<b>Background</b> Cardiovascular risk scores are used to estimate absolute risk of disease and identify patients who will benefit most from treatments to lower risk. As a key risk factor for cardiovascular disease, blood pressure is accounted for in many risk scores, but is inherently variable and may be influenced by both biological and measurement factors. This thesis aims to determine how routinely collected blood pressure measurements should best be used for accurate estimation of cardiovascular risk. <b>Methods</b> This thesis describes four main studies. A patient survey and prospective study establish the current practice of blood pressure measurement. Secondary analyses of data from blood pressure monitoring trials determine how risk estimates may be affected by the use of different summary measures of blood pressure. A systematic review evaluates the evidence of an association between blood pressure variability and cardiovascular risk. Finally, a cohort study in the Clinical Practice Research Datalink determines if inclusion of blood pressure variability in cardiovascular risk scores may improve risk estimation. <b>Results</b> Current practice of blood pressure measurement may differ from that in risk score derivation studies. However, these differences have limited effects on cardiovascular risk estimates with few patients reclassified across risk thresholds. Increased long-term variability in blood pressure is in itself a risk factor for cardiovascular disease over and above mean blood pressure but its inclusion in a cardiovascular risk score does not materially improve the accuracy of risk estimates. <b>Conclusions</b> Healthcare professionals should continue to estimate risk for primary prevention of cardiovascular disease using the blood pressure measurements available to them, whether measured at home or in the clinic. There is also no additional benefit of considering measures of long-term blood pressure variability in risk estimation.