History of Bipolar
II. Manic Depressive Illness Prior to 1955.
1. David Healy in his book Mania, published this year by Johns Hopkins press.
He went through the records of hospitals in North West Wales, and he found that from 1875 to 1924, there were 127 patients admitted to the psych hospital with symptoms that today would be seen as bipolar. That is one case per 100,000 population per year—an extremely rare disorder. Average of onset was 32; youngest admission was 17.
2. The epidemiology of Depression, Charlotte Silverman, 1968.
She cites various first admission studies from the 1940s, 1950s and 1960s, and they show rates of one in 10,000 to one 50,000.
As for prevalence—people who have it at any one time—Silverman cites studies from these three decades showing it to be from around one in 3,000 to one in 10,000. So extremely rare.
1. David Healy on manic-depressive illness, as seen in records of a Welsh hospital:
“Manic-depressive patients got well and went home and on subsequent admissions to hospital were often described as being in exactly the same state that they had been in during the course of their previous admission.” The 127 patients he identified nearly all went home well, usually in six months. In total, this group had 345 admissions, or about three each.
2. Winokur, in his book Manic Depressive Illness. 1969.
He’s a professor at Washington University:
In Lundquist’s 1945 study, he identifies 95 manic patients. 75% well by 10 months; 42% became ill once more in course of 20 years, I believe. 85% socially recovered. Only 8.5% developed a chronic course.
Pollack studies 8,000 manic depressive cases from 1909 to 1920, from NYS Department of mental hygiene. In this period of time, over half had not had a second attack. Only 20% suffered 3 attacks or more.
“The outcome of manic depressive disease is characterized by not becoming chronic . . . patients may have one or more attacks, with symptom-free intervals. When the patients recover, there is no difficulty resuming their usual occupations.”
3. Goodwin and Jamison, Manic Depressive illness, 1990.
Goodwin cites five studies from the pre-drug era that followed patients from 10 to 32 years. He shows chronic rates ranging from one to 11 percent; the fifth, which was a very small study of 39 patients, said 56% become chronic. A sixth 35-year followup study of 100 patients that was published in 1979, and thus spills over a bit into the drug era, said 22% became chronic, 14% had fair outcomes, and 64% had good outcomes.
4. Huxley, Disability and its treatment in bipolar disorder patients. Bipolar disorders 2007:9:183-196.
Up until early 1970s, 85% employment rate for those with bipolar. Also notes absence of cognitive impairment prior to 1970s in various tests.
So what we see is episodic illness, almost like a fever that strikes now and then. Get ill, recover fully. No cognitive impairment. And then some get ill again. Three episodes on average. But for most, return to complete premorbid functioning between episodes.
III. The Emergence of Bipolar Illness As Separate from Depression
The emergence of manic-depressive illness as a separate disorder from depression occurs in 1970s. Then we get what is called bipolar 1, which is when you have had an episode of both mania and depression that have required hospitalization, and then bipolar II, an episode of lesser mania and of lesser depression that didn’t require hospitalization. In DSM III, in 1980s, you formally get bipolar disorder carved out as being distinct from unipolar disorder.
IV. The Case for Lithium
The evidence base for use of drugs in bipolar basically arises out of use of lithium to knock down mania, and then as a use for reducing risk of recurrent episodes.
Placebo-controlled trials for acute mania
1985, Tyrer. Lithium in the Treatment of Mania, Journal of Affective Disorders, 8 (1985), 251-257.
Now in terms of placebo controlled trials of lithium for knocking down acute episodes of mania, there are only few. A 1985 review of the literature could only find four, and concluded that all suffered from methodological flaws. But the overall response rate to lithium was said to be 76%, and that was higher than placebo. However, because lithium can take 6 to 10 days to produce much an effect, the common practice was to use antipsychotics to quiet the patient, and then to give lithium.
Lithium as a maintenance treatment
1994. Baker, J.P. Outcomes of Lithium Discontinuation, Lithium, 5, 187-192.
There was evidence that many people with manic-depressive illness would suffer a second episode at some time, and thus it would be good to have a drug that could prevent this. Lithium became the drug of choice for a time. The studies that proved this had this design: You took patients who were stabilized on the drug and then divided into two groups: One would be withdrawn, usually abruptly, and the other maintained on the drugs.
Metanalysis of 19 such studies. Results:
53.5% of those withdrawn from lithium relapsed
37.5% of those on lithium relapsed.
Gradual withdrawal. However, within those 19 studies, the researcher found several where he could group patients according to whether they had undergone slow withdrawal (greater than a two-week period.) Small patient sample, but only 23 of 78 patients relapsed, or 29%.
The importance of this is that the slow withdrawal method had a lower relapse rate than those maintained on lithium. So is lithium preventing relapse, or are what you seeing is that once you go on lithium, you are at high risk of relapse if you come off it quickly?
V. The Changing Course of Bipolar.
A. Conversion of Unipolars into Bipolars.
One of the problems that was quickly realized with the use of antidepressants is that it triggered manic episodes in many depressed patients, and thus converted them into bipolar patients.
1. 1985. Angst, J. “Switch from depression to mania, a record study over decades between 1920 and 1982. Psychopathology 18 (1985): 140-154.
These Swiss investigators found that from 1940 to 1957, only six percent of depressed patients admitted to a psychiatric hospital in Zurich, were of the bipolar type. However, following the introduction of first-generation antidepressants, this percentage rose to 12% during the years 1959 to 1963, and then to 26% during the years 1969 to 1973. Bipolar illness was suddenly four times more common than before. Moreover, they determined that whereas prior to 1959 only 2% to 7% of hospitalized unipolar patients “switched” to a bipolar state during the course of their treatment, this rate jumped to 9-25% after that date. So threefold rate.
2. 2001, Goldberg and Harrow. Risk for bipolar illness initially hospitalized for unipolar depression. American Journal of psychiatry, 158, 1265-1270.
Now this question of switching has remained something of a controversial subject. There is debate as to how much this increased risk is. But in this 15-year study, 27% converted to bipolar. Other studies have shown that as many as 40% of long-term antidepressant users do. But this is one source for the new flood of bipolar patients.
B. Worsening of symptoms
The next thing that was quickly noticed was that the course of bipolar illness changed dramatically. Researchers noted that drug-treated patients were cycling much more rapidly, they were more constantly symptomatic, and they suffered more relapses.
More rapid cycling
1. 1980. Kukopolos. Course of the manic-depressive cycle and changes caused by treatment. Pharmakopsychiatr Neuropsychoapharmakol. 13(4). 156-7.
Antidepressants associated with rapid cycling.
2. Kukopolos. Rapid Cyclers, Temperament and antidepressants. Comprehensive Psychiatry, 1983. Vol. 24, No. 3,m 1983.
“The general impression of clinicians today is that the course of recurrences of manic-depressive illness has substantially changed in the last 20 years. The recurrences of many patients have become more frequent. One sees more manias and hypomanias and therefore more bipolar cases than before, more rapid cyclers, and more chronic depressions.”
They noted that 16% of their manic-depressive patients—and this was by Kraepelin’s old definition of manic-depressive illness—were “rapid cyclers,” who now suffered 6.5 episodes per year. This group of rapid cyclers also became poor responders to lithium.
“In the majority of cases, the change of the previous course into rapid cycling coincided with antidepressant drug treatments.”
So here, early on, is the recognition that drugs may be worsening the course of the disorder:
“The possibility that in some manic-depressive patients antidepressants can worse the further course of he disease raises the question of how they should be treated . . . concern about the future course of the disease should prevail over the need of immediate clinical success.”
“It certainly seems paradoxical that a treatment that is therapeutic for depression can worse the further course of the disease.”
3. 2003. Koukopoulos. Duration and stability of the rapid-cylcing course: a long-9term personal followu- of 109 patients. Journals of affective disorders, 73 (2003):75-85.
They found that antidepressants caused 88% of rapid-cycling in patients. Associated with a much worse long-term course.
4. Mallakh, Antidepressant-associated chronic irritable dysphoria (ACID) in STEP-BD patients. Journal of Affective Disorders. Article in Press, 2008.
Patients treated with an antidepressant were ten times more likely to develop ACID than those who were not. However, it appears to be related to antidepressant-related manic switch.
1. 1981. Symonds, Lithium and the changing incidence of mania. Psychological medicine, 11, 193-196.
This was a study of bipolar patients in England. The researchers reported that readmission rates increased during the 1970s, when antidepressants and lithium became the standard of care.
2. 1990. Harrow. Outcome in manic disorders. Archives of General Psychiatry 47, 665-671.
They reported a 1.7 year followup of 73 bipolar patients. (This is from the same Chicago study we found in schizophrenia). More than 40% treated with lithium suffered a relapse. They concluded: “Manic patients taking lithium carbonate did not show better outcome than those not taking lithium carbonate.”
3. Tohen, Outcome in mania, Archives of General Psychiatry, 1990, 47(12):1106-11.
This was a four-year follow-up of 75 patients. They reported that only 28% of patients treated with drugs were in remission at end of four years. This is mostly a lithium treated group. So a lot of chronicity.
4. 1995. Gitlin, Am. J. psychiatry. Am. Journal of Psychiatry. Nov. 152 (11):1635-40.
This study, by guys at UCLA, followed 82 patients for a mean of 4.3 years. Despite continual maintenance treatment with lithium (and other drugs), 73% relapsed within five years into mania or depression. Of those who relapsed, two-third had multiple relapses. So now we see beginning of descent into multiple relapses, chronicity, etc.
And even for those who did not relapse, “considerable affective morbidity was observed.” Conclusion:
“Even aggressive pharmacological maintenance treatment does not prevent relatively poor outcome in a significant number of bipolar patients.”
More Symptomatic In Between Acute Episodes
Instead of return to normal functioning between episodes, have low-level symptoms.
5.2002. Judd. The long-term natural history of the weekly symptomatic status of bipolar disorder. Archives of General psychiatry 59 (6):530-7.
This followed up 146 patients first entered the NIMH Collaborative Depression Study from 1978 through 1981. Then followed up for a mean of 12.8 years, weekly.
Patients with bipolar one were symptomatically ill 47.3 of weeks. They were depressed 32% of the time, manic 9% of the time, and going through mixed symptoms 6% of the time. Patients with BP-1 changed symptoms status on average of six times per year and polarity more than 3 times a year. So rapid cycling, etc.
6. 2003. Post, Morbidity in 258 bipolar outpatients followed for 1 year with daily prospective ratings on the NIMH life chart method. J. Clinical Psychiatry, 2003, June 64 (6):680-690.
This is with the Stanley Foundation Bipolar Network and Biological Psychiatric Branch, NIMH. This is a daily followup for one year of patients admitted between1996 to 1999. They were on a mean of 4.1 psychotropic medications. “Despite comprehensive pharmacologic treatment” they were depressed 33.2 percent of the time, manic 10.8% of the time, and 62.8 percent had four or more mood episodes per year.
This is a new course. Highly symptomatic, cycling a lot. No longer getting that return to asymptomatic morbidity.
7. 2003. Post. A prospective investigator of the natural history of the long-term weekly symptomatic status of bipolar II Disorder. Archives of General Psychiatry 60,
They were symptomatic 54% of the time. Mostly depressed (50.3% of the weeks.) Seems like they went off and on drugs. They conclude: BP-II “Is a chronic disorder.”
Now chronically and permanently ill
2007. Elinson. Working, receiving disability benefits, and access to mental health care in individuals with bipolar disorder. Bipolar disorders 2007, 9:158-165.
This looked at 1,855 individuals with bipolar in a registry maintained by the Western Psychiatric Institute at the University of Pittsburgh. These were people who on average had been diagnosed 20 years earlier. They found the following:
Two-thirds had been depressed in last six months
Close to one-third had been manic in last six months
More than one-quarter have had mixed or rapid cycling in last six months.
So still, although on drug cocktails, were very symptomatic even many years after iniial diagnosis.
C. Decline In Functioning
As was noted before, in pre-drug era, after patients recovered from an acute episode, they returned to regular “premorbid” functioning. Employment rates of 85% or so for bipolar patients. The fever remitted, so to speak. That has changed dramatically for drug-treated bipolar patients.
1. 1988. Dion, Hospital Community Psychiatry, 39 (6), 652-7.
At six-month followup, 80 percent of patients were asymptomatic. But functional outcome was terrible. Only 43 percent of patients were employed, and 21 percent were working at their expected level of employment. “The results suggest that factors other than symptoms are related to the functioning of patients with bipolar disorder, and that treatment should be targeted to the patient’s disability as well to symptom amelioration.”
2. 1998, Keck. 12-month outcome of patients with bipolar disorder following hospitalization for a manic or mixed episode. Am. J. Psychiatry, 155 (5) 646-52.
At end of 12 months, 48% of group recovered “syndromally,” 26% recovered symptomatically, and functional recovery in only 24%.
3. Tohen, Two-year syndromal and functional recovery in 219 cases of first-episode major affective disorder with psychotic features. Am. J. Psychiatry, 2000, Feb. 157 (2):220-8.
Only 36.9% recovered functionally.
4. 2002. Kupfer, Demographic and clinical characteristics of individuals in a bipolar disorder case registry. Journal of Clinical Psychiatry. 2002, February 63 (2):120-5.
This is an analysis of 2839 patients in a the Stanley Center Bipolar Disorder Registry. Despite the fact that 60% completed some college and 30% were college graduates, 64% were unemployed. Patients were taking multiple medications, and more than one-third were on 3 or more drugs. “Our present findings point to the chronicity and severity of bipolar disorder as experienced in the community.”
5. 2003. Tohen, The McLean-Harvard First-Episode Mania Study: prediction of recovery and first recurrence. Am. J. Psychiatry 2003, 160: 2099-21077.
This is an update of their 2000 report. There are now 166 patients in the study. Only about 37% achieve functional recovery at end of two years. 40% had relapsed within two years.
“Antidpressant treatment was marginally related to longer time to recovery and earlier relapse.”
6. 2007. Pope. Determinants of social functioning in bipolar disorder. Bipolar Disorders 2007, 9:38-44.
They note that Coryell in 1993 found that psychosocial impairments “were pervasive and persistent even in individuals who experience sustained remission of clinical symptoms.” A 2001 review confirmed a 30% to 60% prevalence of social and occupational adjustment problems in individuals with BP whether or not they had inter-episode symptoms.
IN this study, only one-third employed or a parent. Depressive symptoms is what makes people so impaired.
D. Cognitive Impairment
1. Zarate, Functional Impairment and cognition in bipolar disorder. Psychiatric Quarterly, 71, no. 4, winter 2000. 309-329.
Earlier, cognitive impairment has been held to be uncommon in bipolar disorder. Earlier studies conducted prior to 1975 found no consistent findings in cognitive deficits in bipolar patients.
However, they note: Psychopharmacological treatment has been suggested to influence neuropsychologoical test performance with neurooleptics and anticholinergic drugs having the greatest effect. (Bellini, 1988.)”
So they see drug-caused cognitive impairment as a possible reason for the poor function. They write:
“Cognitive deficits in schizophrenic patients ;have been shown to be highly associated with poor outcome even more than positive and negative symptoms. Recently there has been some suggestion that there may be impairment of cognition in patients with bipolar disorder. Whether this is the case and whether this deficit in cognition is associated with the impairment in functioning seen in bipolar patients achieving syndromal recovery is an intriguing question.
Cognitive Deficits/Now Like Schizophrenia
1. 2001 Dickerson, “Outpatients with schizophrenia and bipolar I disorder: do they differ in their cognitive and social functioning?” Psychiatry Research 102 (2001):21-27.
The 26 bipolar patients in this study are mostly on drug cocktails—antipsychotics, anti-convulsant mood stabilizers, antidepressants, anticholinergic agents, and half on lithium. They were taking a mean of 3.2 medications.
The 74 schizophrenia patients are also mostly on drug cocktails. Antipsychotics, antidepressants, lithium, mood-stabilizers, etc. They were taking a mean of 2.4 medications, but higher doses of antipsychotics.
Results: They were compared on 41 cognitive and social functioning variables. The two groups did not differ significantly on 36 of the variables.
In terms of cognitive function, they found “a similar pattern of cognitive functioning in patients with bipolar disorder as compared to those with schizophrenica.”
And “on most measures of social functioning, our patients with bipolar disorder were not significantly different from those in the schizophrenia group.”
2. 2004. Dixon. Effect of symptoms on executive function in bipolar illness. Psychological Medicine, 34 (5):811-21.
“Deficits in response initiation, strategic thinking and inhibitory control were evident in all the bipolar groups.”
3. 2005. Martinezn-Aran. Psychotherapeutics and psychosomatics, 74 (5):295-302.
“Bipolar patients showed deficits in “several cognitive measures related to attention, memory and executive function.”
4. 2007. Martinzez-Aran. “Functional outcome in bipolar disorder: the role of clinical and cognitive factors.” Bipolar disorders 2007: 9 103-113.
In this study, looked at cognitive impairment. 75 bipolar patients and 35 healthy controls. The bipolar patients were euthymic when tested.
These patients showed cognitive impairment on verbal memory and executive function measures. But high functioning bipolar patients also performed poorer (than controls) on new learning, as well as on recall, attentional measures, and semantic verbal fluency.
“In general, patients complain about having difficulty in remembering names and conversations both in the distant past and more recently.”
Verbal memory is now as bad in bipolar patients as schizophrenia patients. Executive function in bipolar patients halfway between schizophrenia and normals.
In this study, with respect to executive and verbal memory functioning, this implicates dysfunction in the prefrontal cortex (and other structures.)
Because of this cognitive impairment, “Function outcome is not appreciably better in studies in which patients have very low levels of residual symptomatology.”
“The more medications the patients received, the greater the psyosocial functioning impairment.”
Says still don’t know if it’s a question of the drugs impairing psychosocial function, or rather it is that the impaired are getting more drugs.
5. Huxley. Disability and its treatment in bipolar disorder. Bipolar Disorders 2007:9:183-196.
Cognitive functioning. When you discontinue lithium treatment, you get statistically significant improvements in tests of memory and response times, followed by worsening after lithium was restarted.
He notes that other cognitive impairments may be chemically induced (by antipsychotics, etc.)
6. Malhi, 2007. Neuropsychological deficits and functional impirment in bipolar depression, hypomania, and euthymia. Bipolar disorders 2007, 9, 114-125.
Modest impairment in executive functioning, memory and attention in both hypomanic and depressed bipolar patiens, with additional fine motor skills impairment in the latter. Notes that in euthymic patients, subtle impairments in attention and memory suggest that an absence of symptoms does not necessarily equate to recovery.
E. Physical Illness
2005. David Kupfer. The Increasing Medical Burden in Bipolar Disorder. JAMA, May 25, 2005. Vol. 293, no. 20. 2528-2530.
Have to get these studies related to Increased diabetes, obesity, thyroid disease.
Obesity and being overweight are highly prevalent in patients with bipolar disorder and related to both adverse psychiatric and adverse medical outcome. “It is not yet clear whether obesity is a feature of bipolar disorder itself or of the treatments used to manage it.”
However, he says: Treatment factors such as toxicity from medications that currently lead to poor outcomes in these patients must be acknowledged.”
F. Psychiatry’s Recognition of Drug-Induced Worsening.
a) First identified by Kukopolos in 1983
b) In 2000, Zarate Paper.
Zarate, Functional Impairment and cognition in bipolar disorder. Psychiatric Quarterly, 71, no. 4, winter 2000. 309-329.
This group of researchers, which is from the Bipolar and Psychotic Disorders Program at the University of Massachusetts, reviewed the decline in outcomes.
The problem: “In the era prior to modern pharmacotherapy, poor outcome in mania was considered a relatively rare occurrence . . . In contrast, more modern do not describe such a favorable outcome in patients with bipolar disorder. “
The cause: They said one reason for the worsening outcomes in the literature may be due to changes in design and diagnostic criteria. But then they pointed their finger squarely at the drugs:
“Medication induced changes may be yet another factor in explaining the discrepancies in recovery rates between earlier and more recent studies. It is possible that as clinicians e have been contributing to a worsening of the course of the illness.”
a) Antidepressants increase risk of switching and rapid cycling
b) Antipsychotics increase depressive episodes
c) Neuroleptics lead to lower functional recovery rates as well
3. 2003. Ghaemi, “Antidepressants in bipolar disorder: the case for caution.” Bipolar disorders 2003:5 421-33.
In his review, he concludes:
“There are significant risks of mania and long-term worsening of bipolar illness with antidepressants.”
The antidepressants lead to “an increased risk of cycle acceleration with antidepressants.” A three-time increase in manic episodes. Associated with rapid cycling in at least 20% of patients.
Conclusion: The use of antidepressants leads to “long-term mood destabilization” in a significant percentage of patients.
4. Kupfer’s paper above.
Acknowledged drugs causing worsening physical health related to poor health.
5. We see papers linking the drugs to cognitive decline.
6. 2008. Ghaemi. Treatment of Rapid-cylcing Bipolar disorder: Are Antidepressants mood destabilizers? AJP, 165:3, March 2008.
Rapid cycling is first identified in the 1970s, after introduction of lithium. These patients do worse in long-term followup.
New data from Schenck (above) note that “the major predictor of worse outcome was antidepressant use, which about 60% of patients received.”
“This shows that “antidepressants are associated with worsened course of illness even after adjustment for severity of baseline depression.”
In my own clinical experience, “most cases of refractory bipolar disorder, usually of he rapid cycling variety, are due to the mood-destabilizing effects of antidepressants.” He notes this is different from an AD-induced switch to mania. “Mood destabilization is a long-term phenomenon, reflecting more mood episodes over time than would have occurred by natural history.”
G. Bipolar Transformed.
Before: 1 in 10,000.
Now: 1 in 25.
For modern prevalence, see: Merikangas. Lifetime and 12-month prevalence of bipolar spectrum disorder in the national comorbidity survey Replication. Archives of General Psychiatry, 64, May. 543-552.
Zarate and Tohen:
“Prognosis for BPD was once considered relatively favorable, but contemporary findings suggest that disability and poor outcomes are prevalent, despite major therapeutic advances.”
Before: 85% had good long-term outcomes.
Now: Only 33% achieve full social and occupational functional recovery to their own premorbid levels.”
Different Symptom Course:
Tarate and Zohen:
Before: Going back to Kraepelin, the course of BPD was understood to be “marked by time-limited acute episodes of mania and major depression with recovery to euthymia and a favorable functional adaptation between episodes.”
Now: They understand that bipolar disorder is characterized by “slow or incomplete recovery from acute episodes, continued risk of recurrences, and sustained morbidity over time even with continuous long-term use of modern treatments.”
Before: 85% employment
Now: 35% employment
Before: No evidence of it.
Now: Dysfunction in multiple realms; nearly as bad as for schizophrenics.
Physical Illness/early mortality
Before: Not sure of this
Now: Obesity, cardiovascular problems, early death by 25 years.
Outcomes in Bipolar II.
Before: Didn’t exist. Unrecognized.
Now: Zarate and Tohen:
“Remarkably there is some evidence that functional outcome in type 11 BPD may be even worse than type 1.”
50% symptomatic, poor functional recovery, some cognitive impairment.
Now seen as multi-system disease
“Mood disorders are increasingly recognized as multisystem disorders that affect immunologic, endocrine, vascular and neural functions.”
Before: Rare disease. One in every 15,000 people hospitalized for it. Most recovered.
Now: The WHO has announced that bipolar is now the sixth leading cause of disability in the world today.
WHAT ABOUT THE STEP-BD STUDY?
FIRST: THE STEP-BD PAPER
The Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD)
drug study, located at http://content.nejm.org/cgi/content/full/NEJMoa064135
, paid for by NIMH, was "designed to evaluate the effectiveness of
treatments for bipolar disorder and to provide results that are generalizable
to routine clinical practice." They took a bunch of people classified as
"bipolar" who were on mood stabilizers and added either a placebo or
an antidepressant to the med mix. (They selected 2 antidepressants thought not
to cause mania as much as others: bupropion and paroxetine.)
They started with 366 people and split them into two equal groups (adding a
placebo to the mood stabilizer in one group, and adding an antidepressant to
the mood stabilizer in the other group.) By the end of the 16 week study, only
125 people (1/3 of the original number) remained in the study. They objective
was to drive people to 'durable recovery,' which they defined as 8 consecutive
weeks of "euthymia."
The study got a P value of 0.40 -- that's 40% -- on the 'durable recovery'
outcome metric. (Just to put that in perspective, a P value of 0.50 would mean
that the number of 'durable recoveries' would be identical in the two groups.)
So to sum up the study:
* They gave drugs to everyone -- there were no drug-free people in the study
* 2/3 of the participants dropped out of the 16 week study (and 2/3 dropped out
of both treatment groups -- dropout rates didn't differ)
* It didn't matter whether they just used mood stabilizers or they added
antidepressants to the mix -- 'durable recoveries' were the same either way
It's no surprise that Elias Zerhouni said ""We know that medication
is an important component in the treatment of bipolar illness" --
apparently they knew it without even testing a non-drug option!
SECOND: THE PSYCHOSOCIAL TREATMENTS PAPER:
In this study, located at http://archpsyc.ama-assn.org/cgi/content/full/64/4/419
, they once again took a bunch of people who were on drugs. This time, instead
of mixing up the drugs (with no effect), they added therapy. But therapy
actually helps people...
Here's what they did:
They either put a subject in the control group or one of the therapy groups.
The control group got some videotapes. (It is not reported whether microwavable
popcorn was provided along with the videotapes.) They also got some workbooks
and a few one-hour one-on-one study sessions during which a 'treatment
contract' was discussed. This control group was called "Collaborative
Care" (CC) in the study. I have taken the liberty of renaming it
"Cassette Care" (CC). (It is possible they used DVDs, but I'm still a
Four therapies were tested: Cognitive-Behavioral Therapy (CBT), Interpersonal
and Social Rhythm Therapy (IPSRT), and Family-Focused Therapy (FFT).
Figure 3 (http://archpsyc.ama-assn.org/cgi/content/full/64/4/419/YOA60070F3)
of the study shows the results:
Cassette care had the lowest 'recovery' rates: that's the red line on top.
Slightly better than 'cassette care' was CBT. (Hey, at least CBT involves a
live human therapist.)
The other two therapies -- FFT and IPSRT are clearly better. I.e., therapy
works. (But do we need charts with colored lines to know this?)
Elias Zerhouni says that this study "suggests that adding specific,
targeted psychotherapy to medication may help give patients a better shot at
lasting recovery." Thanks, Doc. I have a further suggestion: psychotherapy
without medication, maybe even of the "untargeted" kind (if there is
such a thing as "untargeted" psychotherapy.)
I think that measuring statistically meaningful effects of therapy on a large
sample of human beings is challenging, and whether or not drugs make someone
high (or "euthymic," if it's a psychiatric drug) for 8 consecutive
weeks doesn't seem like a very interesting metric to me. (Just because you can
measure something, doesn't mean it's useful...)
It's been suggested by many people, after the STEP-BD paper, that longer drug
treatments should be tried to see if we can get any results. I would respond
with two points:
* Drug treatment studies are designed to be short-term -- they measure how high
you are on the drugs. I don't personally think they'll work as well
* If 2/3 of the patients dropped out in a 16 week study, how many would drop
out in a longer study? I don't think longer studies are feasible.