Heterogeneity of treatment outcomes for psychiatric patients

Yan Feng (primary)
Pragmatic Clinical Trial Unit
Queen Mary University of London
Stefan Priebe (secondary)
Unit for Social and Community Psychiatry
Queen Mary University of London


Progress in predicting treatment outcomes – despite extensive research – has been disappointing as standard analyses do not reveal predictors of any relevance. The past two decades has seen rapid development of Machine Learning (ML) methods for improving the prediction of treatment outcomes. In this project, student will analyse large observational data and apply ML methods to develop models for identifying predictors, which could be used to predict likely size of benefit for each patient before treatment. This is one of the first studies to apply ML to predict outcomes of complex interventions in psychiatric care.


Feng Y, Devlin N, Bateman A, Zamora B, Parkin D (2019). The Distribution of the EQ-5D-5L Profiles and Values in Three Patient Groups. Value in Health. 22(3): 355-361.

Priebe S, Kelley L, Omer S, Golden E, Walsh S, Khanom H, Kingdon D, Rutterford C, McCrone P, McCabe R (2015). The Effectiveness of a Patient-Centred Assessment with a Solution-Focused Approach (DIALOG+) for Patients with Psychosis: A Pragmatic Cluster-Randomised Controlled Trial in Community Care. Psychother Psychosom. 84(5): 304-313.

Schlackow I, Kent S, Herrington W, Emberson J, Haynes R, Reith C, Wanner C, Fellström B, Gray A, Landray MJ, Baigent C, Mihaylova B on behalf of the SHARP Collaborative Group (2017). A policy model of cardiovascular disease in moderate-to-advanced chronic kidney disease. Heart. 03:1880-1890.

Shatte ABR, Hutchinson DM, Teague SJ (2019). Machine learning in mental health: a scoping review of methods and applications. Psychological Medicine. 49:1426–1448.

Genes, development and STEM* approaches to biology
Area of Biology
Techniques & Approaches
Mathematics / StatisticsSimulation / Modelling