Hypoglycaemia and serious adverse events in older people living with diabetes and dementia – a population-based cohort study

Application Number
Lay Summary

Diabetes and dementia are common illnesses that can occur together in older people. Diabetes UK estimate that approximately 5 million people will have diabetes in the next 10 years, whereas Alzheimer’s UK estimate that over 1 million people will have dementia. Currently, 1 in 5 patients with dementia also have diabetes.

Medication to control blood sugar can provoke low blood sugars (hypoglycaemia; a particularly serious side-effect that may cause serious long-term harm). Although dementia is significantly associated with hypoglycaemia, there is currently little evidence regarding long-term consequences of hypoglycaemia in patients with diabetes and dementia. We will analyse healthcare data from the Clinical Practice Research Datalink to determine serious complications associated with hypoglycaemia (falls or fractures, use of emergency healthcare and hospitalization, heart attacks, and death) in patients with diabetes and dementia compared to those with only diabetes.

Understanding the potentially serious consequences of hypoglycaemia is crucial in helping patients, carers and doctors make decisions on choice, intensity and monitoring of medication for diabetes and concomitant dementia.

The results will provide valuable evidence for national guidance and co-ordination of health and social care policy (e.g. provision of meals at the appropriate time) for vulnerable patients with diabetes and dementia.

Technical Summary

Background: Treatment of diabetes in older people with dementia is challenging, as clinicians try to achieve a comfortable balance between the pursuit of tight blood sugar control (in accordance with national targets), against pragmatism and avoidance of serious side-effects, such as hypoglycaemia. There is a paucity of evidence regarding risk of hypoglycaemia in patients with diabetes and dementia, and the relationship between hypoglycaemia and serious adverse events (falls or fractures, use of emergency healthcare and hospitalization, heart attacks, and death).
Objectives: To describe serious (medically recorded) hypoglycaemic events in older people with multimorbidity (diabetes and dementia), and to determine risk of associated serious adverse events after serious (medically recorded) hypoglycaemia (falls or fractures, use of emergency healthcare and hospitalization, heart attacks, and death). For ease of reference, all mention of hypoglycaemia is intended to mean serious (medically recorded) hypoglycaemia.
Methods Population-based cohort of patients with diabetes (± dementia) based on CPRD with linkage to Hospital episode statistics, and Office for National Statistics datasets.
Data Analysis Kaplan-Meier survival curves will be used to display the survival curve for each adverse outcome after incident exposure to a hypoglycaemic event. The association between hypoglycaemic exposure and serious subsequent adverse event, will be evaluated using Cox proportional hazard regression models with adjustment for appropriate confounders to estimate a Hazard Ratio and 95% confidence interval.

Health Outcomes to be Measured

The primary outcome will be falls and/or fractures. Secondary outcomes will be emergency healthcare and hospitalization, cardiovascular events (acute coronary syndrome, stroke), and overall mortality..


Yoon Loke - Chief Investigator - University of East Anglia
Katharina Mattishent - Corresponding Applicant - University of East Anglia
George Savva - Collaborator - University of East Anglia
Kathryn Richardson - Collaborator - University of East Anglia


HES Accident and Emergency;HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation