Completed Lungs & Breathing NIHR-supported project Public Health & Healthcare

Reducing COPD exacerbation frequency and severity through early and accurate diagnosis and risk stratification using integrated primary and secondary care data

In plain English

AI plain-English summary

Every year, tens of thousands of people with COPD are rushed to hospital with a sudden, severe flare-up that could have been prevented with earlier warning. The problem is that current risk tools are blunt—they miss patients whose condition is quietly deteriorating until it is too late. This project will build smarter risk-stratification models by linking GP records with hospital data and applying natural language processing to spot subtle patterns in clinical notes that conventional coding overlooks. The researchers will also evaluate whether integrated Respiratory Hubs—centralised clinics that coordinate care across primary and secondary services—can reduce exacerbation frequency and severity. If successful, the work could give GPs a practical, data-driven tool to identify rising-risk patients months before a crisis, allowing targeted interventions such as medication adjustments or pulmonary rehabilitation. The project will also produce a blueprint for joining up COPD data pathways across the NHS, supporting the national Respiratory Data Strategy. This is applied health services research with a direct route to policy and practice, not fundamental science.

View original technical description
The aim of this study is to inform regional and national practitioners and policy makers about the benefits of integrated Respiratory Hubs for COPD care and to explore advanced AI approaches to risk stratification of COPD patients in primary care, by Identify and curate clinical datasets from provider service’s EHRs across the pathway for COPD care; Develop harmonised coding set/key terms for COPD care- diagnosis, treatment, and outcomes including future risk profile; Analyse factors which predict poor clinical outcome (future exacerbation risk, ED attendance and hospital admission) to create risk stratification models for targeted clinical interventions; Develop novel stratification approaches to identify rising risk COPD patients in primary care using Natural Language Processing Models (NLP) and ML validated against conventional coding and hospital outcomes and Provide a blueprint for integrating COPD pathway data to inform national rollout of the HDRUK support Respiratory Data Strategy.

Researchers

Tom Wilkinson (Principal Investigator)

Related Research

Grants with similar aims, by meaning.

Novel Integrated Health and Social Care Data Enabling Earlier Detection of COPD and Intervention Delivery Through Optimised Care Pathways
What happens between first symptoms and first acute exacerbation of COPD? Mapping and prediction study
my Smart COPD exacerbation management (mySmartCOPD)
Understanding variations in outcome in asthma and COPD: use of routine clinical data
A digitally integrated clinical platform with self-management and machine-learned risk-predictive algorithms for COPD

Original classification

Data, Health and Society (DHS)

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.