Introduction: Based on population statistics of blood pressure values, the frequency of the false positive/negative diagnosis of hypertension could be decreased approximately by 20 % if the systematic errors of systolic and diastolic blood pressure measurements were less than 1 and 3 Hgmm respectively. The aim of this study was to compare the outcome of the comparative statistical classification with and without the consideration of the measurement uncertainty (U). Methods and results: Following the extensive analysis of the validation protocols and the ISO81060-2:2018 guideline, seven uncertainty components were defined to compute the combined uncertainty (uc) for of the systolic BP measurement: uc = 3.12 Hgmm. Reference and observed systolic BP values (n=304) from a shared database were modelled by Monte Carlo simulation and were resampled (m=1000) with and without expanding the sampled values with U under 95% coverage probability. The confidence interval for the mean differences between the reference and the observed values was reconstructed at the level of 95% coverage probability. The number of reliable comparisons was defined as the number of differences within the confidence interval: DiffCI. The number of DiffCI events obtained without U calculation for the reference and for the observed systolic BP values was 203 out of 304. The mean and standard deviation of DiffCI events obtained by Monte Carlo simulation expanding with U calculation only for the reference values and both for the reference and observed values were as follows respectively: DiffCI mean=150.90, DiffCI std=6.65, DiffCI mean=129.75, DiffCI std=7.88. Conclusion: Including the concept of the measurement uncertainty in the comparative validation process of noninvasive blood pressure measurement methods significantly changes the number of reliable comparisons. This highlights the importance of the direct bio-signal metrology approach compared to the general biostatistical approach based on indirect measurement.