Research | MINDSET

My research examines how complex real-world data can be used to understand mental health, inequalities, and care across the life course. I work at the intersection of psychiatric epidemiology, causal inference, population data science, and health services research, with a particular interest in questions that cannot be answered adequately using any single dataset or methodological approach.

Across the MINDSET programme, I combine linked administrative records, electronic health records, longitudinal and environmental data, clinical free text, and intensive measures of lived experience. Alongside substantive epidemiological questions, I am interested in the methodological and governance challenges involved in drawing meaningful inference from data that were often collected for purposes other than research.

My current work is organised around four connected areas.

Mental health trajectories and inequalities

Severe mental illness develops and unfolds within social, relational, environmental, and institutional contexts. I am interested in understanding how these experiences combine over time to influence the onset, course, and consequences of mental ill health, and why outcomes differ between people and populations.

I lead WISDOM, a UKRI Mental Health Platform-funded research programme examining the social and emotional drivers of severe and enduring mental health needs. The programme brings together expertise and data resources across the Mental Health Platform, including work spanning social health, complex emotions, mental health data science, and other complementary perspectives.

This work is developing approaches that integrate evidence across different scales and forms of data, including linked population data, electronic health records, environmental exposures, longitudinal cohorts, and intensive longitudinal methods. A central aim is to understand not only which factors are associated with poor mental health outcomes, but how combinations of experiences and exposures may shape trajectories of vulnerability, resilience, service contact, and recovery.

Related interests include:

social and environmental determinants of severe mental illness
inequalities in access to and outcomes of mental health care
longitudinal and life-course approaches to psychiatric epidemiology
ecological momentary assessment and intensive longitudinal data
spatial and environmental epidemiology
integration of evidence across heterogeneous data sources

Administrative data and neurodivergence

Linked administrative data provide opportunities to study health, education, and social outcomes across entire populations, including groups who are often underrepresented in conventional research.

Through an Administrative Data Research UK Fellowship, I lead research using the ECHILD database, which links national health and education records for children and young people in England.

My work examines inequalities in the experiences and outcomes of children with neurodivergent special educational needs, including patterns of school attendance, exclusion, alternative provision, educational participation, and health-care use. I am particularly interested in how individual characteristics interact with schools, local systems, and place to shape these outcomes.

This programme also provides a setting for developing and applying methods for analysing large, hierarchical administrative datasets, including:

longitudinal population-based cohort designs
multilevel and fixed-effects approaches
spatial analysis
analysis of repeated and correlated outcomes
linked health and education records
transparent and reproducible research in secure data environments

More broadly, this work asks how administrative data can be used to identify inequalities without reducing people’s experiences to the categories through which public systems record them.

Addiction treatment, services, and data

A longstanding strand of my research concerns substance use, addiction treatment, and inequalities in access to care.

My earlier work examined opioid treatment systems, access to evidence-based treatment, substance use inequalities, and variation in treatment provision across populations and places. My current work increasingly focuses on the data infrastructure through which addiction treatment is understood, evaluated, and improved.

I am part of the leadership team for UNITED (Using a National, Interdisciplinary Team to Enhance Drug and alcohol treatment data), an NIHR Team Science programme bringing together researchers, treatment providers, people with lived experience, and other partners.

I lead work examining the use of England’s National Drug Treatment Monitoring System (NDTMS) for research and the development of principles for good practice in the use of treatment data. This includes consideration of how routinely collected treatment information can better support epidemiological research, service improvement, policy, and clinical practice while remaining attentive to data quality, interpretation, governance, and the experiences of people represented within those data.

My broader interests in this area include:

access to and inequalities in addiction treatment
treatment pathways and outcomes
co-occurring substance use and severe mental illness
routinely collected addiction treatment data
linkage between treatment, health, mortality, and other administrative records
participatory and interdisciplinary approaches to improving treatment data

Clinical data and research infrastructure

Electronic health records contain rich longitudinal information about mental health and clinical care, but much of their value is distributed across structured records, free text, organisational systems, and local practices.

Since 2019, I have worked with the North London NHS Foundation Trust Research Database, and I currently serve as its Deputy Lead. The database enables ethically governed research using routinely collected mental-health records and forms part of the research partnership between the Trust and UCL.

My role spans research development, data governance, methodological support, and collaboration with researchers, clinicians, informatics teams, public contributors, and other NHS research databases. I also work with colleagues across the UK to improve approaches to the use and accessibility of electronic mental-health records for research.

Methodologically, I am interested in extracting information that is poorly captured in conventional structured fields. This includes work using natural language processing and clinical free text, as well as emerging approaches using large language models within appropriately governed research environments.

This strand of the programme therefore combines substantive psychiatric research with research infrastructure: not only asking questions using clinical data, but helping to create the conditions in which high-quality research using those data can take place.

Connecting the programme

These areas are deliberately connected rather than separate programmes of work.

Across neurodivergence, severe mental illness, and addiction, many of the same challenges recur: important experiences are fragmented across systems; populations are represented differently depending on who collected the data and why; social and institutional contexts shape outcomes; and analytical convenience can easily be mistaken for meaningful inference.

My aim is therefore to develop a research programme in which methods, infrastructure, substantive epidemiology, and engagement with the people represented in data inform one another.

This includes a commitment to:

interdisciplinary and team-based research
methodological rigour and appropriate causal inference
open and reproducible research
responsible use of sensitive and routinely collected data
meaningful public and lived-experience involvement
developing researchers and collaborative research capacity