BackgroundPrecise blood pressure (BP) measurements are central for the diagnosis of hypertension in clinical and epidemiological studies. The purpose of this study was to quantify the variability in BP associated with arm side, body position, and successive measurements in the setting of a population-based observational study. Additionally, we aimed to evaluate the influence of different measurement conditions on prevalence of hypertension.MethodsThe sample included 967 men and 812 women aged 45 to 83 years at baseline. BP was measured according to a standardized protocol with oscillometric devices including three sitting measurements at left arm, one simultaneous supine measurement at both arms, and four supine measurements at the arm with the higher BP. Hypertension was defined as systolic BP (SBP) ≥140 mmHg and/or diastolic BP (DBP) ≥90 mmHg. Variability in SBP and DBP were analysed with sex-stratified linear covariance pattern models.ResultsWe found that overall, no mean BP differences were measured according to arm-side, but substantial higher DBP and for men also higher SBP was observed in sitting than in supine position and there was a clear BP decline by consecutive measurement. Accordingly, the prevalence of hypertension depends strongly on the number and scheme of BP measurements taken to calculate the index values.ConclusionsThus, BP measurements should only be compared between studies applying equal measurement conditions and index calculation. Moreover, the first BP measurement should not be used to define hypertension since it overestimates BP. The mean of second and third measurement offers the advantage of better reproducibility over single measurements.Electronic supplementary materialThe online version of this article (doi:10.1186/s12872-017-0468-7) contains supplementary material, which is available to authorized users.