Pharmacological treatments for generalised anxiety disorder

Abstract

In their systematic review and network meta-analysis, April Slee and colleagues1Slee A Nazareth I Bondaronek P Liu Y Cheng Z Freemantle N Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis.Lancet. 2019; 393: 768-777Summary Full Text Full Text PDF PubMed Scopus (131) Google Scholar asserted that several pharmacological treatments were effective for generalised anxiety disorder. We would like to highlight several limitations of their analysis. First, the investigators did not assess the overall risk of bias within or across included studies. Accepting their assessments of risk of bias for individual domains and using the categories proposed by the Cochrane Handbook for Systematic Reviews of Interventions2Higgins JPT Green S Cochrane handbook for systematic reviews of interventions. 2011https://handbook-5-1.cochrane.org/Date accessed: February 7, 2019Google Scholar (appendix), which provides guidance on summary assessment across domains within studies, we found that only one (1%) of the 89 included studies had low risk of bias, 18 (20%) had unclear risk, and 70 (79%) had high risk of bias. Second, Slee and colleagues did not evaluate the certainty of the evidence they used for their analysis. Using the Grading of Recommendations Assessment, Development, and Evaluation approach,3Schünemann H Brożek J Guyatt G Oxman A GRADE handbook for grading quality of evidence and strength of recommendations. 2013https://gdt.gradepro.org/app/handbook/handbook.htmlDate accessed: February 7, 2019Google Scholar the certainty of the evidence from the placebo comparisons should arguably be downgraded to “very low” because of high risk of bias, together with potential publication bias4Roest AM de Jonge P Williams CD de Vries YA Schoevers RA Turner EH Reporting bias in clinical trials investigating the efficacy of second-generation antidepressants in the treatment of anxiety disorders: a report of 2 meta-analyses.JAMA Psychiatry. 2015; 72: 500-510Crossref PubMed Scopus (73) Google Scholar and indirectness resulting from short trial duration and unrepresentative study populations.5Hoertel N Le Strat Y Blanco C Lavaud P Dubertret C Generalizability of clinical trial results for generalized anxiety disorder to community samples.Depress Anxiety. 2012; 29: 614-620Crossref PubMed Scopus (57) Google Scholar Third, the authors did not discuss the clinical relevance of their effect estimates. The largest effect for any of the included drugs compared with placebo was a mean difference of 3·6 points (95% credible interval 2·39–4·83) on the Hamilton Anxiety scale (range 0–56 points). This effect estimate might be inflated because of biases4Roest AM de Jonge P Williams CD de Vries YA Schoevers RA Turner EH Reporting bias in clinical trials investigating the efficacy of second-generation antidepressants in the treatment of anxiety disorders: a report of 2 meta-analyses.JAMA Psychiatry. 2015; 72: 500-510Crossref PubMed Scopus (73) Google Scholar and might not be a clinically important difference. Finally, by not including any harms outcomes, Slee and colleagues provided no basis for assessing if benefits outweigh harms for the pharmacological treatment of anxiety. We declare no competing interests. Download .pdf (.09 MB) Help with pdf files Supplementary appendix Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysisTo our knowledge, this is the largest contemporary review of pharmacological agents for the treatment of generalised anxiety disorder by use of network analysis. There are several effective treatment choices for generalised anxiety disorder across classes of medication. The failure of initial pharmacological therapy might not be a reason to abandon a pharmacological treatment strategy. Full-Text PDF Pharmacological treatments for generalised anxiety disorderThe network meta-analysis by April Slee and colleagues1 studied the effect of 22 drugs for the treatment of generalised anxiety disorder and showed that several drugs were effective and well tolerated. We, however, consider that the inclusion of 16 randomised controlled trials from China in this network meta-analysis is highly questionable. Only one2 of these trials was registered on a clinical trial registry platform and provided evidence of ethical approval. Ten trials were published by Chinese predatory journals; these journals tend to accept articles with no or inappropriate peer review, meaning that any study can be published as long as the publication fee is paid. Full-Text PDF Pharmacological treatments for generalised anxiety disorder – Authors' replyKlaus Munkholm and colleagues raise several interesting issues; however, we disagree that our results1 have a very low level of certainty. Our protocol was published a priori2 and we went to extensive and exhaustive lengths to identify many completed trials, providing the most comprehensive summary of evidence to date using robust methods. More than half of the trials in our analysis were done for regulatory purposes and were held to high standards of study conduct. A 2018 network meta-analysis for depression3 also found high risk of bias according to the scale recommended by the Cochrane Handbook for Systematic Reviews of Interventions;4 in fact, none of those studies achieved a low risk of bias across all categories. Full-Text PDF