Article

Do antidepressants cause suicidality in children? A Bayesian meta-analysis

Department of Pediatrics, The Ohio State University, Columbus, Ohio, United States
Clinical Trials (Impact Factor: 1.93). 02/2006; 3(2):73-90; discussion 91-8. DOI: 10.1191/1740774506cn139oa
Source: PubMed

ABSTRACT

To quantify the risk of suicidal behavior/ideation (suicidality) for children who use antidepressants, the FDA collected randomized placebo-controlled trials of antidepressant efficacy in children. Although none of the 4487 children completed suicide, 1.7% exhibited suicidality. The FDA meta-analyzed these studies and found sufficient evidence of an increased risk to require a black-box warning on antidepressants for children.
The FDA considered different drug formulations and psychiatric diagnoses to be equivalent in their effect on suicidality. If this assumption does not hold, the FDA analysis may have underestimated the variance of the risk estimate. We investigate the consequences of relaxing these assumptions.
We extend the FDA analysis using a Bayesian hierarchical model that allows for a study-level component of variability and facilitates extensive sensitivity analyses.
We found an association between antidepressant use and an increased risk of suicidality in studies where the diagnosis was major depressive disorder (odds ratio 2.3 [1.3, 3.8]), and where the antidepressant was an SSRI (odds ratio 2.2 [1.3, 3.6]). We did not find evidence for such an association in the complement sets of trials. Although the results based on the hierarchical model are insensitive to model perturbations, the robustness of the FDA's meta-analysis to model assumptions is less clear. These data have limited generalizability due to exclusion of patients with baseline risk of suicide and the use of relatively short duration trials.
Because of model specification and interpretation issues raised in this paper, we conclude that the evidence supporting a causal link between antidepressant use and suicidality in children is weak. The use of Bayesian hierarchical models for meta-analysis has facilitated the incorporation of potentially important sources of variability and the use of sensitivity analysis to assess the consequences of model specifications and their impact on important regulatory decisions.

Download full-text

Full-text

Available from: Joel B Greenhouse, Aug 23, 2014
  • Source
    • "The association of blood phenylalanine levels with IQ was meta-analyzed using a hierarchical mixed-effects model, estimated using Markov chain Monte Carlo methods (Gelman et al 2003). The advantages of using a Bayesian approach to meta-analysis were recognized over a decade ago (Smith and Spiegelhalter 1995), and they have been applied extensively ever since (Tweedie et al 1996; Sutton and Abrams 2001; Brophy and Joseph 2001; Brophy and Bélisle 2003; Babapulle et al 2004; Kaizar et al 2006; Afilalo et al 2008). It allows for straightforward, probabilistic inference across studies, and readily combines both fixed and random effects. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Though the control of blood phenylalanine (Phe) levels is essential for minimizing impairment in individuals with phenylketonuria (PKU), the empirical basis for the selection of specific blood Phe levels as targets has not been evaluated. We evaluated the current evidence that particular Phe levels are optimal for minimizing or avoiding cognitive impairment in individuals with PKU. This work uses meta-estimates of blood Phe-IQ correlation to predict the probability of low IQ for a range of Phe levels. We believe this metric is easily interpretable by clinicians, and hence useful in making recommendations for Phe intake. The median baseline association of Phe with IQ was estimated to be negative, both in the context of historical (median = -0.026, 95 % BCI = [-0.040, -0.013]) and concurrent (-0.007, [-0.014, 0.000]) measurement of Phe relative to IQ. The estimated additive fixed effect of critical period Phe measurement was also nominally negative for historical measurement (-0.010, [-0.022, 0.003]) and positive for concurrent measurement (0.007, [-0.018, 0.035]). Probabilities corresponding to historical measures of blood Phe demonstrated an increasing chance of low IQ with increasing Phe, with a stronger association seen between blood Phe measured during the critical period than later. In contrast, concurrently-measured Phe was more weakly correlated with the probability of low IQ, though the correlation is still positive, irrespective of whether Phe was measured during the critical or non-critical period. This meta-analysis illustrates the utility of a Bayesian hierarchical approach for not only combining information from a set of candidate studies, but also for combining different types of data to estimate parameters of interest.
    Full-text · Article · Nov 2012 · Journal of Inherited Metabolic Disease
  • Source
    • "However, this result was sensitive to variation in the clinical effectiveness parameters with non-directive counselling dominating TF-CBT and TF-CBT + SSRI under certain assumptions. Moreover, while the best available evidence suggests that TF-CBT + SSRI is more effective (and more cost-effective) than TF-CBT alone, there is a continuing debate about risks of suicidal behaviour/ideations associated with prescribing SSRIs to children and adolescents [31,66]. The limited clinical evidence used in the present study did not produce any support for the elevated risk of suicidal behaviour/ideations in the population of children and adolescents treated for PTSD secondary to sexual abuse with a combination therapy of TF-CBT + SSRI [33]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Post-traumatic stress disorder (PTSD) is diagnosed in 20% to 53% of sexually abused children and adolescents. Living with PTSD is associated with a loss of health-related quality of life. Based on the best available evidence, the NICE Guideline for PTSD in children and adolescents recommends cognitive behavioural therapy (TF-CBT) over non-directive counselling as a more efficacious treatment. A modelled economic evaluation conducted from the Australian mental health care system perspective estimates incremental costs and Quality Adjusted Life Years (QALYs) of TF-CBT, TF-CBT combined with selective serotonin reuptake inhibitor (SSRI), and non-directive counselling. The "no treatment" alternative is included as a comparator. The first part of the model consists of a decision tree corresponding to 12 month follow-up outcomes observed in clinical trials. The second part consists of a 30 year Markov model representing the slow process of recovery in non-respondents and the untreated population yielding estimates of long-term quality-adjusted survival and costs. Data from the 2007 Australian Mental Health Survey was used to populate the decision analytic model. In the base-case and sensitivity analyses, incremental cost-effectiveness ratios (ICERs) for all three active treatment alternatives remained less than A$7,000 per QALY gained. The base-case results indicated that non-directive counselling is dominated by TF-CBT and TF-CBT + SSRI, and that efficiency gain can be achieved by allocating more resources toward these therapies. However, this result was sensitive to variation in the clinical effectiveness parameters with non-directive counselling dominating TF-CBT and TF-CBT + SSRI under certain assumptions. The base-case results also suggest that TF-CBT + SSRI is more cost-effective than TF-CBT. Even after accounting for uncertainty in parameter estimates, the results of the modelled economic evaluation demonstrated that all psychotherapy treatments for PTSD in sexually abused children have a favourable ICER relative to no treatment. The results also highlighted the loss of quality of life in children who do not receive any psychotherapy. Results of the base-case analysis suggest that TF-CBT + SSRI is more cost-effective than TF-CBT alone, however, considering the uncertainty associated with prescribing SSRIs to children and adolescents, clinicians and parents may exercise some caution in choosing this treatment alternative.
    Full-text · Article · Apr 2012 · Child and Adolescent Psychiatry and Mental Health
  • Source
    • "Specifically, all of the 95% credible intervals for the 80 EE Kaizar et al. Clinical Trials 2006; 3: 73–98 www.SCTjournal.com Table 2 Posterior summary measures for fixed effects models (␶ 2 ϭ 0, ␻ 2 ϭ 0) and random effects models (␻ 2 ϭ 0)"
    [Show abstract] [Hide abstract]
    ABSTRACT: To quantify the risk of suicidal behavior/ideation (suicidality) for children who use antidepressants, the FDA collected randomized placebo-controlled trials of antidepressant efficacy in children. Although none of the 4487 children completed suicide, 1.7% exhibited suicidality. The FDA meta-analyzed these studies and found sufficient evidence of an increased risk to require a black-box warning on antidepressants for children. The FDA considered different drug formulations and psychiatric diagnoses to be equivalent in their effect on suicidality. If this assumption does not hold, the FDA analysis may have underestimated the variance of the risk estimate. We investigate the consequences of relaxing these assumptions. We extend the FDA analysis using a Bayesian hierarchical model that allows for a study-level component of variability and facilitates extensive sensitivity analyses. We found an association between antidepressant use and an increased risk of suicidality in studies where the diagnosis was major depressive disorder (odds ratio 2.3 [1.3, 3.8]), and where the antidepressant was an SSRI (odds ratio 2.2 [1.3, 3.6]). We did not find evidence for such an association in the complement sets of trials. Although the results based on the hierarchical model are insensitive to model perturbations, the robustness of the FDA's meta-analysis to model assumptions is less clear. These data have limited generalizability due to exclusion of patients with baseline risk of suicide and the use of relatively short duration trials. Because of model specification and interpretation issues raised in this paper, we conclude that the evidence supporting a causal link between antidepressant use and suicidality in children is weak. The use of Bayesian hierarchical models for meta-analysis has facilitated the incorporation of potentially important sources of variability and the use of sensitivity analysis to assess the consequences of model specifications and their impact on important regulatory decisions.
    Full-text · Article · Feb 2006 · Clinical Trials
Show more