Statins are a commonly prescribed drug which help reduce the risk of heart disease and death. They work by reducing the levels of cholesterol in your body. When cholesterol levels are measured with a blood test some people show greater reductions in cholesterol levels than others, even when prescribed the same amount of statins. We are not sure if the people with better reductions in their cholesterol levels are taking their tablets more often as prescribed or if there are other reasons. It is possible that having different health conditions and use of other medications may affect how well a persons cholesterol levels are reduced or not.
A growing number of people have two or more physical or mental health conditions this is known as multimorbidity. This research aims to explore which combinations of conditions may be important to managing an individuals cholesterol levels and risks. For example, we will look to see if there is a difference in the way people with diabetes and depression respond to statins compared to those with diabetes and high blood pressure. The Clinical Practice Research Datalink is a large database of routinely collected data which will allow us to study the relationship between certain groups of conditions and cholesterol levels. We hope our work will help develop personalised care for patients on statins.
Background and aims: Statins reduce lipid levels, which helps modify the risk of cardiovascular morbidity and death. However, there is variability in lipid response and there is some evidence that the reason for this may be modifiable. Given the growing prevalence of multimorbidity (two or more physical or mental health condition) in the population we aim to understand if this has an influence on lipid response to statins. This study aims to explore which groupings of conditions may be most important to managing lipid levels and preventing future cardiovascular disease.
Methods: We will undertake a prospective cohort study using the CPRD database of all patients started on statin therapy (those with pre-existing cardiovascular disease will be excluded). The exposure groups will be pre-specified common multimorbidity clusters in patients with cardiovascular disease, identified from the literature, which may influence lipid response and/or cardiovascular disease risk. We will use 20 clusters across 4 age-strata as described by Zhu et al (1) and additionally incorporate key statin drug interactions as agreed by 3-4 clinicians. The baseline unexposed group will be those without any identified pre-existing conditions.
Outcomes: Short-term outcomes will be 2-year reduction in lipid level by total cholesterol, LDL-C or non-HDL C. Long term outcomes will use a composite of first major adverse cardiovascular events (MACE) endpoints; all-cause mortality, and proxy measures of health service utilisation (hospitalisation and GP visits).
Statistical analysis: Association between pre-specified multimorbidity clusters and lipid response will be assessed using multivariate logistic regression. Cox proportional hazards model will be used to assess the association between the same multimorbidity clusters and future MACE. Both analyses will be adjusted for baseline CVD risk factors, sex and other potential confounders. Informative censoring of survival time will be taken into account by considering patients who have died or left the practice. A further competing-risk analysis for cause-specific (or sub-) hazard ratio (HR) will be used to calculate the cumulative incidence of the CVD outcomes of interest.
Short-term outcomes: 2-year reduction in lipid level by total cholesterol; LDL-C; non-HDL C.
Long-term outcomes: first major adverse cardiovascular events (MACE) endpoints; and all-cause mortality; hospitalisation; GP visits.
Joe Kai - Chief Investigator - University of Nottingham
Mohana Ratnapalan - Corresponding Applicant - University of Nottingham
Caroline Mitchell - Collaborator - University of Sheffield
Nadeem Qureshi - Collaborator - University of Nottingham
Ralph Kwame Akyea - Collaborator - University of Nottingham
Stephen Weng - Collaborator - University of Nottingham
Yana Vinogradova - Collaborator - University of Nottingham
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation