A suicidal ideation is based on the score (>1) of item 9 of the Montgomery-Åsberg Depression Rating Scale. HAD indicates Hospital Anxiety and Depression Scale; SSRI, selective serotonin reuptake inhibitor.
Courtet P, Lopez-Castroman J, Jaussent I, Gorwood PAPM. Antidepressant Dosage and Suicidal Ideation. JAMA Intern Med. 2014;174(11):1863-1865. doi:10.1001/jamainternmed.2014.4509
In a recent issue of JAMA Internal Medicine, Miller and colleagues1 proposed that the risk of a suicide attempt doubles among patients 24 years or younger if the initial prescription of selective serotonin reuptake inhibitors (SSRIs) exceeds the recommended dosage. Their findings unveiled the importance of antidepressant dosage on suicidal acts but could not take into account a key risk factor, suicidal ideation (SI), and some potential clinical confounders. Indeed, the effect of type and dosage of antidepressant on the worsening of SI at treatment onset could inform clinical decisions but has been barely studied.
The French National Medical Council approved the study, and written informed consent was obtained from each patient. We analyzed a subgroup of 767 adult outpatients included in a prospective naturalistic study2 after the onset of antidepressant treatment for a major depressive episode. The patient’s regular physician selected the antidepressant drug without restraints. Depression severity and SI were assessed through the Hospital Anxiety and Depression Scale and by item 9 of the self-rated Montgomery-Åsberg Depression Rating Scale, respectively. The participants completed these questionnaires at baseline and at weeks 1, 2, and 6. Patients were divided into 2 groups based on whether or not their SI scores increased during the follow-up period. Type of antidepressant (SSRI vs non-SSRI) and its dosage at treatment onset (modal vs high-dose) were compared in accord with the criteria used by Miller et al.1 Associations between SI worsening and different characteristics of patients and treatments were quantified with odds ratios (ORs) and their 95% CIs. Sociodemographic and clinical variables associated with SI worsening (at P < .15) were included in logistic regression analyses to estimate the adjusted OR for treatment type and dosage. The significance level was set at P < .05.
The SI scores increased for 113 patients (14.7%), and 20 patients (2.6%) attempted suicide during the follow-up period. Findings were unchanged when suicide attempters were excluded from the analysis. No significant associations were found between SI worsening and sociodemographic variables, history of depression or suicidal behavior, and baseline anxiety or hopelessness symptoms (Table). In contrast, SI worsening was associated with the use of non-SSRIs (OR, 1.63; 95% CI, 1.08-2.44), even after adjusting for sex, history of alcohol dependence, duration of the current major depressive episode, and baseline depression severity (OR, 1.66; 95% CI, 1.10-2.51). An interaction between dosage and type of treatment was observed for SI worsening (P = .04, Wald χ2 test). Therefore, the analyses were stratified into 2 groups: patients taking SSRIs and patients taking non-SSRIs (Figure). A significant association with SI worsening was found for high-dose agents at the onset of antidepressant treatment if the agent was an SSRI (OR, 2.07; 95% CI, 1.07-3.97) but not if the agent was a non-SSRI (OR, 0.83; 95% CI, 0.47-1.47). Previous failures with antidepressant medication were not associated with SI worsening. Worsening of depression, anxiety, or psychomotor activation during the follow-up period was not associated with antidepressant type or dosage.
Consistent with the results by Miller et al,1 beginning an SSRI regimen at a higher dosage than recommended (occurring in 27.8% of patients) doubled the risk of SI worsening during the first 6 weeks of treatment. This effect was independent of depression severity or suicidal risk and was not restricted to youth. In addition, as previously reported,3 a similar risk of increased SI was found when patients received non-SSRIs. Our data suggest that the use of SSRIs may reduce the risk of SI worsening relative to other antidepressants, but caution should be exercised with the dosage. Patients 25 years or older may also need to “start low, go slow”4 when initiating a treatment for depression. Although further research is needed, these findings may help physicians choose the most appropriate type and dosage of antidepressant.
Corresponding Author: Philippe Courtet, MD, PhD, Centre Hospitalier Régional Universitaire de Montpellier, Institut National de la Santé et de la Récherche Médicale Unité 1061, Université Montpellier, Pavillon 42, 39 Ave Charles Flahault, Boite Postale 34493, 34093 Montpellier CEDEX 5, France (firstname.lastname@example.org).
Published Online: September 22, 2014. doi:10.1001/jamainternmed.2014.4509.
Author Contributions: Dr Jaussent had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Courtet, Jaussent, Gorwood.
Acquisition, analysis, or interpretation of data: Courtet, Lopez-Castroman, Jaussent.
Drafting of the manuscript: Courtet, Lopez-Castroman, Jaussent.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Jaussent.
Obtained funding: Gorwood.
Administrative, technical, or material support: Lopez-Castroman.
Study supervision: Courtet, Gorwood.
Conflict of Interest Disclosures: Dr Courtet reported receiving during the past 5 years research grants from Servier and honoraria for presentations at congresses or participation in scientific boards from AstraZeneca, Bristol-Myers Squibb, Janssen, Lilly, Lundbeck, Otsuka, Roche, and Servier. Dr Lopez-Castroman reported receiving during the past 5 years honoraria for participation in scientific boards from Servier. Dr Gorwood reported receiving during the past 5 years research grants from Eli Lilly and Servier and honoraria for presentations at congresses or participation in scientific boards from AstraZeneca, Bristol-Myers Squibb, Janssen, Lilly, Lundbeck, Roche, and Servier. No other disclosures were reported.
Funding/Support: The sample was recruited with support by a research grant from Servier (Drs Courtet and Gorwood).
Role of Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.