Examining treatment patterns of antidepressant and antipsychotic prescriptions; and serious adverse events and the associated economic cost of antipsychotic drug use in people with dementia

Date of ISAC Approval: 
07/08/2018
Lay Summary: 
People with dementia sometimes experience associated behavioural and psychological symptoms (BPSD). These are sometimes treated with antipsychotic drugs (APDs) and antidepressants (ADDs) although little is known about how long they are used for, and if treatment is discontinued whether they are reintroduced soon afterwards or alternative drugs are used. We also don't know if certain factors can predict if someone will be prescribed an APD or ADD, for example if they are male or female. It is important to know this to help us understand if these medications are used in an appropriate way. Some research has suggested that people with dementia who use APDs might have a higher risk of stroke, than people who do not use APDs. We are going to look at whether this is the case, and also look at the risk of death and other conditions including heart problems, infection and fractures. For patients that experience stroke, we will determine how much this costs in relation to their hospital admissions and GP appointments which will give us an indication of how much could be saved if stroke was prevented.
Technical Summary: 
Our aims are to explore the treatment patterns for medications prescribed for the Behavioural and Psychological Disorders (BPD) associated with dementia and to examine adverse outcomes associated with antipsychotic use (APD). We will delineate a prevalent cohort of patients with a diagnosis of dementia, and examine medication utilisation including APD dose, persistence and switching to alternative treatments, and antidepressant use (ADD). We will also determine factors which influence this use, using regression analysis. From the dementia cohort, we will use stratified cox-proportional hazards models to estimate hazard ratios (HR) for each of the specified outcomes including death, stroke, cardiovascular disease, venous thromboembolism, infection and fracture; for APD use versus non-use. For stroke outcomes, we will additionally use the findings to inform the parameters of an economic model, estimate the costs associated with this outcome in APD use and not use; and map primary and secondary health care utilisation with associated costs, related to this outcome.
Health Outcomes to be Measured: 
- mortality - myocardial infarction - Congestive heart failure - stroke or transient Ischaemic attack - Pulmonary embolism or Venous thromboembolism - hip fracture - Serious infection - secondary care utilisation - cardiac Arrhythmia - primary care utilisation
Collaborators: 

Professor Darren M Ashcroft - Chief Investigator - University of Manchester
Bruce Guthrie - Collaborator - University of Dundee
Georgios Gkountouras - Collaborator - University of Manchester
Dr Hayley Gorton - Collaborator - Not from an Organisation
Dr Jill Stocks - Collaborator - University of Manchester
Dr Matthew Carr - Collaborator - University of Manchester
Dr Pearl Mok - Corresponding Applicant - University of Manchester
Professor Rachel Elliott - Collaborator - University of Manchester
Professor Tjeerd van Staa - Collaborator - University of Manchester
Tommy Allen - Collaborator - University of Manchester
Professor Tony Avery - Collaborator - University of Nottingham

Linkages: 
HES A&E;HES Admitted;HES Outpatient;ONS;Patient IMD;Practice IMD (Standard);Practice IMD (Standard)