Article Text

Download PDFPDF
3  Visualizing the evolution of evidence: cumulative network meta-analyses of new generation antidepressants in the last 40 years
  1. Yan Luo1,
  2. Anna Chaimani2,
  3. Toshi A Furukawa1,
  4. Yuki Kataoka3,
  5. Yusuke Ogawa4,
  6. Andrea Cipriani5,
  7. Georgia Salanti6
  1. 1Department of Health Promotion and Human Behavior, School of Public Health in the Graduate School of Medicine, Kyoto University, Kyoto, Japan
  2. 2Université de Paris, Research Center of Epidemiology and Statistics (CRESS-UMR1153), INSERM, INRA, Paris, France
  3. 3Hospital Care Research Unit, Hyogo Prefectural Amagasaki General Medical Center, Amagasaki, Japan
  4. 4Department of Healthcare Epidemiology, School of Public Health in the Graduate School of Medicine, Kyoto University, Kyoto, Japan
  5. 5Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford Health NHS Foundation Trust, Oxford, UK
  6. 6Institute of Social and Preventive Medicine, University of Bern, Bern, Switzerland


Background Meta-analysis and network meta-analysis (NMA) in particular is rapidly becoming the standard form of evidence synthesis. It is exceedingly important to present such evidence in a timely, clinically relevant and comprehensible manner for patients, clinicians and policy makers. However, it is often challenging because the evidence synthesized in NMAs is often voluminous and can be rapidly changing.

Objectives We propose a way to help evidence end users capture and understand the dynamic evolution of evidence through interactive visualization. Methods: We built a Shiny web application that performs and presents cumulative NMAs based on R netmeta. We used a comprehensive dataset of both published and unpublished double-blind randomized controlled trials (RCTs) of 21 antidepressants in the acute treatment of major depression. For the purpose of increasing clinical relevancy, we set two primary outcomes, which were efficacy (treatment response) and acceptability (all-cause discontinuation), and treatment effects were summarized via odds ratios. We also evaluated the confidence in evidence using the CINeMA (Confidence in Network Meta-Analysis) framework for a series of consecutive NMAs (1990, 1995, 2000, 2005, 2010 and 2016). Users can change several conditions for the analysis: the period of synthesis (from 1978 to 2016), the publication status of RCTs (only published data or full data), and the RCT type (head-to-head trials only or including placebo-controlled trials). The Shiny app displays the results of any user-selected NMAs, including network plots, two-dimensional plots and league tables combining efficacy and acceptability for each drug, forest plots, funnel plots, among others. The league tables and two-dimensional plots that combine not only efficacy and acceptability but also the level of confidence in the evidence are also presented for NMAs which have been assessed of confidence, in order to demonstrate the importance of credibility in the evidence in decision-making.

Results The dataset includes 190 head-to-head trials and 270 placebo-controlled trials. The web app is accessible at It shows that over the years the drugs with relative superiority changed dramatically; but as the evidence network grew larger and better connected, the overall confidence improved. It reveals that newly approved drugs often showed initially exaggerated results, which tended to diminish and stabilize after approximately a decade. It also highlights the potential difference between head-to-head trials and placebo-controlled trial, and the magnitude of publication bias in antidepressant trials, which have been extensively addressed in other studies.

Conclusions Our interactive visualization via Shiny in R helps to present the voluminous evidence in a succinct but extensive way, which allows the users to capture the dramatic changes and special features in evidence evolution, and may help decision making. It could be considered as a tool to improve accessibility of evidence in the future.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.