GME Logo

User login

User menu

Issue 64, Aug 2017
Share This
Two simple but neglected red flags for subsequent suicide

Birtwistle J, Kelley R, House A, Owens D. Combination of self-harm methods and fatal and non-fatal repetition: A cohort study. J Affect Disord. 2017 Apr 27;218:188-194. PubMed PMID: 28477496.


Predictors of suicide risk in persons who have had self-harm are important in order to plan their care.

This study aimed to assess whether two specific clinical features (discussed below) predicted the risk of subsequently dying by suicide and/or of non-fatal repetition of a suicide attempt.


During the study period of three years, over 10,000 consecutive emergency department visits due to self-harm (by over 6000 patients) were identified.

Consecutive persons who presented to the emergency department after self-harm were enrolled if they were 12 years or older.

Of these persons, 58% were female and 42% were male.

Self-harm was defined as intentional self-poisoning or self-injury, irrespective of what the person said the motivation for the act was. Overdose on illicit drugs or on alcohol was included in self-poisoning.

Combined self-poisoning and self-injury was defined as the person reporting having done both as part of a single act that led to the emergency department visit. Even if the act had taken place over several hours or even over two consecutive days, it was still considered to be a case of combined self-poisoning and self-injury.

In two-thirds of cases, the combined acts involved self-poisoning and self-cutting.


Determining the subsequent outcome for persons included in this study involved the following two steps:

1. These persons were followed up to determine if they had had non-fatal repetition of self-harm.

2. National statistical records were searched to see if these persons had subsequently died by suicide or from another cause. As is conventional in suicide research in the UK, deaths due to undetermined reasons were considered to be suicides.


It is unfortunate, in my opinion that the extent of follow-up varied from person to person but the study tried to take this into account using statistical methods.  


The initial self-harm emergency department visits were due to self-poisoning (72% of episodes), self-injury (21%), or a combination of methods (7%).

So, using a combination of methods was not rare.

Those who had a self-harm episode using combined methods were statistically significantly more likely to subsequently die by suicide. They were also more likely to subsequently have a non-fatal repetition of self-harm.

The non-fatal repetition or subsequent suicide also occurred earlier in persons who had used combined methods.

Also, those who come to the hospital for self-harm again within a month (rapid repetition) were nearly four times more likely to subsequently die by suicide.

Males were more almost three times as likely as females to die by suicide.

Those who were more than 65 years old and self-harmed were more than seven times as likely as young people to subsequently die by suicide.


When assessing a person who presents with self-harm, two important predictors of higher risk are:

- Having used more than one method of self-harm during that episode

- Having had another episode of self-harm within the previous month (rapid repetition)

Clinical Commentary

It is notoriously difficult to predict the risk of suicide in a particular person.

This study draws our attention to two high risk factors that make sense but ones we may not be explicitly looking for.

One problem with suicide risk assessment is that there are many factors that increase risk but individually have a small effect. But, those who had self-harmed using combined methods were three times more likely to subsequently die by suicide. Also, those who had came to the hospital twice for two episodes of self-harm within a month of each other (rapid repetition) were at four-fold greater risk of subsequently dying by suicide.

These findings suggest that, in addition to other risk factors that we know to look for, these two are clinically important risk factors that we should look for.

Expand Content
Share this Section:
What does antidepressant-induced hypomania/mania mean?

Barbuti M, Pacchiarotti I, Vieta E, Azorin JM, Angst J, Bowden CL, Mosolov S, Young AH, Perugi G; BRIDGE-II-Mix Study Group. Antidepressant-induced hypomania/mania in patients with major depression: Evidence from the BRIDGE-II-MIX study. J Affect Disord. 2017 May 24;219:187-192. PubMed PMID: 28558366.



What should we conclude if a patient without a history of bipolar disorder is given an antidepressant and switches to hypomania or mania?

Does it mean that the person has bipolar disorder? Or should it be considered to be a substance-induced mood disorder?

This paper analyses data on antidepressant-induced hypomania/mania from a large, international, multicenter study of persons with a major depressive episode.


Data on more than 2800 patients with a major depressive episode in an international study called the BRIDGE-II-MIX study were analyzed. The patients came from over 200 centers in eight countries (the USA was not among them).

Note: these patients had a major depressive episode, not necessarily major depressive disorder. The major depressive episode could be due to either:


1. Major depressive disorder without a history of antidepressant-induced hypomania/mania (67% of the sample)

2. Major depressive disorder with a history of antidepressant-induced hypomania/mania (MDD-AIHM; 17% of the sample)

3. Bipolar disorder (17% of the sample)


Unfortunately, in my opinion the definition of hypomanic or manic episode was not specified. It was based only on the retrospective judgment of an experienced psychiatrist. So, the duration of the episode and whether or not the symptoms continued after the antidepressant was stopped did not matter with regard to calling the episode “antidepressant-induced hypomania/mania”.

Also, data on which antidepressant was used and how long it was continued for were not gathered.  


Compared to persons with MDD, persons with MDD-AIHM had higher rates of:

1. Bipolar disorder in first-degree relatives (25% versus 10%)

2. Three or more lifetime major depressive episodes (82% versus 47%)

3. First major depressive episode before the age of 30 years (61% versus 46%)

4. Current major depressive episode duration of less than one month (46% versus 29%)

5. Presence of atypical features (12% versus 5%)

6. Presence of DSM-5 criteria for mixed features (17% versus 3%)

7. Comorbid anxiety disorders (but only a small difference – 34% versus 27%)

8. Comorbid eating disorders (10% versus 5%)

9. Comorbid borderline personality disorder (13% versus 4%)

10. Rates of treatment resistance (47% versus 23%)

11. Mood lability after being treated with an antidepressant (60% versus 19%)

12. Irritability after being treated with an antidepressant (53% versus 16%)

13. Suicide attempts (30% versus 19%)

14. Being treated with more than three medications (51% versus 25%)


On the other hand, patients with MDD-AIHM did not differ from those with MDD in rates of alcohol or other substance use.

What about when compared to persons with bipolar disorder? Those with MDD-AIHM were remarkably similar to those with bipolar disorder on most variables.


However, persons with MDD-AIHM were more likely to have:

1. Had three or more lifetime major depressive episodes (82% versus 71%)

2. Comorbid ADHD (4% versus 2%)

3. Treatment resistance to antidepressants (47% versus 30%)

4. Mood lability with antidepressant treatment (60% versus 43%)

5. Irritability with antidepressant treatment (53% versus 37%)


Based on a variety of clinical features, patients with MDD-AIHM were more similar to patients with bipolar disorder than to those with major depressive disorder.

These findings support including patients with antidepressant-induced hypomania/mania under bipolar disorders.

Clinical Commentary

The problems of both overdiagnosis and underdiagnosis of bipolar disorders continue to be common.

When there is a history of antidepressant-induced hypomania/mania, greater uncertainty is introduced.

Antidepressant-induced hypomania/mania is not rare. Of those with a major depressive episode, 17% had a history of it.

In DSM-IV, antidepressant-induced hypomania/mania was not considered to indicate bipolar disorder. But DSM-5 changed how this was conceptualized. DSM-5 states the following:

“A full hypomanic episode that emerges during antidepressant treatment (e.g., medication, electroconvulsive therapy) but persists at a fully syndromal level beyond the physiological effect of that treatment is sufficient evidence for a hypomanic episode diagnosis. However, caution is indicated so that one or two symptoms (particularly increased irritability, edginess, or agitation following antidepressant use) are not taken as sufficient for diagnosis of a hypomanic episode, nor necessarily indicative of a bipolar diathesis.”

Expand Content
Share this Section:
How to treat hypochondriasis

Fallon BA, Ahern DK, Pavlicova M, Slavov I, Skritskya N, Barsky AJ. A Randomized Controlled Trial of Medication and Cognitive-Behavioral Therapy for Hypochondriasis. Am J Psychiatry. 2017 Jun 29:appiajp201716020189. [Epub ahead of print] PubMed PMID: 28659038.  


Hypochondriasis is a condition that is notoriously difficult to treat.

Several studies have suggested that CBT and other psychotherapies can be efficacious. There is very little research on medications to treat hypochondriasis but a couple of studies have found that an SSRI can be helpful.

This is one of the largest studies ever done to evaluate the treatment of hypochondriasis. Also, this is the first study to evaluate the use of a combination of CBT and medication for the treatment of hypochondriasis.


Patients with DSM-IV hypochondriasis (N=195) were recruited. The hypochondriasis was at least moderate in severity.

If comorbidity was present, the hypochondriasis had to be of earlier onset, more severe, and causing more distress than the other disorder.


The participants were randomly assigned to receive 24 weeks of treatment with one of the following:

1. Fluoxetine

2. Cognitive-behavior therapy (CBT)

3. Both fluoxetine and CBT

4. Placebo only


The persons evaluating the outcome were blinded to what treatment each participant was receiving.

Fluoxetine was started at 10 mg/day and increased, as needed and tolerated, up to 80 mg/day.

The CBT was conducted according to a scripted manual that had previously been developed for the treatment of hypochondriasis. Initially, six sessions of one hour each and one per week were done. Then, two more sessions were done at two-week intervals. After that, booster sessions were done once a month.


The CBT involved:


Reformulation of dysfunctional assumptions about symptoms

Modification of confirmatory bias

Reduction of maladaptive sick role behaviors

Identification of situations that exacerbated health anxiety, and

Reduction of bodily hypervigilance


A variety of structured questionnaires were used to assess the diagnosis, severity of symptoms, and change in symptoms.  


Overall, the response rates at 24 weeks were:


Placebo only — 30%

Monotherapy (Fluoxetine alone or CBT alone) — 42%

Combined therapy (Both fluoxetine and CBT) — 47%


The findings in this study were statistically significant when analyzed in some ways but not in others. The main hypothesis of the study was confirmed using a trend analysis – that monotherapy would be more efficacious than placebo and that combined treatment would be better than monotherapy.

However, when the rating scales for hypochondriasis were analyzed as continuous measures, fluoxetine was the treatment that mainly led to improvement.

In one of the ways of analyzing the results, fluoxetine was more effective than placebo and had a faster rate of improvement, but CBT was not.

Patients who received fluoxetine also had statistically significantly less anxiety and better quality of life than those who received placebo. However, the improvement with fluoxetine was not simply due to improvement in depression because the depression scores did not show statistically significant improvement over time.

Higher doses of fluoxetine were more efficacious than lower doses.

Even though hypochondriasis improved, somatization did not improve with any of the treatments. I take that to mean that patients continued to experience physical symptoms, but their interpretation of the symptoms changed.


Fluoxetine was specifically effective for hypochondriasis.

Combined treatment with fluoxetine and CBT led to slightly better outcomes.

The authors speculated that CBT did not do as well in this study as in previous studies perhaps because many patients discontinued the CBT, only six initial sessions of CBT were given, and exposure and motivational interviewing were not part of the treatment.

Clinical Commentary

In the past, we did not consider an SSRI to be a primary treatment for hypochondriasis. So, to many readers, the efficacy of fluoxetine for the primary symptoms of hypochondriasis may come as a surprise, as it did for me.

One of the most striking findings of this study is that even with six months of treatment with both fluoxetine and CBT, less than 50% of patients had even a 25% improvement. This highlights how difficult it is to treat hypochondriasis and how important it is for new treatments to be developed.

By the way, in DSM-5, the term hypochondriasis has been dropped. Most patients who were previously diagnosed as having hypochondriasis are considered to have Somatic Symptom Disorder, but some are diagnosed with Illness Anxiety Disorder.  

Expand Content
Share this Section:
Can TMS be a new treatment option for OCD?

Zhou DD, Wang W, Wang GM, Li DQ, Kuang L. An updated meta-analysis: Short-term therapeutic effects of repeated transcranial magnetic stimulation in treating obsessive-compulsive disorder. J Affect Disord. 2017 Jun;215:187-196. Review. PubMed PMID: 28340445.


Repetitive transcranial magnetic stimulation (rTMS) is being evaluated as a potential treatment for a variety of disorders.

This paper presents the results of a meta-analysis of studies of the use of rTMS for the treatment of obsessive-compulsive disorder (OCD).


A systematic search of the literature was done to identify relevant studies.

Only sham-controlled, randomized clinical trials were included.

The results of these studies were combined using meta-analysis.  


Twenty studies, in which a total of 791 patients had participated, were included. This is the largest meta-analysis done so far on this topic.

rTMS was statistically significantly more efficacious than the sham-control intervention.

The effect size was large (Hedge’s g = 0.7).

With rTMS, stimulation to different areas of the brain has been tried. For OCD, rTMS administered to many different sites was found to be effective:

Supplementary motor area (g=0.6)

Left dorsolateral prefrontal cortex (g=0.5)

Bilateral dorsolateral prefrontal cortex (g=0.65)

Right dorsolateral prefrontal cortex (g=0.9)

Should the stimulation be low-frequency or high-frequency? Both low-frequency (g=0.7) and high-frequency (g=0.7) rTMS were equally effective.

Which patients are more likely to respond? Those patients whose OCD had not failed to respond to medication (i.e., not “treatment-resistant”) and who did not have comorbid major depressive disorder at the time of treatment had greater benefit from rTMS.


rTMS treatment is effective for the short-term treatment of OCD.  

Clinical Commentary

OCD is usually difficult to treat. Partial response and a relapsing course are quite common.

So, this paper’s findings are exciting because it suggests that rTMS can be added to the treatment options for OCD.

Given the effect sizes reported in this meta-analysis, targeting the right dorsolateral prefrontal cortex makes the most sense.

I would love to see research in the future on combining rTMS with SSRIs and/or exposure and response prevention because that is how it is most likely to be used.

Also, it would be important to do research on whether, if stimulating the right dorsolateral prefrontal cortex does not work, targeting other brain areas may work.

Expand Content
Share this Section:
What happens to the newborn when an antidepressant is taken before delivery?

Yang A, Ciolino JD, Pinheiro E, Rasmussen-Torvik LJ, Sit DKY, Wisner KL. Neonatal Discontinuation Syndrome in Serotonergic Antidepressant-Exposed Neonates. J Clin Psychiatry. 2017 May;78(5):605-611. PubMed PMID: 28570796.



This study aimed to determine if, at 2 to 4 weeks after birth, newborns whose mothers had taken a serotonin reuptake inhibitor have signs consistent with neonatal discontinuation syndrome.


In research on neonatal outcomes it is important to separate out the effects of taking an antidepressants from the effects of the mother having a mood disorder.


So, mothers (total n = 214) belonging to the following three groups were included:

1. Women who had a mood disorder and were taking a serotonin reuptake inhibitor during pregnancy (but no benzodiazepines). In most cases (81% of participants), the women were exposed to the serotonin reuptake inhibitor throughout the pregnancy. The most common serotonin reuptake inhibitor taken was sertraline, followed by fluoxetine.

2. Women who had a mood disorder but were not taking any psychotropic medications during the pregnancy.

3. Control group: Women who did not have a mood disorder and did not take any psychotropic medications during pregnancy.


It is very important to realized that because, obviously, women were not randomized to these three groups, they were not similar in other ways, e.g., age, socioeconomic group, etc.

The infants were systematically examined using a standard scale, called the Finnegan scale. Signs evaluated by this scale can be present due to several reasons, including antidepressant discontinuation. The scale included assessments for diarrhea, sleep impairment, temperature, respiratory impairment, tremors, and so on.

The persons doing this evaluation were blinded as to which group the infants belonged to.


At least some of the signs on the scale were present in 30 to 35% of infants and the percentages were not different between the three groups.

Women who had a mood disorder and had taken a serotonin reuptake inhibitor were more likely to have preterm deliveries (24%) than mothers who had a mood disorder but had not taken an antidepressant (7%) or those in the control group (9%).

Infants who were born preterm were more likely to have signs on the Finnegan scale than those born full-term (54% versus 31%).


The presence of neonatal signs at 2 to 4 weeks was associated with premature birth but not directly with presence of mood disorder or with use of a serotonergic antidepressant during pregnancy.

The authors noted that the findings don’t rule out the existence of a neonatal discontinuation syndrome in newborns whose mothers were exposed to a serotonergic antidepressant prior to giving birth. It may be that the neonatal signs had subsided before the age of two weeks when this study’s assessments started. Also, paroxetine has been most frequently associated with neonatal discontinuation syndrome but in this study only one infant had been exposed to paroxetine.  

Clinical Commentary

It may seem that comparing two groups—women with a mood disorder who took an antidepressant and women with a mood disorder who did not—separates out the effects of the mood disorder from those of the antidepressant.

BUT, it is important to realize that in these kinds of studies, for ethical reasons, women are not randomized to receive or not receive an antidepressant.

Thus, for now, the possibility of a neonatal discontinuation syndrome should continue to be considered in the risk-benefit discussion with pregnant women considering taking a serotonergic antidepressant during pregnancy.

To me, the more dramatic finding of this study was that taking a serotonergic antidepressant during pregnancy seemed to more than triple the rate of pre-term delivery compared women who had a mood disorder but were not taking a serotonergic antidepressant (24% versus 7%). If we believe this, it is of great importance in the decision of whether to take an antidepressant during pregnancy.  

Expand Content
Share this Section:

Rajnish Mago, MD
Medical Editor, GME Research Review

GME Research Review is a monthly newsletter edited by Rajnish Mago, MD, who is author of "The Latest Antidepressants" and "Side Effects of Psychiatric Medications: Prevention, Assessment, and Management." Dr. Mago selects, summarizes, and provides a clinical commentary on the latest published research in psychiatry. 

We are always carefully evaluating which research papers to discuss in GME Research Review. Have come across a research paper published in the last 6 months that you thought is clinically relevant? Do you want me to analyze it for you and for the benefit of others? Please email Dr. Mago the citation at [email protected].

To contact GME, email us at [email protected]

GME does not provide medical advice. The website and articles are intended for informational purposes only. They are not a substitute for professional medical advice, diagnosis or treatment. Never ignore professional medical advice in seeking treatment because of something you have read on the GME Website. If you think you may have a medical emergency, immediately call your doctor or dial 911.

Log In

Join GME For Free

Become a GME subscriber and gain full access to our extensive library of 700+ psychiatric medical education videos, free CME webcasts, latest research updates, and more. To sample our content, watch the featured videos below.
Join Today!