Doctors use factors such as age, blood pressure and cholesterol to decide when patients are at risk of heart disease or stroke. A computer program or a formula for estimating risk, based on such factors, is called a cardiovascular risk calculator. One such calculator (QRISK risk prediction tool) is widely used across English GP practices. We are studying risk of heart disease or stroke in patients with kidney disease who are known to be at an increased risk of such events. It is unclear how well QRISK performs in this population, so we will use data on these patients to check the accuracy of QRISK and develop a new cardiovascular risk calculator specific to patients with kidney disease. We will compare the accuracy of this new calculator to that of the previously published risk calculators including QRISK. We will also examine the rate at which patients with kidney disease progress to later stages of kidney disease, compared to the rate at which they develop cardiovascular disease.
Patients with chronic kidney disease (CKD) are at an increased risk of cardiovascular disease but there are no widely accepted risk scores to evaluate risk in individual patients. The QRISK risk prediction tool is widely used in UK to evaluate 10-year risk of cardiovascular disease (ie, coronary heart disease, ischaemic stroke, or transient ischaemic attack) and guide primary cardiovascular disease prevention, including in people with mild CKD, but its performance in CKD is unknown. Firstly we aim to quantify the risk of progression of CKD, and incident CVD, in people with CKD, stratified by level of CKD (aim 1). Secondly (aim 2) we will validate the QRISK3 score in people with mild-to-advanced CKD (estimated glomerular filtration rate [eGFR]<90 ml/min/1.73m2), and (aims 3 and 4) develop and validate a new cardiovascular risk score (CKD-CVD Risk Score) using a primary prevention CKD cohort, derived from CPRD. The new CKD-CVD Risk Score will be derived separately for women and men. We will use a range of demographic and clinical risk factors including those used in QRISK3 as well as including additional factors, such as eGFR. The Harrells C index and the ratio of predicted-to-observed risks will be used in model validation.
Cardiovascular disease
End stage renal disease
Borislava MIHAYLOVA - Chief Investigator - University of Oxford
Richard Stevens - Corresponding Applicant - University of Oxford
Benjamin Feakins - Collaborator - University of Oxford
Claire Simons - Collaborator - University of Oxford
Iryna Schlackow - Collaborator - University of Oxford
Jason Oke - Collaborator - University of Oxford
Pengfei Zhu - Collaborator - University of Oxford
Rafael Perera - Collaborator - University of Oxford
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Townsend Score