Extracting hospital data to inform post-discharge community care in COPD patients: A feasibility study

Conference:

SAPC ASM 2019 - Exeter

Talk Code:

P2.36

Presenter:

Maxine Whelan

Co-authors:

Xiaorong Ding; David Clifton; Peter Watkinson; Mona Bafadhel; Beth Lawson; Carlos Areia; Lionel Tarassenko; Andrew Farmer

Author institutions:

Nuffield Department of Primary Care Health Sciences University of Oxford, Institute of Biomedical Engineering University of Oxford, Nuffield Department of Clinical Neurosciences University of Oxford, Nuffield Department of Medicine and Oxford Respiratory NIHR BRC University of Oxford

Problem

Patients with chronic obstructive pulmonary disease (COPD) can experience health deterioration in the community and often attend hospital as a result. More than 4,000 COPD emergency admissions were recorded in 2013-2017 at the Oxford University Hospitals (OUH) NHS Foundation Trust with nearly 1 in 4 patients re-admitted within three months. A wealth of data is captured in the hospital setting but external access to this data is not always possible, despite potential to improve delivery of primary care. Furthermore, it is currently unclear how feasible extraction and alignment of the hospital records and personal collected data are in this patient population. The aim of the sElf-management anD support proGrammE (EDGE2) study is to establish the feasibility of aligning hospital data with community-acquired data for risk-stratification and predicting adverse events in primary care.

Approach

Up to two hundred patients with COPD aged ≥40 years will be recruited while hospitalised with an exacerbation. Participants will be given a tablet computer, pulse oximeter and physical activity monitor (collectively the EDGE2 system) to use for six months in the community after discharge. Participants will be asked to complete daily COPD symptom and monthly mood questions, to monitor vital signs such as oxygen saturation and heart rate, and to wear the physical activity monitor. Routine collection of vital signs data is now collected at scale using the System for Electronic Notification and Documentation system in OUH and can be extracted from the OUH Research Data Warehouse. Other hospital data will include blood gases and ventilation use during index admission, major comorbidities, preceding hospital admissions and emergency department attendances; each extracted from the electronic patient record, patient administration system and the intensive care database. Re-admissions and any emergency department attendance will be identified in the 6 months after discharge. Consent will be obtained prior to the extraction of any hospital data or the provision of the EDGE2 system.

Findings

This abstract describes the protocol to link hospital data to EDGE2 community-acquired data via mobile application and wearable device. Patient outcomes will be assessed six months after discharge. Feasibility of integrating data and refinement of predictive algorithms are the primary outputs. Recruitment for EDGE2 begins in February 2019.

Consequences

Access to routinely collected hospital data in addition to self-reported COPD symptoms and objectively measured physiological and physical activity data may together provide a better overview of a patient’s condition than current clinical practice. EDGE2 will inform the development of predictive algorithms to identify when patients may be experiencing deterioration in their health, utilising both hospital and primary care data which could be incorporated to provide tailored alerts to patients and clinicians, to encourage timely care intervention in the community.

Submitted by:

Maxine Whelan

Funding acknowledgement:

This work will be funded by NIHR Oxford Biomedical Research Centre (Technology and Digital Health sub-stream) and the Engineering and Physical Sciences Research Council (EPSRC).