Fast tracking informative clinical trials: lessons for mental health

The COVID-19 pandemic has put clinical research and how it is organised, implemented, and disseminated under scrutiny, and has revealed not only capabilities, but also shortcomings and inefficiencies. Many of these observations reflected or magnified problems that preceded the pandemic in many areas of medicine and are likely to remain relevant in the years that follow. For example, it has been highlighted that more than 95% of clinical trials on COVID-19 were underpowered or poorly designed and thus had no possibility of providing meaningful evidence.¹ This problem sounds familiar to anybody observing the clinical trials landscape in psychiatry.

Overcoming such issues in pandemic preparedness has been a key aspect of policy initiatives such as the World Health Assembly's Resolution to Strengthen Clinical Trials and the Independent Pandemic Preparedness Secretariat's 100 Days Mission. We argue that similar efforts to improve the clinical research landscape are needed for non-communicable diseases in general and mental health in particular.

Although the mental health crisis might not dominate the headlines in the same way that a novel viral infection would, the silent pandemics of depression and other mental health conditions pose a major threat to health and wellbeing globally.² Furthermore, the limitations of our evidence base are in many ways reminiscent of the methodological shortcomings observed during the COVID-19 pandemic. Indeed, clinical trials in psychiatry often fail to inform clinical practice,³ and systematic analyses of clinical trials for psychological and pharmacological mental health interventions showed that most trials are dramatically underpowered to detect the effect sizes that one could reasonably expect from them.⁴

Big wins (ie, new interventions with large effects) seem unlikely for many health conditions, including mental disorders.⁵ An intervention with a modest benefit on a common cause of disease would have a much greater public health impact than an intervention with a large impact on a rare cause. Given the high prevalence of mental disorders and the enormous disease burden,² socioeconomic costs,⁶ and effects on morbidity and mortality,⁷ even moderate to small treatment effects would be highly relevant. Thus, we have to ensure that trials are sufficiently powered to reliably detect such modest, but relevant, treatment effects.

Here, the mental health field can learn important lessons from the few exceptional trials that rapidly transformed the evidence base during the COVID-19 pandemic, such as the RECOVERY trial done in the UK. This trial was pragmatic in nature (eg, it had broad inclusion criteria and was open label), fully embedded in routine care (ie, it imposed a minimal additional burden for front-line health-care staff and patients), and sufficiently large to detect modest but clinically relevant effects. The trial produced the first guideline-changing result within 100 days of the protocol first being drafted,⁸ and it has since provided answers for ten additional treatment options, definitively showing what works and, equally important, what does not.⁹

Of course, one cannot translate all aspects of trials such as the RECOVERY trial⁸ directly into randomised controlled trials (RCTs) in mental health. However, trials such as RECOVERY have shown practical ways in which to produce more efficiently and at higher speed clinically meaningful results that have immediate implications for the standard of care worldwide.

Driven partly by the lessons learnt from the pandemic response, the Good Clinical Trials Collaborative, with support from the Wellcome Trust and the Bill & Melinda Gates Foundation, brought together and drew on the expertise and experiences of a diverse, multidisciplinary, global community, to identify and describe five fundamental principles that are required to deliver a good RCT; the full guidance is now available online. The principles, taken together, capture the necessary qualities of a well planned, well run, and clinically relevant RCT. The guidance recognises that the methods and approaches needed to achieve these qualities will differ in small or large ways from trial to trial, but that their validity is universal. These principles can be readily applied to mental health to make sure that future trials do not repeat the errors of the past and instead become increasingly informative (by making sure that each trial answers definitively an important question) and more relevant for clinical practice. In future, it will be important to elaborate what these fundamental principles of informative and relevant clinical trials mean for the mental health field with regard to, for example, proportionate risk management in clinical trials (in comparison with routine care) or appropriate inclusion and exclusion criteria (eg, with regard to comorbidities or suicidality). To achieve this improvement, the mental health field will need a collaborative conversation between people with lived experience, clinicians, researchers, ethics commissions, research funders, industry, regulators, and other key stakeholders.

SMG reports honoraria from Hexal and STREAMED UP and grant funding from the European Commission, German Federal Ministry of Health, German Research Foundation, and National Multiple Sclerosis Society, USA. MJL reports grant support from Bill & Melinda Gates Foundation, Wellcome Trust, Novartis, Boehringer Ingelheim, Merck, Sanofi, Regeneron, Moderna, Schmidt Futures, Google Ventures, Flu Lab, National Health Service England, UK National Institute for Health and Care Research, and UK Medical Research Council; an unpaid advisorship for the European Society of Cardiology; and receipt of drugs for clinical trials from Roche, AbbVie, Regeneron, GlaxoSmithKline, Vir Biotechnology, and Boehringer Ingelheim. NM reports grant support from Bill & Melinda Gates Foundation and Wellcome Trust and unpaid board membership for Cystic Fibrosis Europe. CO reports grant funding from the European Commission, German Federal Ministry of Research and Education, German Research Foundation, and Berlin Institute of Health; consulting fees from Janssen, Peak Profile, LIMES Schlosskliniken; honoraria from Janssen, Neuraxpharm, Fortbildungskolleg, and LIMES Schlosskliniken; participation in the Data and Safety Monitoring Board for the Central Institute of Mental Health Mannheim; and unpaid consultancy for the German National Guidelines for Depression.