A Count-Based Binary Decision Method For Target Blood Pressure Achievement in Home Blood Pressure Monitoring Data Interpretation For Clinical Practices

Abstract

Home blood pressure (HBP) is useful to decide whether blood pressure (BP) is controlled. However, applying HBP to daily clinical practices is still challenging without easy access to the average HBP. Therefore, we developed a simple method to make a binary decision for the control of HBP through high BP counts. We simulated 100 cases of HBP series for each combination of 3 numbers of BP readings (K = 16, 20, 24) and 4 levels of the standard deviations (SDs = 5, 10, 15, 20). A high BP was defined as an individual BP ≥ 135/85 mmHg, and an uncontrolled HBP was defined as a mean HBP ≥ 135/85 mmHg. Validation for the decision method was conducted using actual HBP data. The C-statistics and the accuracy of the high BP counts for the uncontrolled HBP were generally high (> 0.85) for all combinations of Ks and SDs but decreased as SDs increased but remained steady as Ks increased. In the validation, the C-statistic of the high BP count-to-total BP reading (C-T) ratio was 0.985, and a C-T ratio ≥ 0.5 showed a sensitivity of 0.957, a specificity of 0.907 and an accuracy of 0.927. The count-based decision method can provide an accurate quick assessment of the controlledness of HBP.