28\u2005Reanalysis of the treatment for adolescents with depression study (TADS) under the restoring invisible and abandoned trials initiative (RIAT)

Abstract

Objectives Increasing numbers of children and adolescents are being prescribed antidepressants, despite uncertainty about the benefits and harms. A majority of antidepressants remain contraindicated for use in juveniles because they significantly increase the risk of self-harm and suicidal behaviour. Despite this, dispensing rates have increased in many countries; for example, by 35.5% in Australia in 10–14 year olds between 2009 and 2015 (Karanges et al., 2014). This trend was led largely by guidelines from the United States that asserted the safety and efficacy of fluoxetine for adolescents. Funded by the US National Institutes of Health (NIH), the Treatment for Adolescents with Depression Study (TADS) was an influential trial of fluoxetine and counselling (with over 48 publications and 1000 citations). Its methodology and reporting have been strongly criticized. We will report a reanalysis of the original TADS dataset to assess the validity of its published reports and to investigate any unreported findings. Method The Restoring Invisible and Abandoned Trials (RIAT) initiative provides an important mechanism for accountability in clinical trials. RIAT’s systematic methodology will address selective and misleading outcome reporting in clinical trials through rigorous reanalysis of trial data. We will reanalyse the TADS dataset, reporting on key efficacy and safety outcomes as specified by the original protocol, alongside additional safety outcomes. The TADS study recruited 12–17 year old adolescents with a primary diagnosis of major depressive disorder (MDD) and randomised them to fluoxetine, psychotherapy (cognitive-behavioural therapy; CBT), combined fluoxetine and CBT, or placebo-based intervention. The TADS team reported that fluoxetine was efficacious and safe for use in adolescents and that participants who received combined therapy demonstrated the best health outcomes. We will follow the original TADS protocol’s statistical analysis or use best practice methods where it is silent, conducting regression analyses and ANCOVAs with post-hoc reporting where appropriate. Results We will present results from our reanalysis of the efficacy and safety of fluoxetine and counselling in the TADS study. We will discuss any identified discrepancies in reporting and reproducibility of trial protocol with reference to the public dataset and published methodology. Conclusions TADS informed the subsequent US Food and Drug Administration (FDA) approval for removing any lower age limit on fluoxetine prescriptions, and therefore led to similar practices internationally. Uncritical prescribing to children and adolescents is particularly irresponsible because of short-term risks such as suicidality and long-term risks, many of which are unknown. Our work examines research integrity in TADS, one of the most important drug trials currently underpinning the overdiagnosis and overtreatment of adolescents with depression. Our RIAT-led methodology will clarify TADS findings and discourage uncritical reliance on the original publications. We anticipate our findings will influence clinical guidelines and improve the health and safety of adolescents with depression. Independent reanalysis of influential trials such as TADS supports the safety and efficacy of treatments for psychiatric disorders in vulnerable patient populations. Further, this work advocates for open access to data and constructive criticism and feedback from the independent research community.