Time Line of Depression Being Overcome To

Depression Becoming Chronic and "needing" Drugs

Major Research Support by Robert Whitaker, best selling author & science writer.  Please visit Robert Whitaker's List of Books and Support This Incredible In Depth Research.

In the 1940s, less than one per 1,000.  Back then, depression is basically a disorder for people 35 and older.  The Epidemiology of Depression, by Charlotte Silverman. Published in 1968. Silverman was chief of epidemiology studies at the National Institute of Mental Health.

 Pollack, H. Recurrence of attacks in manic depressive psychoses. America Journal of psychiatry, 11:567, 1931.  He is with the NYS Department of Mental Hygiene, and he looks at 8,000 manic-Depressives, admitted to a hospital between 1909 and 1920. He separates out those who had been admitted for depression, and found that more than 50% had a single episode, another 30% had two or three episodes, and only 17% became chronically ill.

Rennie, T. Prognosis in manic-depressive psychoses, American Journal of Psychiatry 98:801. (1942).   He reports a 95% recovery rate in those with depression alone, and a median duration of six months.

Lundquist, G. Prognosis and course in manic-depressive psychoses, Acta Pscychiatrica Neurol. (Suppl 35), 1945.  In a study of 216 depressives, found half were well within six months, and 75% within 10 months. He found that of those who got well, 60% to 70% never had another episode over the next  20 years. This was a study of first-admissions. (Swedish study?)

Conclusions by NIMH officials and other Leading Figures:

 George Winokur, a professor of psychiatry at Washington University, writes in his 1969 book,  Manic-Depressive Illness:  “In general, assurances can be given to a patient and to his family that subsequent episodes of illness after a first mania or even a first depression will not tend toward a more chronic course.”

Dean Schuyler, who was coordinator of the depressions section at the NIMH in 1974, wrote in this book The Depressive Spectrum:  “Most depressive states are self-limiting, that is they will run their course and terminate with virtually complete recovery without specific intervention.”

Paul Wender and Donald Klein, who were leaders in Biological Psychiatry, said in their 1981 book, Mind, Mood and Medicine:   “Vital depressive reactions tend to be self-limited, that is they tend to disappear in time.”

 ENTER THE Case for Antidepressants

Now with this understanding of depression, that it generally cleared up on its own, and once it did, the majority of people would not have a second episode, the thought was that antidepressants could simply speed up the process. They could get the symptoms to remit faster.

First NIMH Trials

1.  Medical Research Council, Clinical trial of the treatment of depressive illness. British Medical Journal, 1965; 881-886.  This is Britain’s equivalent of the NIMH. It studies an MAOI and a tricyclic. It found that the tricyclic was modestly superior to placebo, while the MAOI treatment “has been singularly unsuccessful.” It did not beat placebo (A sugar pill).

2. Smith, A. Studies on the effectiveness of antidepressants drugs. Psychopharmacology Bulleting, 1969, 5:1-53.  In 1969, the NIMH conducted a review of studies of antidepressant drugs. It concluded the poorer the design, the greater the benefit for the drug. The better the trial design, the lower the rate of improvement for drug-treated patients and the higher the rate of improvement for placebo patients. In well-controlled studies, 61% of drug-treated patients improved (usually in a six week period), versus 46% of nondrug treated patients. Thus, a 15% edge, which the NIMH concluded: “The differences between the effectiveness of antidepressant drugs and placebo are not impressive.”

This led Charlotte Silverman, at the NIMH to confess in 1968:  “Psychotropic drugs have not yet proved to be very helpful in depression.”

3. 1970. Raskin, Differential Response to chlorpromazine, imipramine, and placebo. A study of subgroups of hospitalized depressed patients. Archives of General psychiatry. August 23 (2):164-173. 1970. This is called the Second Collaborative Depression Study, funded by the NIMH.

 By end of study, differences favoring imipramine over placebo were not apparent. However, by excluding black patients who had done poorly, they were able to report that there was a small benefit for the drug.

Antidepressant versus Active Placebo

4. R. Thomson, Side Effects and Placebo amplification. British Journal of psychiatry, 1982; 140:64-68. This poor difference led some researchers to think the drugs were not really antidepressants at all, but rather “placebo amplifiers.” The thought was that the drugs provoked physical sensations—sedation, stimulant—that reinforced the patients’ belief that they were getting a magic bullet for the disorder, and that belief was what made the antidepressant slightly better than the placebo.

So a New England psychologist, Richard Thompson, decided to look at all the trials he could find that had used an active placebo. This is a biological agent not meant to be an antidepressant, but may cause something like dry mouth.  He found seven studies, and he found the drug was superior to placebo in only one of the seven. So no better than an active placebo.

The Second NIMH Trial.

 5. The NIMH now runs a second trial, which is called the NIMH Collaborative Research Program. Elkin, General Effectiveness of treatments. Archives of General Psychiatry, 1990;47:682-688. This had four arms. Interpersonal therapy, cognitive behavior therapy, a tricyclic, and placebo. Now after 16 weeks:  “There were no significant differences among treatments, include placebo, for the less severely depressed and functionally impaired patients.” Only the severely depressed patients fared better on imipramine than on placebo.”

The SSRI Story

6. The SSRI Drug Trials. Kirsch, “Initial Severity and Antidepressant Benefits: A meta-analysis of data submitted to the food and drug administration. PLOS Medicine, February 2008, Volume 5, issue 2, 260-268.  This group of researchers, using freedom of information requests, obtained data on ALL clinical trials submitted to the U.S. FDA for the six most widely prescribed SSRIs from 1987 to 1999. There were 47 trials in all they analyzed. Only 20 of 47 showed any measurable advantage for the drug; in the other 27, either they were the same or better for placebo. When they grouped all patients together, they found on average patients on drugs improved 10 points on the Hamiltom scale, and placebo patients slightly more than 8 points. The difference was 1.8 points, and the British health agency had recently decreed that anything less than 3 points was clinically irrelevant. They concluded:

“Recent metanalysis shows ssris have no clinically meaningful advantage over placebo.”

The only group they found any benefit for was the “most severely depressed patienits.”  “There is little reason to prescribe new-generation antidepressant medications to any but the most severely depressed patients.”

B. The Case For Maintenance Therapy

Now, given that the initial conception of depression was that people got better anyway, and that drugs were simply to speed up the recovery process, there wasn’t the idea that you should stay on the drugs. So really, no case for maintenance therapy arose for years.

The Case Against Antidepressants

Long-term Outcomes: the chronicity data:  Now since nearly everyone was thought to get better in depression, little thought at first about improving long-term outcomes. Most people recovered and returned to complete pre-morbid functioning.

1968. DiMascio. Effects of imipramine on individuals varying in level of depression. AJP 1968, 127 (sup):55-58. The first hint that exposure to antidepressants might make patients more chronically ill came in 1968, with a research report that first use of a tricyclic in patients who had only slightly ill subsequently had an increase in depression levels.

1973, Van Scheyen, Recurrent vial depressions. Pscyiatric Neurol Neurochir, Mar-Apr, 76 (2):93-112.  Next, a researcher did a naturalistic follow up study of patients treated either with an antidepressant or no drug. He found that treatment with a tricylic antidepressants was associated with an increased number of recurrences. In other words, the depression was more likely to come back. He wondered:

 “Whether such an increased number of depressive phases would not be regarded as a side effect or a paradoxical effect which, after protracted therapy, is produced by the tricyclic antidepressants so far most commonly used.”

 1981. Kovacs, Depressed outpatients treated with cognitive therapy or pharmacotherapy. A one-year followup. Jan, 38 (1):33-39.  No placebo group, but there is a finding that after one year, those treated with drugs were more symptomatic and twice as many had relapsed than those treated with cognitive behavior therapy.

1986. Blackburn, A Two-year naturalistic followup of depressed patients treated with cognitive therapy, pharmacotherapy, or both. Journal of affective disorders 10 (1986), 67-75.  This is a UK study of 40 or so patients. At end of two years, we had these results:

44% of drug-treated group were depressed

8% of cognitive therapy group were depressed

 In the drug group, 78% had suffered at least one relapse in the two-year followup, versus 23% of cognitive therapy group. So relapse is three times as high.

BIG NIMH trial. Shea, Course of depressive symptoms over follow-up; findings from the National Institute of  Mental Health Treatment of Depression collaborative research program. Archives of General Psychiatry. 1992;49:782-787.  This is the follow-up to the study to the one that found no short-term benefits except for the severely ill. The follow up findings were even more worrying:

                                                 18-month relapse rates for           Recovered and well at

                                                those recovered at 8 weeks          18 months (all patients)

 Cognitive behavior                                36%                                                     30%

Interpersonal therapy                            33%                                                     26%

Placebo                                                  33%                                                     20%

Imipramine                                           50%                                                     19%

So we see here a substantial increase in relapse rates for the drug treated patients, and stay well rates are lowest for the drug group.  Key was also social functioning. When you looked at all patients in the aggregate, social functioning was found to be “superior in the interpersonal” group than to either CBT or drug. But it wasn’t superior to the placebo group. Researchers from the State University of New York revisited the data by looking at drop outs. They concluded that if you add this group into your data: 

 “Patients receiving a antidepressant were the most likely to see treatment following termination, highest probability of relapse and exhibited the fewest weeks of reduced or minimal symptoms during the followup period.”

1993, Frank, Efficacy of Treatments for Major Depression. Psychopharmacology Bulletin, vol. 29, 4, 457-475.  These are researchers at University of Pittsburgh. This is a review of the literature. They wrote: "The number of studies that have looked at the efficacy of drug therapy on the long-term wellness of the patient with major depressive disorder is limited. The data available are, in general, not very positive.”

 The problem was that this form of care just wasn’t producing good long-term outcomes. Only about 50% of patients initially responded to the drug, and then even among the good responders, long-term outcomes weren’t good. That’s what the NIMH trial had showed, and in this review, they found other studies that showed the same thing.

Two other studies reviewed by Frank:

             In one trial, only 48% of the good responders, when maintained on the drugs for two years, stayed well. In another, only 32% remained well.

 The Problem:  Even the good responders kept on the drugs were descending into second and third episodes, and it was even worse for drug-treated patients who were then withdrawn from the drugs, which was the recommended form of care. So staying on drugs ended up in recurrent illness, and going on drugs and then withdrawing seem to end up in recurrent illness as well.

Why did it used to be that people got well and many stayed well. Why had the course of the illness changed?

The Star*D trial.

 Rush, Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: A star*D report. Am. J. Psychiatry (2006): 163:11: 1905-1916. Finally, the NIMH in the late 1990s mounted a new trial to study the outcomes of drug-treated outpatients. Here they are:

             3,671 patients entered into the trial.

             36.8 percent saw their symptoms remit in first three months.

            Of those who remitted, 50% relapsed within 12 months.

            At end of year, only 18% who initally remitted and stayed well.

            If included study drop-outs, only about 11% of beginning group and stayed well.

They then kept trying those who didn’t remit on a new regimen, and in that way got many initial non-responders to remit. But each step ends up with fewer and fewer patients.

Editorial, Star*D: What Have We Learned. Am. J. Psychiatry, (2007): 164:2:201-203. In an editorial titled “What Have We learned,” the study investigators concluded:  

“This highly representative clinical sample of depressed outpatients has revealed that major depression is often chronic, severe and associated with substantial general and psychiatric comorbidity.”

HUH?  Now It Becomes Chronic! 

The Worsening the Course Question

 1994. Fava. Do Antidepressant and antianxiety drugs increase chronicity in affective disorders?” Psychotherapy and Psychosomatics 61 (1994):125-31.  He writes:

 “Within the field of psychopharmacology, practitioners have been cautious, if not fearful, of opening a debate on whether the treatment is more damaging than the cure . . . I wonder if the time has come for debating and initiating research into the likelihood that psychotropic drugs actually worsen, at least in some cases, the progression of the illness which they are supposed to treat.”

 “Do antidepressants drug sensitize (the brain) to depression.”  While they may provide some benefit over the short term, he worried that “long-term use of antidepressants may also increase the biochemical vulnerability to depression.”

American Response. This was reported in Psychiatric News, May 20, 1994. A trio of American doctors told him to knock it off. They said: “The industry is not interested, the NIMH is not interested, and the FDA is not interested. Nobody is interested.”  I have seen this personally by the attacks by local Mental Health America and the Psychiatrist at Health & Human Services.

 1995, FAVA, Holding On: Depression, sensitization by antidepressant drugs, and the prodigal experts.” Psychotherapy and psychosomatics, 63:137-141. He now steps forward with a biological explanation. “Drug treatment may recruit processes that oppose the initial acute effects of a drug or of receptor alterations. When drug treatment ends, these processes may operate unopposed, at least for a time.” This  is the reason why “abrupt drug removal is associated with a variety of potentially untoward responses.” So you see the problem with drug withdrawal, but at the same time, he noted, there is also a “loss of anti-depressant effect with long-term treatment.”

So if you stay on drugs, not so good, and if you go off, even worse.Now he raises this question: At some point, can the brain ever renormalize?

Best evidence was that it took at least “several months” may be needed “to reestablish a predrug level of neurophysiological and neuropsychological homeostasis.” He noted that with schizophrenia, you get tardive dyskinesia, which is a kind of irreversible brain damaged, and wonders if at some point with antidepressants you get “irreversible receptor modifications” that make you more vulnerable to depression.

1998 Viguera and Baldessarini. Discontinuing antidepressant treatment in major depression. Harvard Review of psychiatry, 1998:5:293-306. This paper looked at rates of relapse following discontinuation of drugs. And what it found basically proved Fava’s point, and that was that longer you are on drugs, the more the receptor modifications might become irreversible and thus make you prone to chronic illness. As Fava later noted, they had reported that:

            “Whether one treats a depressed patient for three months or three years, it doesn’t not matter when one stops the drugs. A statistical trend suggested that the longer the drug treatment, the higher the likelihood of relapse.”

Chronic Antidepressant Treatment Alters Serotonergic Regulation of GABA Transmission in Prefrontal Cortical Pyramidal Neurons (2004) Neuroscience 65-73.  "These electrophysiological results suggest that chronic antidepressant treatment resulted in a down-regulation of synaptic function of forebrain 5-HT receptors."    This means that once you go on the drug, you may experience worse symptoms down the road, and thus, need even more of their product to get the same reduction in behavior, thought or feeling.  The authors go on to discuss the interactive role of serotonin and GABA...claiming that GABA must have a role in mood disorders.  Either way, the ongoing confirmation of down-regulation (aka neurophysiological disruption) is my interest.

2003. Fava, Can Long-term treament with antidepressant drugs worsen the course of depression?” Journal of Clinical Psychiatry 64 (2003): 123-133. He ran through the literature.

a) The literature showed “very unfavorable long-term outcome of major depression treated by pharmacologic means.”

 b) the literature showed “tolerance to drugs with maintenance therapy”, and by that he meant the drugs lose their efficacy.

 c) The literature showed that once you relapse, you increasingly have trouble finding any drug that works

 d) The literature showed that  “withdrawal symptoms” that increased risk of relapse.

 As a result, he said, depression was being reconceptualized as “a chronic disease.”   He concludes: “At present, it is impossible to know whether antidepressant drugs fail to improve the long-term course of depression or worsen its course.”

 Confirming Evidence: What happens to people diagnosed with major depression and they are never exposed to antidepressants?

 Naturalistic Studies: Now how can we answer that question? What it would be good to see is how untreated patients fared over the long-term in naturalistic settings. And the point here is those not getting drugs are not patients who worry about getting placebo, but think they might get better on their own.

So can we go back and find patient pools where you had a lot of patients present with depression, and then some chose to go on drugs at first treatment and others did not, and then we can find which group did better over the long-term. That will give us some sense of how the drug-treated course at first admission compares to the natural outcomes course.

Ronalds, The outcome of anxiety and depressive disorders in general practice. Britisth Journal of Psychiatry 1997:17: 427-433.  This is a British study in which the data was extracted from a larger inner-city practice with 13,000 patients. They identified 148 who presented with a first episode of depression and anxiety. 95 with no drugs and 53 with drugs.  Measured on Hamilton score (i.e. depression rating). Now those who didn’t go on drugs, on average, weren’t quite as ill as those on drugs. However, after six months, those not on drugs doing much, much better than those who got treated with drugs. Even the amount of their improvement was much greater: 62% reduction in their Hamilton score, versus 33% for those on drugs.

1998. Goldberg, The effects of detection and treatment on the outcome of major depression in primary care, a naturalistic study in 15 cities. British Journal of General Practice, 48, 1840-1844.

 This is a study funded by the World Health Organization. They looked at outcomes in 540 patients, divided into four groups. Three where depression was recognized, and treated either with drugs, sedatives, or no drugs, and then undetected. They hypothesized that those treated with drugs would do best, sedative second best, then detected/no drugs, and undetected/no drugs. They were followed for a year.  Here are the results:                       Well                 Cont. Depression     Another Diag. Remained

                                                                                                                     Mentally ill

 Antidepressants            31.3%                          51.6%              10.9%              62.5%

Sedatives                      24.5%                          44.9%               14.3%             58.2%

No Drug                       39.5%                          25.2%              24.4                49.6%

Unrecognized               41.7%                          28.3%               20.0%                        48.3%

 Key is drug treated group worst, and no drugs the best. They conclude:

  “The study does not support the view that failure to recognize depression has serious adverse consequences.”

2000. Weel-Baumgarten, “Treatment of depression related to recurrence:10-year followup in general practice.” Journal of Clinical Pharmacology and Therapeutics 25, 61-66. This was a study by Netherlands researchers in which they went back and looked at patients diagnosed with major depression before 1985, and thus when many people weren’t put on drugs. They found 222 patients, with followup records going for 10 years.  Here are the results:

                                Number             One episode           Two Episodes            More than two

 Initial Drug                    134                  67 (50%)            26 (19%)                    41 (31%)

Initial No Drug                88                  67 (76%)              9 (10%)                    12 (14%)

If you look at the no initial drug treatment group, you see that 76% had one episode and that was it. This is the old course of the disorder. Another 9% had only two episodes. Only 14% had three or more episodes and thus began to end up in the chronic category. Now here drug treatment was much less than in U.S. and overall pretty favorable outcomes. But 31% of those initially treated with antidpressants ended up in the chronic category—the risk was more than doubled.

 They concluded gently: “A diagnosis (of depression) does not necessarily mean a need for treatment. Even when a diagnosis of major depressive disorder has been made, spontaneous recovery should be considered for a number of cases in general practice and watchful waiting could prove worthwhile.”

2003. Dewa. Pattern of antidepressant use and duration of depression-related absence from work. British Journal of Psychiatry (2003): 183, 507-513. Canadian study. Premise: They said it was curious that with the increase in treatments for depression, it was strange that in Canada there was a rise in disability for depression. Hypothesis:  

They speculated that “undertreatment” was a cause, and therefore conducted a study to see if there was an association between antidepressant use and return to work for those who went on short-term disability for depression. Their hypothesis was that those who got the drugs would be more likely to return to work quickly, and also that fewer would move on to long-term disability.

So they do an observational study using administrative date from three major Canadian financial and insurance sector companies. They have a combined workforce of 63,000 employees. The study period is from 1996 to 1998. After 10 days of missing work in a row for an illness, can go on short-term disability. If not better in six months, go on long-term disability.

 The results for the 1,281 people who went on short-term disability:

                         Number       Returned           Long-term            Quit/retired

                                              to Work            Disability           Fired

 Non-drug         565              471 (83.4%)        52 (9.2%)         41 (7.3%)

Drug                 717              524  (73%)         136 (19.0%)      56 (7.8%)

 In terms of time they spent on short-term disability, it was 77.3 days on average for he non-drug group, and 104.7% for the drug group.They wrote:  “Those who did not use antidepressants returned to work sooner than those who did . . . Does the lack of antidepressant use reflect a resistance to adopting a sick role and consequently a more rapid return to work?”

2004. Patten. The Impact of antidepressant treatment on population Health. Population Health Metrics, 2004, 2:9.  This is an epidemiological study drawn from two longitudinal health studies done by the government of Canada. One is called the National Population Health Survey and the other is the Canadian Community Health survey. These two studies collected data from 1994 to 2000. The studies drew on health records of 130,000 people including 9508 that were diagnosed with depression.

 The results: Symptoms: Those off drugs spent on average (the median) 10.8 weeks depressed each year. Those on drugs spent on average 18.7 weeks depressed each year.

 Relapse rates in remitters:  People who were symptom free for a year, who then developed major depression in the following year, it was 11.1 percent relapse rates for those who were on drugs, and 3.5% for those off drugs.

Bottom line:  Nearly twice as symptomatic, and for those who got well for a year, a three times higher relapse rate.

Conclusion:  They note that their findings are consistent with Fava’s hypothesis that “antidepressant treatment may lead to a deterioration in the long-term course of mood disorders. This hypothesis, although not widely accepted, predicts that episode duration and incidence would be higher in those reporting antidepressant use than in those not using these medications.”  “That future research is needed to address the lack of epidemiological evidence confirming the population-health benefits of increased antidepressant treatment.”

2006, Georgetown University Medical Center  Jan. 20, URL:
http://www.sciencedaily.com/releases/2006/01/060119230939.htm  Commonly Used Antidepressants May Also Affect Human  Immune System.  SSRIs manipulating the neurotransmitter serotonin, but now researchers know it also affects our immune system in ways "that are not yet understood."  investigators found, for the first time, that  serotonin is passed between key cells in the immune system, and that the  chemical is specifically used to activate an immune response. "They do not  know...if it could have a damaging effect...bolster immunity to the point that they trigger autoimmune  disease...at this
point we just don't know how these drugs might affect  immunity."

National Disability Data

 We noticed in the Canadian study on returning to work that the researchers had spoken about the increased number of people disabled in Canada due to depression. So now let’s see if we see it in other countries as well.

2000. Moncrieff, Trends in sickness benefits in Great Britain and the contribution of mental disorders. Journal of Public Health medicine, 22, 1, 59-67. From 1984 to 1995, there was a 244% increase in disability due to mental disorders. And the big one is depression and neurotic conditions. They went up from a total of 40 million days of incapacity to 120 million days of incapacity. Three fold increase as Prozac came in.

2004. Helgason. Antidepressants and pubic health in Iceland. Britisth Journal of Psychiatry, 184, 157-162.  They raise the question: Has (extensive use of antidepressants) had any impact on public health and the burden of depression?  Study. Iceland has population of 286,000. They looked at number of prescriptions of SSRIs, outpatient visits, and admissions to psychiatric departments for period of 1989-2000. Antidepressant use in Iceland begins with prozac in 1988.  Results.

             The number of patients admitted to psychiatric facilities increased 3.9% annually.

The total number of admissions (some patients coming back more frequently) increased 5.4% per year.

The number of out-patient consultations increased 2% per year.

The percentage of the population disabled by depressive disorders rose from .4% to .7%. (1/250 to 1/143.) They concluded that if the drugs were useful, one would expect they would have “reduced disability, morbidity and mortality.” Instead, they had found increased disability and morbidity.”

2005. Rosenheck. The Growth of psychopharmacology in the 1990s. Evidence-based practice or irrational exuberance. International Journal of Law and psychiatry, 28: 467483.  Rosenheck is at Yale, and does a lot of research with the VA. He reports on two increases in disability during the 1990s, which is a period that use of antidepressants soared. 

a) SSA Disability benefits for affective disorders:

                  1994:  425,138

                 2002:   939,711                   More than doubled.

                 As percentage of adult population, from one in 357 to 1/227.

 b) National Health Interview Survey

 The proportion of working age Americans who reported themselves to be disabled due to depression increased from .10% to .30% during the 1990s. The percentage who report themselves unable to work because of depression rose from .7%  to 2.1 percent.  

Depression Transformed

Prevalence

 NIMH Fact sheet: “Major depressive disorder is the leading cause of disability in the U.S. and established market economies worldwide.” Now hits 5% of the U.S. adult population in any year.

Now the Course: In 1999, the APA Textbook of Psychiatry, Essentials of Clinical Psychiatry, said:  It used to be thought that “most patients would eventually recover from a major depressive episode. However, more extensive studies have disproved this assumption.”

Today, it says, after a first episode:  

Only 50% recover within six months

Only 15% of unipolar patients will only have one episode; 85% will have at least a second episode.

 Between 15% and 54% of patients now have three or more episodes, and with each new episode remissions become “less complete and new recurrences develop with less provocation.”

 Depression, the APA NOW concluded, “is a highly recurrent and pernicious disorder.”

Summary:

1. In the pre-drug era, depression was a disorder that primarily affected those over 35.

2. Fewer than one in 1000 people were hospitalized with depression.

3. Prognosis was good. More than 50% would recover within six months, and 75% within 10 months. The majority who recovered would not have another episode in the next 20 years.

4. When antidepressants were introduced, idea was they could just speed up time to recovery. Fifty years of short-term studies shows that the drugs may be slightly more efficacious on knocking down target symptom over short term, but really only in severely depressed patients. No evidence better than active placebo.

5. Starting in 1968, researchers started worrying that might be making people more depressed over long term. Patients placed on the drugs then relapsed at high rates when they stopped taking them, and even if they continued taking them, had fairly high rates of relapse. A high percentage of people treated with antidepressants were becoming more chronically ill.

6. In the 1990s, Giovanna Fava raised the hypothesis that the drugs were making people more biologically vulnerable to depression over time. He pointed to the drug-induced abnormalities in receptor densities as a possible reason.

7. We then looked at five naturalistic studies from three countries, and a WHO study from 15 countries. All showed that patients treated without drugs did better over the long term.

8. We then looked at disability numbers due to depression from several countries. All showed rising disability rates in the 1990s, which was when prescribing of antidepressants really soared.

9. Finally, we saw that the disorder was transformed. Before drugs, was seen as one with a good prognosis, and often only one episode. Now see as highly chronic, and the prevalence is many times higher than it was in the pre-drug era.