QSIG Midday Talk: Using prediction models in mental health care

Colourful picture of squares with different coloured heads in

This presentation will be by PhD student Yanakan Logeswaran. Below is his description of the presentation and biography. We will send the joining link to our mailing list on March 31st and again on the morning of the 13th.

In recent years, clinical prediction models (CPMs) have emerged as an avenue for risk stratification and the provision of highly individualised care, through careful consideration of differences across individuals’ characteristics, genes, environments, and lifestyles. In psychiatry, whilst many CPMs have shown promise in improving care, few have actualised their potential and been implemented for routine usage in clinical practice as in certain branches of medicine (e.g., cardiovascular disease and cancer). This lack of translational success is in part due to the insufficient consideration of implementation research in precision psychiatry, and also highlights the importance of discussing the development and use of CPMs with relevant stakeholders to improve their utility and relevance to real-world needs in healthcare. To further our understanding of patient and clinical perspectives on prediction models, one of my PhD studies aims to explore through qualitative interviews the views of service users (who have received a diagnosis of a severe mental illness) and clinicians on the use of clinical prediction models in mental health care.

I plan to present the protocol for this qualitative study and to share some insights from pilot interviews, and would appreciate any input with regard to my analytical approach (and otherwise!). 

I am a third-year PhD student at the Biostatistics and Psychosis Studies Department at the IoPPN. My PhD is focussed on the early detection and prevention of psychosis and other severe mental disorders, particularly through developing and using prediction modelling methods to better identify at-risk individuals and to predict their clinical outcomes.