A Tool to Predict Which Antidepressant is Most Effective

Principal investigator: Claire Gillan

Funding Period: 2017-2021

Dr Claire Gillan is creating an internet-based tool which could predict how effective antidepressants will be for different individuals.

The project

Imagine if we could accurately predict which types of antidepressant would be most effective for people living with depression?

Right now, one in ten people experience depression in the UK. Millions turn to their doctor each year – but without precise tools to know which treatment is most suitable, many may be given treatment that doesn’t work for them. For drug treatments, such as antidepressants – it often takes several attempts before finding something that is works.

This trial and error approach means many people may endure months of little improvement and unnecessary side effects. Recovery and wellbeing often suffers as a result.

So, Dr Claire Gillan at Trinity College Dublin will be investigating whether a simple new tool that uses internet-based techniques could provide the information needed to transform treatment of depression. She’s designing a tool that could make sure the right treatments are reaching the right patients – by predicting how well each individual will respond to treatment.

The process

To create this tool, Claire collected data online from people who are new to antidepressants.

1,100 participants shared information about themselves online and took part in online cognitive tests to determine and track their symptoms. Follow-up rates were high, with 907 participants retained at follow-up. Each participant completed up to 400 clinical and cognitive assessments at baseline;

This data enabled Claire to create a machine learning algorithm to estimate how well people with certain characteristics will respond to specific treatments.

The algorithm is now being tested in local clinics with patients who have been diagnosed with depression – and with other conditions like Obsessive Compulsive Disorder (OCD).

By using the internet in this way, Claire will be exploring important new ways to conduct treatment research.

The outcome

Key Findings:
- Internet-based methods can efficiently gather large, high-quality datasets for mental health research, with rapid recruitment and high retention;
- Both iCBT and antidepressant arms showed significant reductions in depression and other psychiatric symptoms over four weeks, with the largest improvements in depression;
- Metacognitive biases (such as confidence in one’s abilities) are malleable and improve alongside reductions in anxious-depression symptoms;
- Engagement with digital CBT was high, and greater engagement was associated with greater improvement in depression;
- Some cognitive correlates of symptoms differ between treatment-seeking individuals and the general population, highlighting the complexity of translating findings into clinical practice.

The project has leveraged over €2 million in further funding for follow-up studies.

Papers:

Gillan CM, Rutledge RB. Smartphones and the Neuroscience of Mental Health. Annu Rev Neurosci. 2021
Lee, Chi & Palacios, Jorge & Richards, Gillan, Claire Et Al. The Precision in Psychiatry (PIP) study: Testing an internet-based methodology for accelerating research in treatment prediction and personalisation. BMC Psychiatry2023

The impact

Clinical and Research Impact: The project has developed and validated a protocol for large-scale, internet-based treatment research in psychiatry, which accelerates the pace and scale of research in this field. The data and methods are expected to help develop algorithms for more precise allocation of treatments in primary care, supporting clinicians in making smarter prescribing decisions;

Broader Influence: The approach and findings are influencing the design of future studies, including the use of smartphone apps and alternative data sources (e.g., social media, sensor data) to reduce participant burden and improve data quality;

Capacity Building: The project has fostered collaborations with industry partners (e.g., SilverCloud Health), led to further major grants, and contributed to the development of new research strands, such as network theory of mental disorders;

Claire Gillan

In January 2017, Dr Claire Gillan will open a new lab and take up a post as assistant professor of Psychology at Trinity College Dublin. She completed a Ph.D. at Cambridge University and gained a postdoctoral fellowship at New York University. She is passionate about understanding and treating mental health problems through research.