Skip to main content
Figure 5 | BMC Medical Research Methodology

Figure 5

From: Is blood pressure reduction a valid surrogate endpoint for stroke prevention? an analysis incorporating a systematic review of randomised controlled trials, a by-trial weighted errors-in-variables regression, the surrogate threshold effect (STE) and the biomarker-surrogacy (BioSurrogate) evaluation schema (BSES)

Figure 5

Stroke relative risk reduction and systolic BP difference - errors-in-variables. Upper graph: scatterplot weighted by trial size and by trial errors-in-variables (eiv) weighted least squares regression of stroke relative risk and systolic blood pressure difference reduction assuming measurement error (reliability coefficient = 0.9). Lower graph: scatterplot weighted by trial size and by trial eiv weighted least squares regression of stroke relative risk and systolic blood pressure difference reduction assuming measurement error (reliability coefficient = 0.7). The upper and lower bold solid lines are the upper and lower 95% prediction limits, the dashed inner lines are the 95% confidence limits, the dot-dash centre line is the mean regression line. The arrow indicates where the lower 95% prediction line intersects with the x axis. This is the Surrogate Threshold Effect (STE) for stroke reduction and is the systolic blood pressure difference needed to impute a stroke reduction benefit in a new trial given eiv regression.

Back to article page