Bookmarks
Diabetes In-Patient Hypoglycaemia Prediction Model
Safe People
Organisation name
Oxford University Hospitals NHS Foundation Trust
Organisation sector
Government Agency (Health and Adult Social Care)
Applicant name(s)
Rustam Rae
Funders/ Sponsors
Neil Hill
DEA accredited researcher?
Unknown
Sub-licence arrangements (if any)?
No
Safe Projects
Project ID
NIBDAPC_2021_0001
Lay summary
We wish to use anonymised patient data to confirm the efficacy of a model that can predict people at risk of hypoglycaemia in during their hospital admission. If this works it may be possible to use this model in real-time to identify individuals at risk and take pre-emptive steps to prevent or mitigate the risk of hypoglycaemia.
Public benefit statement
Between May and October 2018, 525 hypoglycaemic harms were recorded amongst participating NHS Trusts in the The National Diabetes Audit. As well as these harms being linked to poorer outcomes, hypoglycaemia requires significant medical and nursing resources to manage and the cost of an inpatient hospital stay is increased by 40% for those exposed. For example, if the number of hypoglycaemic episodes in people with diabetes and acute stroke were halved, this could lead to savings of more than £6,000,000. Current means of detecting blood glucose levels in hospitalised patients are inadequate. Hypoglycaemia is recognised only if patients are symptomatic and are able to communicate this to healthcare professionals. This is clearly not feasible for large numbers of patients (including those with reduced level of consciousness, difficulty communicating, or cognitive impairment). In addition, around 30% people with diabetes have impaired awareness of hypoglycaemia. Healthcare professionals establish blood glucose levels by checking capillary blood glucose (CBG) levels. This method is sensitive but its efficacy in preventing harm from hypoglycaemia is a function of how often it is performed. Approximately 20% of hospital beds in the UK are occupied by people with diabetes, and such a large amount of bedside CBG testing is unfeasible for most hospitals. Changes in nursing staffing levels on general wards may also impact on capacity to undertake regular CBG monitoring. This means the risk of hypoglycaemia is set to rise. Mitigating this increased risk is essential to prevent harm.
Request category type
Public Health Research
Other approval committees
Project start date
25/06/2021
Latest approval date
25/06/2021
Safe Data
Dataset(s) name
NIHR HIC Diabetes Dataset
Data sensitivity level
De-Personalised
Common Law Duty of Confidentiality
Not applicable
National data opt-out applied?
Not applicable
Request frequency
One-off
Release/Access date
13/08/2021
Safe Setting
Access type
TRE
Safe Outputs
Link to research outputs