression. In some cases, these contribute to the major depressive episode, and in some cases, MDD leads persons to those habits and conditions of life. To name just a few, individuals with MDD tend to be affected by greater economic uncertainty; they are more likely to lack a close significant life partner; they are more likely to abuse drugs or alcohol; and they are more likely to develop chronic life-style-related diseases linked to smoking or obesity. Mental health professionals, not being sorcerers, cannot wave a wand and solve these problems, but they can discuss them with their patients who are experiencing MDD, perhaps leading to understanding and change.
So, how do health-care professionals, regardless of their specialization, evaluate whether a person is affected by depression, and, if so, how severe?
Instruments for assessing depression
The word “instruments” may create a misleading image in your mind – it certainly did in mine when I first heard it used in his context. No, these instruments are not machines that attach to your body to detect your state of mind, nor are they quick brain scans that zero in on the brain region that governs your mood. They are questionnaires, and they have been “validated” by comparing the results with the observations of skilled and experienced mental health professionals, so that if a patient’s score on a questionnaire or a scale indicates that the patient is experiencing an episode of major depression, the chances are very good that a psychologist or psychiatrist would come to the same conclusion.
Among the most widely used are the PHQ-9, which is a nine-question Patient Health Questionnaire, the CESD-R, which is the 20-question Center for Epidemiologic Studies Depression Scale Revised, the HAM-D, which is the Hamilton Rating Scale for Depression, and the MADRS, which is the Montgomery-Asberg Depression Rating Scale.
The PHQ-9 is the quickest and easiest. If primary care physicians use any instrument, this is the one most likely to be used. It poses 9 questions, as follows:
Over the last 2 weeks, how often have you been bothered by any of the following problems? (The patient is asked to select one of the following answers: “Not at all,” “Several days,” “More than half the days,” or “Nearly every day.”
- Little interest or pleasure in doing things
- Feeling down, depressed, or hopeless
- Trouble falling or staying asleep, or sleeping too much
- Feeling tired or having little energy
- Poor appetite or overeating
- Feeling bad about yourself – or that you are a failure or have let yourself or your family down
- Trouble concentrating on things, such as reading the newspaper or watching television
- Moving or speaking so slowly that other people could have noticed. Or the opposite – being so fidgety or restless that you have been moving around a lot more than usual.
- Thoughts that you would be better off dead, or of hurting yourself in some way<
Each answer is scored from zero, for “Not at all,” to 3, for “Nearly every day.” A total score from 15 to 19 indicates that the patient is experiencing a moderate episode of MDD, and 20 or above, a severe episode.
The CESD-R scale asks similar, but more detailed questions. The HAM-D is designed to be used by mental health professionals, and it covers the same general topics as the two questionnaires, but adds topics such as GI and sexuality symptoms, hypochondriasis, and symptoms likely to be characteristic of frank mental illness. Clinical trials of antidepressant drugs and related psychoactive drugs are more likely to use the HAM-D or the MADRS than the simpler instruments.
Having established that a person is in the grip of a major depressive episode, we come to the really big question:
At best, how would major depressive disorder be treated?
Just because I have been expressing Doc-Gumshoe-like skepticism about psychopharmacology does not mean that I am a partisan of psychotherapy instead of medication. The gloom-inducing data indicate that neither approach could be characterized as highly successful. Both psychotherapy and psychopharmacology are moderately successful at best, and the success rates for each are about the same.
However, an encouraging finding is that when the two approaches are used at the same time – psychotherapy and antidepressant medication – the results are superior in a number of respects. A large meta-analysis, of 52 studies with 3,623 patients, concluded that combined treatment resulted in clinically meaningful and statistically significant improvement over either form of treatment alone. The size of the effect was moderately large; the benefit of combined treatment compared with placebo was about twice as large as the benefit of pharmacotherapy compared with placebo. The authors conclude that the effects of psychotherapy and pharmacotherapy are largely independent of each other, with each contributing about equally to the effects of combined treatment. The effects were found to remain strong and significant for up to two years after treatment. (Cuijpers P et al, World Psychiatry 2014)
Generally speaking, when comparing psychotherapy with drug therapy, one of the important differences is that the benefits of psychotherapy are more enduring. A meta-analysis comparing the rates of relapse more than two years after treatment, with follow-up continuing to the four-and-a-half year mark, found that the rates of relapse were somewhat lower in patients treated with psychotherapy than in those receiving pharmacotherapy – about 50% for psychotherapy versus about 70% for pharmacotherapy.
Evaluating the effects of any treatment for depression, or for any so-called “mood disorders,” for that matter, is extraordinarily difficult. Strictly speaking, it is not possible to compare psychotherapy with placebo; the best that can be done is to recruit a comparable group of subjects, follow them in the same way that the treatment group is being followed, employing the same instruments to attempt to measure their level of depression, and comparing those results with the results for the treatment group.
But even in studies of antidepressant drugs, measuring the effects of the drug accurately is difficult. Yes, there can be a genuine placebo group that is given dummy pills. But both groups are observed and tested, and individuals in the placebo group get the same level of attention as do the subjects who are receiving the active drug. Perhaps as a result of this attention and interaction, the placebo response in antidepressant trials is exceptionally high, sometimes so high that it is difficult to claim a significant and meaningful difference between the active drug and placebo. Needless to say, this is a nightmare for drug developers, especially if they have identified an agent with a promising mechanism, that is well tolerated by patients, and that has demonstrated strong results in earlier, non-placebo-controlled trials. A meta-analysis of antidepressant trials found that on average, in published placebo-controlled trials, the active drug was about 37% more effective than placebo. This is considered to be on the low side of the moderate effectiveness range.
Even though, strictly speaking, the benefit of a drug is the difference between the change from baseline in patients treated with active drug compared with placebo patients, I venture to suggest that the overall benefit with antidepressant treatment is appreciably higher than that. Instead, perhaps the benefit should be calculated as the difference between treatment and no treatment whatsoever. As I pointed out earlier, all the subjects in these placebo-controlled trials have health-care interactions that could well qualify as treatment, especially compared with the total absence of treatment, which is the experience of about half of all persons affected by major depression.
That was the optimistic interpretation, namely, that antidepressants can work better than the clinical trial data indicate. However, the less optimistic interpretation is that the actual experience of many patients taking antidepressants is likely worse than the clinical trial data, because of the very large proportion of patients who get antidepressant prescriptions from physicians without going through any of the recommended procedures for establishing a baseline index for depression, without monitoring or follow up, and with minimal attention to the incidence of adverse effects.
A strategy to improve outcomes in patients with MDD
It hasn’t escaped the attention of health professionals in the field that patients with major depression don’t have very good outcomes. If the goal is remission – and, after all is said and done, what else should the goal be? – the proportion of patients attaining this goal is rather low. In patients with chronic MDD, about 30% at most achieve remission after 12 weeks of treatment, and about 40% of patients with MDD continue to meet DSM-IV criteria after one year of treatment. Most clinical trials, however, do not state remission as an outcome measure. Instead, they base success or failure on improvement on one of the standard depression measurement scales. And, as I noted earlier, the rate of relapse after antidepressant treatment is very high – about 70% of patients treated with antidepressant therapy alone had experienced a recurrence of depression by four and a half years after their initial treatment.
Another defect of clinical trial data is that the subjects in these trials are poor matches for the general population of persons with MDD. Clinical trial subjects are volunteers recruited through advertisements, meaning that they are already aware that they are experiencing depressive episodes. But they tend to have uncomplicated depressive disorders – minimal comorbid medical or psychiatric conditions, and not associated with substance abuse or suicidal ideation or behavior. That excludes a large number of patients, and those tend to be the most complicated and difficult to treat patients. In short, most clinical trials do not evaluate treatment for MDD under “real world” conditions.
A large study was conducted to attempt to remedy the shortcomings of clinical trials, and also to find ways of improving outcomes even after early setbacks. The study, Sequenced Alternatives to Relieve Depression (STAR*D), called for four treatment phases or levels, each lasting up to 14 weeks. After each phase, there were critical decision points, at which patients were carefully evaluated and could choose to switch to another antidepressant, add another antidepressant to the original agent, or add to or switch to cognitive therapy. All these augmentations of the original therapeutic regimen were based on careful assessments. The STAR*D model resulted in improved remission rates at every treatment level, although patients were more likely to attain remission during the first two levels of treatment than during the last two. The cumulative remission rates for the four levels were:
- Level 1 33%
- Level 2 57%
- Level 3 63%
- Level 4 67%
We should note that of the 2,876 subjects in the trial, most had comorbid psychiatric disorders, about 75% had recurrent depression, and about 18% had a history of attempted suicide. These were not easy patients to treat. The trial was funded by NIH and conducted at 23 psychiatric and 18 primary care clinics, and there were no differences in remission or response rates between psychiatric and primary care settings. [Trivedi MH et al, Neuropsychopharmacology 2007;Gaynes BN et al Clev Clin J Med 2008]
The first-line antidepressant used in STAR*D was citalopram, or Celexa, which is no longer one of the more widely-used antidepressants, having been largely supplanted by its enantiomer, escitalopram, or Lexapro. However, a considerable number of other agents were employed in the trial. My discussion of STAR*D specifically does not focus on the drugs used, but on the overall strategy, to enroll patients in the trial that were really representative of the population of persons with MDD in this country, and to keep trying difference therapeutic combinations in the effort, finally, to attain remission. The evidence that the strategy worked is that at the culmination of the four phases of the trial, two-thirds of the patients had attained remission, which is a far higher proportion that seen in trials on individual drugs. And it’s worth emphasizing in particular that the patient population in this study was far more challenging to treat than the subjects in most clinical trials.
A few conclusions, advanced with modest confidence by Doc Gumshoe
It seems clear that the underlying brain physiology is the same, regardless of the etiology of the depressive episode. The same neurotransmitters are at work regardless of whether you’re unaware of any eternal cause for your depressed mood, or whether you lost your job and starvation is imminent. In either case the same antidepressant will likely do something to lift your mood, but again, in either case, some understanding and help is indicated. It’s evident that neither mode of therapy is optimal on its own: the most effective treatment combines pharmacotherapy with psychotherapy / cognitive behavioral therapy. What’s not quite so clear is what this therapy should consist of, and there is wide variation, which depends of many factors – personal, situational, and economic, among others.
One more conclusion: with depression, as with many other diseases and disorders, if the first remedy isn’t effective, modifying the treatment plan leads to improvement in outcomes. The STAR*D study showed that, even though after the first attempt only one-third of the patients attained remission, that elusive goal could ultimately be reached in two-thirds of those real-world patients. A strategy of carefully assessing progress and modifying the treatment based on those assessments resulted in doubling the proportion of patients that reached remission. All of which tells us that, yes indeed, if we keep trying, there’s a way to shoo those blues away!
* * * * * * *
You will have noticed that I scarcely mentioned any specific drugs or described their mechanisms of action. That will have to wait for another post, coming up soon – this one is quite long enough! Best to all, Michael Jorrin (aka Doc Gumshoe)