Abstract
The association between visit‐to‐visit systolic blood pressure variability and cardiovascular events has recently received a lot of attention in the cardiovascular literature. But, blood pressure variability is usually estimated on a person‐by‐person basis and is therefore subject to considerable measurement error. We demonstrate that hazard ratios estimated using this approach are subject to bias due to regression dilution, and we propose alternative methods to reduce this bias: a two‐stage method and a joint model. For the two‐stage method, in stage one, repeated measurements are modelled using a mixed effects model with a random component on the residual standard deviation (SD). The mixed effects model is used to estimate the blood pressure SD for each individual, which, in stage two, is used as a covariate in a time‐to‐event model. For the joint model, the mixed effects submodel and time‐to‐event submodel are fitted simultaneously using shared random effects. We illustrate the methods using data from the Atherosclerosis Risk in Communities study.
Highlights
Systolic blood pressure (SBP) is universally recognised as an important risk factor for cardiovascular disease (CVD) and is routinely included in risk scores for CVD risk prediction.[1,2] The prognostic value of SBP is primarily based on the mean of measurements over multiple outpatient visits, whereas substantially less attention has been given to the variability of SBP across visits
In addition to assessing mean clinic SBP levels over time, measurements of visit-to-visit SBP variability may improve the accuracy of CVD risk prediction, which is crucial for the optimisation of patient care
We focus here on the association between SBP variability and CVD, the methods and results presented would be relevant for any application where the association between the variability in a longitudinal outcome and a time-to-event is of interest
Summary
Systolic blood pressure (SBP) is universally recognised as an important risk factor for cardiovascular disease (CVD) and is routinely included in risk scores for CVD risk prediction.[1,2] The prognostic value of SBP is primarily based on the mean of measurements over multiple outpatient visits, whereas substantially less attention has been given to the variability of SBP across visits (ie, visit-to-visit SBP variability). The standard deviation (SD), coefficient of variation (CV), average real variability (ARV), and SBP variability independent of the mean across multiple visits have been widely used to quantify visit-to-visit SBP variability.[3,4,5,6,7] these measures are estimated on a person-by-person basis and are subject to considerable measurement error. This measurement error causes regression dilution bias in the estimated association between visit-to-visit SBP variability and CVD.[8]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call