People with schizophrenia die younger than people without serious mental illnesses, partly because of physical illnesses such as obesity, diabetes, high blood pressure and high cholesterol all of which increase the chances of a heart-attack or stroke. Schizophrenia is treated with drugs called 'antipsychotics'. Taking more than one antipsychotic over a long time can make these physical illnesses more likely although at present we do not have firm evidence about this. We want to find out whether taking more than one antipsychotic drug increases the chance of a person getting one of these serious physical illnesses. In this study, we would like to use GP practice data to answer this. We will obtain data on a large group of people with schizophrenia who are prescribed one or more than one antipsychotic drug. We will then follow them for up to 15 years to find out whether they are diagnosed with one of these illnesses. We will also try to find out if higher doses or certain combinations of antipsychotics are particularly harmful.
Answering these questions can help reduce:
- Harms caused by drugs
- Risks of serious illnesses in schizophrenia patients
- Costs due to unnecessary and potentially harmful prescriptions
People with schizophrenia are more likely to suffer from serious physical conditions i.e. diabetes, hypertension and hypercholesterolaemia. This study aims to investigate the prospective association between antipsychotic polypharmacy and diabetes, hypertension and hypercholesterolemia and to investigate the mechanism of any association.
Using the Clinical Practice Research Datalink (CPRD), we will conduct a retrospective cohort study of schizophrenia patients who are: registered with a participating GP practice and prescribed 1 antipsychotic medication at study entry.
The outcome is the incidence of diabetes, hypertension and hypercholesterolaemia. This will be defined as having a prescription of an antihypertensive, cholesterol-lowering and/or anti-diabetic drug (primary outcome) or having a prescription and an ICD10 code or a READ code for diabetes, hypertension or hypercholesterolaemia (secondary outcome). We will follow patients up for 15 years or until a diagnosis of or prescription for an outcome, transfer out of the surgery, last record of up-to-standard contact, absence of new prescriptions for anti-psychotic medication or death.
We will use marginal structural models to investigate the association between time-exposed to anti-psychotic polypharmacy (APPP) and prescription of an outcome medication. Least Angle Regression will be used to investigate competing hypotheses for an association.
Prescription of a medication for hypertension, diabetes or hypercholesterolemia during follow-up
- A READ code diagnosis of hypertension, diabetes, hypercholesterolemia
- An ICD-10 code for hypertension, diabetes, hypercholesterolemia
Sarah Sullivan - Chief Investigator - University of Bristol
Theresa Redaniel - Corresponding Applicant - University of Bristol
Blanca Bolea - Collaborator - University of Toronto
David Kessler - Collaborator - University of Bristol
Emily Catherine Eyles - Collaborator - University of Bristol
Hannah Edwards - Collaborator - University of Bristol
Jelena Savovic - Collaborator - University of Bristol
Manjula Nugawela - Collaborator - University College London ( UCL )
Sharea Ijaz - Collaborator - University of Bristol
Simon Davies - Collaborator - University of Toronto
HES Admitted Patient Care;Mental Health Services Data Set (MHSDS);Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation