
Professor Daniel Stahl
Professor in Medical Statistics and Statistical Learning
Contact details
Biography
Daniel Stahl joined the Institute of Psychiatry, Psychology & Neuroscience (IoPPN), King’s College London in 2006 and currently holds the position of Professor in Medical Statistics and Statistical Learning at the Department of Biostatistics & Health Informatics.
He is Deputy Head of the Department and Deputy Lead of the NIHR Maudsley BRC theme “Trials, Genomics and Predictions”.
Before joining the IoPPN, Daniel worked as a Statistician for the Departments of Developmental Psychology and Primatology at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. Prior to that, he served as a Postdoctoral Research Fellow at the School of Mathematics and Statistics, University of St. Andrews, Scotland (2000-2001) and the Institute of Insect Virology, State Research Centre for Agriculture, Neustadt/Weinstrasse, Germany (1999-2000). He also worked as a Research Scientist in Primate Husbandry at the German Primate Centre in Göttingen in the years 1996-1997.
Daniel obtained a “Diploma-Biology” degree from Eberhardt-Karls-Universität Tübingen, Germany in 1991. In 1998, in collaboration with the German Primate Centre in Göttingen and Emory University in Atlanta, he acquired a PhD in Behavioural Biology from the same university. He then went on to further his education by completing a postgraduate degree in "Biometry for Scientists" (MSc equivalent) at the Mibeg Institute in Tübingen, Germany.
In recent years, Daniel's research interest has specialised in prediction modelling, a cornerstone of 21st-century precision medicine that intersects statistics, health informatics, and clinical research. His group “Precision Medicine and Statistical Learning” is devoted to propelling precision psychiatry through the development of models for risk prediction, therapy response identification, and trials' enrichment.
Key achievements include creating a transdiagnostic risk calculator for early psychosis detection and a model predicting major depression recurrence. Both these tools are under implementation and validation phases and show promise for clinical decision-making.
In 2019, Daniel founded the Maudsley BRC Prediction Modelling Group at the IoPPN, which enhances collaboration among researchers and clinicians, fostering communication and information exchange on precision medicine's predictive modelling applications.
His research also explores identifying predictors, mediators and moderators of treatment success and uses model-based cluster analysis to discover subgroups among psychiatric patients. In addition, he has played a pivotal role as a Senior Trial Statistician in numerous clinical trials.
Finally, his group addresses the translational gap in prediction modelling within psychiatry. By focusing on effective communication strategies and combating bias and stigma, they aim to enhance the acceptance of predictive models among clinicians and service users.
Research interests
- Clinical Prediction Modelling and Precision Psychiatry
- Comorbidity and Prediction Modelling
- Statistical & Machine Learning
- Model Selection
- Mediation and Moderation
- Missing Data in Prediction Modelling
- Cluster Analyses
- Clinical Trials
Teaching
- Co-Programme Lead for the "Applied Statistical Modelling & Health Informatics" MSc/PG Dip/PG Cert programs.
- Module Lead for "Prediction Modelling" and offers courses on "Introduction to Path Analysis and Structural Equation Modelling."
- Provides statistical support to IoPPN students and researchers
- Lead of the UKRI and MRC-funded "Health Data Science" online learning and research environment, an integral part of King's “Innovation Scholar Programme: Big Data Skills Training.”
Over the years, Daniel has taught both introductory and advanced statistics courses to MSc students at the IoPPN and those in the Doctorate of Clinical Psychology programme. These courses span various topics, such as measurement scale development, prediction modelling, R programming, multiple testing and model selection, factor analyses, structural equation modelling, mediation analyses, longitudinal data analyses, and methods for handling missing values.
Expertise and public engagement
The BRC prediction modelling group page serves as a platform to inform the public about their research in clinical prediction modelling. Daniel organises data science and prediction modelling days specifically designed for school students, aiming to inspire and educate them in these fields.
Research

Precision Medicine and Statistical Learning
Precision Medicine & Statistical Learning

The SOUth London Diabetes (SOUL-D) Study
SOUL-D is a prospective cohort study of the association between depression and diabetes outcomes in people with newly diagnosed type 2 diabetes.
Project status: Ongoing

Psychological interventions to improve glycaemic control in type 1 and type 2 diabetes: A systematic review and meta-analysis
A systematic review of psychological interventions in type 1 and 2 diabetes to assess whether their effectiveness in improving glycaemic levels has improved.
Project status: Completed

Virtual Reality-enhanced Cue Exposure Treatment for people with cocaine dependence
Development, evaluation and testing of a Virtual Reality-enhanced Cue Exposure Treatment integrated with a wearable device to address craving, prevent relapse and improve treatment outcomes of people with cocaine dependence.
Research

Precision Medicine and Statistical Learning
Precision Medicine & Statistical Learning

The SOUth London Diabetes (SOUL-D) Study
SOUL-D is a prospective cohort study of the association between depression and diabetes outcomes in people with newly diagnosed type 2 diabetes.
Project status: Ongoing

Psychological interventions to improve glycaemic control in type 1 and type 2 diabetes: A systematic review and meta-analysis
A systematic review of psychological interventions in type 1 and 2 diabetes to assess whether their effectiveness in improving glycaemic levels has improved.
Project status: Completed

Virtual Reality-enhanced Cue Exposure Treatment for people with cocaine dependence
Development, evaluation and testing of a Virtual Reality-enhanced Cue Exposure Treatment integrated with a wearable device to address craving, prevent relapse and improve treatment outcomes of people with cocaine dependence.