Psychiatric epidemiologist · Administrative data researcher · Applied data scientist

Website · ORCID · UCL Division of Psychiatry

I’m a Senior Research Fellow at UCL, working at the intersection of psychiatric epidemiology, population data science and mental health services research.

My research uses linked administrative data, electronic health records and longitudinal population data to understand mental health, substance use, inequalities and interactions with health, education and other public services. I am particularly interested in how routinely collected data can be used well: methodologically, computationally and responsibly.

• Psychiatric and social epidemiology
• Mental health and substance use
• Linked administrative and electronic health-record data
• Causal and longitudinal methods
• Natural language processing and large language models for clinical text
• Research data infrastructure, governance and reproducible analysis
• Evaluation and improvement of routinely collected public-service data

Most of my GitHub work falls into a few overlapping areas:

Research code Reproducible workflows for data management, statistical analysis, visualisation and research reporting, primarily in R, SQL and Python.

Research infrastructure Tools and configuration for making analytical workflows more robust, maintainable and portable, including work designed for secure and air-gapped research environments.

Open research Code, documentation and supporting materials associated with publications, methods development and collaborative research projects where these can be shared publicly.

Web and automation A smaller collection of academic web, self-hosting and automation projects reflecting my interest in useful technical infrastructure beyond individual analyses.

My current work includes:

• UKRI Mental Health Platform, developing cross-cohort and electronic health-record research on severe and complex mental health problems.
• UNITED, an NIHR Team Science project examining how drug and alcohol treatment data in England can be strengthened for research, policy and practice.
• Research using linked administrative data to study mental health, education, social circumstances and public-service outcomes.

• Reproducible Data Analysis for Mental Health Research — Carpentries-based R and Python workshop materials for mental health researchers.
• Mapping and Spatial Analysis in R — introductory spatial-analysis teaching materials developed for the UCL R User Group.
• Risk of Suicide After Cancer Diagnosis in England — worked teaching examples using routinely collected health data in R, Python and Stata.
• MINDSET academic website — source, data pipelines and validation infrastructure for my academic research website.

I try to make research code transparent, reproducible and maintainable, while recognising the constraints of real-world administrative and clinical data. That means documenting analytical decisions, separating configuration from logic where practical, preserving provenance, and avoiding methodological sophistication for its own sake.

Much of my research necessarily takes place in secure environments and cannot be published directly. Public repositories therefore represent only part of the underlying work.

For publications, projects and more about my research, see justinyang.me.