Editor’s Note: Over the next several months, Mad in America is publishing a serialized version of Peter Gøtzsche’s book, Critical Psychiatry Textbook. In this blog, he covers how the textbooks discuss the use of depression pills in pregnancy, as well as how the textbooks discuss psychotherapy and psycho-education. Each Monday, a new section of the book is published, and all chapters are archived here.
Depression pills are among the most commonly used drugs by females in their reproductive age. In Denmark, 8% in the age group 18-44 years take them.263 This is worrying, as the drugs seem to increase miscarriages, voluntary terminations, birth defects, and behavioural abnormalities in newborns,483,484 and they cause many other serious harms in the offspring.336
The advice about pregnancy was inconsistent and confusing. The textbooks generally put the blame on the disease, not on the pills. For example, one book warned that depression doubles the risk of developing cardiovascular disease16:259 and potentially increases heart malformations and neonatal complications.16:584
Another textbook was confusing, contradictory, and misleading. It warned that depression increases the risk of abnormal bleeding during pregnancy, spontaneous abortion, premature birth, foetal death, eclampsia, other birth complications, poor quality of life for the child, and lack of breastfeeding.17:364 However, on the same page, the authors noted that depression pills are possibly associated with a slightly increased risk of premature birth and perinatal complications, and 13 pages later, that the drugs likely increase the risk of malformations.17:377
After another 291 pages, this book contradicted itself again and tried to have it both ways in a most confusing fashion:17:668 Untreated depression can cause premature birth and perhaps also malformations. Depression pills can perhaps increase spontaneous abortions, but the newest studies speak against malformations. Nonetheless, the authors noted that paroxetine is possibly associated with heart malformations and neonatal complications and that there is an increased risk of pulmonary hypertension in newborns, which can be deadly.
If you don’t know what to say, it is prudent to say nothing instead of confusing your readers totally. I cannot make any sense out of the above, and it got worse. This book noted that the Danish National Board of Health recommend always to consider psychotherapy for pregnant women who are depressed.17:365 Indeed; none of them should get pills. But lo and behold, just one page earlier, the book advised that pregnant women who have been depressed earlier should be treated prophylactically with depression pills to reduce the risk of relapse from about 70% to about 25%.17:364 It is impossible to justify this horrific recommendation.
The Board of Health also contradicted itself. It recommended routine screening of pregnant women for depression and subsequent treatment with depression pills, although the available data do not support these recommendations.485 It acknowledged that SSRIs increase the occurrence of spontaneous abortions, decrease birth weight, likely increase the occurrence of birth defects, increase the risk by a factor of five for developing pulmonary hypertension, which is a lethal harm estimated to occur in 6-12 newborns per 1,000, and increase neonatal complications such as irritability, tremor, hypertonia, and difficulty sleeping or breast feeding.485 An article about this appropriately called it neonatal abstinence syndrome.486
Excuse me, but have they gone mad at the Board of Health? A large Danish cohort study of 500,000 children showed that the risk of heart septum defect is doubled.487 This is not trivial, as 1% of the treated foetuses will get a septum defect. Cardiac birth defects are exactly what we would expect to see because serotonin plays a major role for the functioning of the heart. We have seen deadly valvular defects and deadly pulmonary hypertension in adults who took diet pills that increase serotonin levels, and these drugs have been withdrawn from the market.6:144
The Board’s recommendation of screening was so absurdly harmful that I wrote a little sketch about it,488 which a psychologist and I spontaneously performed as the introduction to my lecture about psychiatry by reading it aloud from my computer. It is on YouTube with English subtitles.489
Among its many weird postulates in relation to pregnancy, this book also claimed that the risks of depression and behavioural disorders are increased in 18-year-old children of mothers who were not treated during pregnancy for their depression.17:365
As I didn’t believe this could be true for drugs that don’t work, I looked up the evidence the authors referred to, which was a 2014 clinical guideline for the use of psychiatric drugs during pregnancy produced by the Danish Psychiatric Association, Danish Society for Obstetrics and Gynaecology, Danish Paediatric Society, and Danish Society for Clinical Pharmacology.490 With so many knowledgeable people involved, one would expect the guideline to be reliable, but it can best be described as being blatantly dishonest.
The guideline stated that there is “an increased incidence of depression in 18-year-old children of mothers who were not treated during pregnancy for their depression (Pearson et al., 2013)” and that “untreated depression during pregnancy seems to increase the risk of developing behavioural disorders in the child (Pedersen et al., 2013).”
None of this was true. The article by Pearson et al. didn’t say anything about whether the women were treated or not during their depression. What the paper showed was that if a mother was depressed, the risk of her offspring becoming depressed was increased, but only for mothers with low education.491 This has nothing to do with treating or not treating a depression, but with poor living conditions, which is often also the case for the offspring. When living conditions are depressing, people become depressed. No great wonder here.
The article by Pedersen et al. did not document at all that untreated depression increases the risk of behavioural disorders in the child.492 This was clear already in the abstract: “Prenatal antidepressant exposure was not associated with abnormal SDQ scores compared with prenatal exposure to untreated prenatal depression or to no exposure.” But the abstract also reported the results of what we call a fishing expedition. When a result is negative, it is very bad research practice to report on subgroups of patients or on selected items on a scale, but this is what the authors did: “Untreated prenatal depression was associated with abnormal SDQ scores in the subscales of conduct [adjusted odds ratio (aOR) 2.3 (95% CI, 1.2-4.5)] and prosocial problems [aOR 3.0 (95% CI, 1.2-7.8)] compared with unexposed children.” They not only selected items on a scale, they also did not compare untreated depression with treated depression but with people who were not depressed at all but were healthy!
In the tables, there wasn’t a single significant difference between the total score or any of the subitems in the score when treated women with depression were compared with untreated women with depression. But the authors had been fishing again to find what they reported in the abstract for conduct: “Including only women with normal MDI [depression] score at time of follow-up”. For prosocial problems, I was unable to find the adjusted odds ratio of 3.0, which was claimed to be statistically significant. It was nowhere in the paper, but there was this information: “The prosocial association was no longer statistically significant, OR 2.2 (95% CI, 0.8-6.5)” (when including only women with a normal depression score).”
It is amazing that such rubbish with tortured data analyses can get published, but the research literature is full of this. A systematic review found that subgroup analyses in trials were more common in high-impact journals; and in trials without statistically significant results for the primary outcome, industry-funded trials were twice as likely to report subgroup analyses as non-industry-funded trials and twice as likely not to have prespecified the subgroup hypotheses.493
This textbook noted that valproate and carbamazepine are contraindicated due to a high risk of neural tube defects.17:669 I wonder why the authors did not warn against all antiepileptic drugs.
Psychotherapy and psychoeducation
Psychotherapy is not a magic bullet against psychiatric disorders. It doesn’t always work, but it is the best intervention we have.
The textbooks were sometimes contradictory and misleading. One noted that 50% of patients with depression are not treated; that many of them likely have mild depression; and that psychotherapy will shorten the disease phase, prevent chronicity, and provide obvious relief for the patients.16:257 Unfortunately, the book advised that SSRIs or tricyclics could be used instead of psychotherapy for moderate depression or in combination with it. For severe depression, psychotherapy was not advised, but hospital admission, tricyclics, tricyclics plus psychosis pills, and electroshock were.16:272
This is a familiar theme. The worse the disease, the more the patients shall be harmed by treatments that don’t help them. This is not evidence-based medicine.
The book where all the authors are psychiatrists denigrated psychotherapy stating that pills can be combined with talk therapy with advantage, which also increases compliance.18:238 So, pills first, even though they don’t work, and psychotherapy is only aimed at keeping the patients on pills that harm them and which many patients would rather avoid. When asked what they prefer, six times as many people prefer psychotherapy for pills,494 but they get the exact opposite. A 2002 survey of US child and adolescent psychiatrists showed that 91% of their patients were treated with psychiatric drugs.495 Only in the remaining 9%, was psychotherapy used without drugs.
In Sweden, the National Board of Health recommends that all adults with mild to moderately severe depression are offered psychotherapy, but only 1% get it.496
This illustrates that psychiatry is a perverse trade. It doesn’t help the patients as they want to be helped, but helps itself.
This textbook recommended watchful waiting or supportive conversations for mild depression, psychotherapy for moderate depression, and depression pills for more severe depression.18:123 The authors claimed that the preventative effect of drugs was more pronounced than that of psychotherapy,18:126 which is false and was contradicted by another book, which noted that the effect of psychotherapy lasts longer than that for drugs.16:277 As expected, studies with long-term follow-up show that psychotherapy has an enduring effect that outperforms pharmacotherapy,180,497-501 and when psychiatrists believe pills prevent relapse, they mistake withdrawal effects for relapse (see Chapter 7, Part Four and Chapter 8, Part Twelve).
A third textbook advised psychotherapy for moderate and severe depression,17:359,17:363 but not for severe depression requiring hospital admission.17:359 It noted that psychotherapy should most often be considered when the patient is in remission17:363 and claimed that a halving of the depression score was obtained in 60% of the patients treated with drugs and psychotherapy.17:359
This statement is meaningless. It cannot be interpreted without knowing what happened to the other 40% of the patients. If their score increased markedly, the overall effect might be zero. Evidence-based medicine is not about what happened in some selected subgroup of patients, but about what happened on average. These authors considered psychotherapy a secondary option, which contradicted a chapter about psychotherapy in the same book where other authors noted that the effect size in a meta-analysis was quite high, and that in many cases, psychotherapy was cost-effective in comparison with drugs.17:675 It seems to be correct that psychotherapy is more cost-effective than other forms of therapy.502
A fourth book also put the pills first even though it noted that the effect of psychotherapy and pills was about the same for mild and moderate depression.20:435 This is misleading because the effect is also about the same in severe depression.503 The book noted that the National Board of Health had found a better effect of combining psychotherapy with pills than of pills alone, but it did not mention that the Board’s guideline strongly recommended to offer psychotherapy, in combination with pills, to patients with moderate or severe depression.504 This was contradicted by another book, which noted that for mild or moderate depression, there was no evidence of a greater effect of the combination than of drug or psychotherapy alone, while it claimed that this was the case for chronic depression.16:278
It was totally confusing. And why would a combination work for chronic depression when it did not work for moderate depression, and what is chronic depression? The guideline from the Board of Health had an important reservation: “Combination therapy has demonstrated an increased effect over monotherapy, but patients have often not been followed beyond the end of the intervention. The working group wants to clarify the long-term effects of combination therapy consisting of antidepressant pharmacotherapy and psychotherapy.”
It is remarkable that three textbooks did not recommended psychotherapy for severe depression,16,18,20 and that a fourth book did not recommend it for depression requiring hospital admission.17 The only book that advised psychotherapy for severe depression was the one about child and adolescent psychiatry,19:214 but, unfortunately, this book advised that psychotherapy should be combined with fluoxetine, which is unsafe and ineffective (see Chapter 8, Part Four).
One book stated that treatment of bipolar in children involves drugs, in addition to psycho-education, but did not say that drugs should only be used if psychoeducation did not work.19:220
Two books stated that psychoeducation may halve the risk of new depressions or manias in bipolar patients and reduce hospital admissions but added that this was probably because of better treatment compliance (with drugs).16:306,17:376 One of the books gave a reference to this statement,17:376 which was a randomised trial of psychoeducation.505
It turned out that the textbook claims about better compliance with drug treatment were false.505 The researchers randomised 120 bipolar patients to 21 weekly group psychoeducation sessions or nonstructured group meetings and the effect was assessed blindly. During treatment, 23 vs 36 patients had a recurrence (P < 0.05); at the end of the follow-up, these numbers were 40 vs 55 (P < 0.001); and there were markedly fewer hospital days, 4.8 vs 14.8 (P < 0.05).
At the 2-year follow-up, a tiny difference was found in lithium levels, 0.76 vs 0.68 mEq/L (P = 0.03), whereas there were no differences in the levels of valproate or carbamazepine, and no differences regarding drug treatment.
The authors wrote in the discussion that, “compared with control patients, psychoeducated patients had higher lithium levels at the 2-year follow-up, which may suggest an effect of psycho-education on pharmacotherapy adherence.”
So, the trial authors did not suggest that the tiny difference in lithium levels could explain the pronounced effects they found of psychoeducation. It is bending the data to the extreme when the textbook authors wrote that this was likely, instead of just accepting that psychoeducation is highly effective.
A textbook noted that, although PET studies are preliminary, there is much to suggest that symptom reduction during psychotherapy may normalise metabolism in certain cerebral areas found to be affected during depression.16:269 Brain scan studies are highly unreliable (see Chapter 3), but this was a rare occasion where they were not used to promote drugs but psychotherapy.
To see the list of all references cited, click here.
Mad in America hosts blogs by a diverse group of writers. These posts are designed to serve as a public forum for a discussion—broadly speaking—of psychiatry and its treatments. The opinions expressed are the writers’ own.
Mad in America has made some changes to the commenting process. You no longer need to login or create an account on our site to comment. The only information needed is your name, email and comment text. Comments made with an account prior to this change will remain visible on the site.