DECODE tool
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This software supports the analysis of multiple long-term conditions (MLTCs) using longitudinal healthcare data derived from the Clinical Practice Research Datalink (CPRD) and linked Hospital Episode Statistics (HES). The methodology combines statistical association testing (Fisher’s exact test with false discovery rate correction), odds ratio estimation, and temporal sequencing of diagnoses to identify patterns of condition co-occurrence and progression.The system operates on structured data describing pairs of conditions (e.g. two conditions occurring in the same patient). Prior to analysis, the data are processed through a cleaning and standardisation pipeline that ensures consistent naming of conditions and expands abbreviations (for example, “CKD” to “Chronic Kidney Disease”). The system then calculates condition prevalence and constructs directed networks, where nodes represent conditions and edges represent the most frequent temporal order of diagnosis. Visualisations include network graphs and statistical summaries to support interpretation of condition patterns across age and sex stratifications.The application includes modular components for sensitivity analysis, identification of common condition combinations, and interactive filtering to support tailored exploration of condition trajectories based on user-selected condition profiles. A Streamlit-based interface enables users to generate population-level insights and explore trajectory patterns relevant to selected conditions. The analysis pipeline is implemented in Python and can be executed using the provided scripts and configuration.The study underpinning this software was conducted using anonymised patient data from CPRD, with linked HES datasets, covering the period 2000–2021. Ethical approval was obtained from the CPRD Independent Scientific Advisory Committee (protocol number: 22_001840). Individual-level patient data are not included in this repository and cannot be shared publicly. The software operates on preprocessed, non-identifiable aggregated data structures.Acknowledgments. Data-driven machinE-learning aided stratification and management of multiple long-term COnditions in adults with intellectual disabilitiEs (DECODE) project (NIHR203981) is funded by the NIHR AI for Multiple Long-term Conditions (AIM) Programme. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. This work uses data provided by patients and collected by the NHS as part of their care and support. We also want to acknowledge all data providers who make anonymised data available for research.© the authors
Fields of Study & Subjects
Chronic Disease Management Strategies
Artificial Intelligence in Healthcare
Machine Learning in Healthcare
Publication details
- Journal / source
- Loughborough University Research Repository (Loughborough University)
- Publication date
- 27 March 2026
- Language
- EN