Blood pressure dynamics during home blood pressure monitoring with a digital blood pressure coach—a prospective analysis of individual user data

Figure 1. User numbers during baseline HBPM interval. (A) Number of users with self-reported diagnosis of hypertension who started the first HBPM interval (n = 3175). Percentages are given in brackets. (B) Number of users with a self-reported diagnosis of hypertension who have started a first HBPM interval and are uncontrolled on the first day of measurement (n = 1309). Percentages for day 2 and 6 are given in brackets.

Figure 2. Home blood pressure values in the first HBPM interval. Shown are plots of predicted smoothing spline functions of time (solid lines) with 95% confidence bands (matt surrounding). (A) Course of systolic and diastolic blood pressure values of users with self-reported hypertension in the first HBPM period. (B) Course of SBP and DBP stratified by initial (day 1) BP category. Smooth functions for systolic BP as a function of time show significant non-linear relationships (normal BP: EDF = 1.98, p < 2 × 10−16; high normal BP: EDF = 1.91, p = 5.4 × 10−3; grade I HTN: EDF = 1.91, p < 2 × 10−16; grade II HTN: EDF = 1.98, p < 2 × 10−16) as well as the diastolic BPs (normal BP: EDF = 1.97, p < 2 × 10−16; high normal BP: EDF = 1.85, p = 1.8 × 10−2; grade I HTN: EDF = 1.88, p = 1.6 × 10−5; grade II HTN: EDF = 1.99, p < 2 × 10−16). The observed EDFs identical or close to 2 indicate a parabolic relationship in all scenarios although the sign of the effects differs, i.e. BP increase as well as BP decrease over time is observed. Analyses are based on data from users with self-reported hypertension who enrolled in HBPM and completed the first day (n = 2582).

Figure 3. Blood pressure classification tracking during the HBPM protocol. (A) Fraction of users who were reclassified compared to the previous day; classification is always based on the cumulative BP mean of consecutive measurement days. (B) Percentage of users who already matched their final BP-category. Classification is based on the cumulative BP-mean of day 1–6. (C) Number of patients who changed their classification compared to the classification from the previous day. (D) Change in blood pressure categories after exclusion of day 1. The crosstab shows how users (grouped by category based on day 1–6 mean, vertical) are classified according to a day 2–6 protocol. (A–D) Analyses are based on data obtained from users with self-reported hypertension who completed the first HBPM -period (n = 2307).

Figure 4. Blood pressure values and BP control at baseline and during follow-up. (A + B) Distribution of absolute systolic (A) and diastolic (B) BP- values at baseline and at follow-up. The dashed-line indicates the mean BP. (C) Fraction of patients by SBP (mean of entire HBPM period) at baseline (solid) and at follow-up (dashed). The vertical line indicates 135 mmHg; the threshold for diagnosing HT. (D) Change in SBP at follow-up according to baseline SBP: solid line indicates steady linear fitting of data. Thin grey line is a LOESS based guide-to-the-eye to visualise the non-linear relationship approximated by a piecewise linear model (bold solid line). The horizontal dashed line (grey) represents the average decrease in SBP. All analyses are based on data obtained from users with self-reported hypertension, who completed a HBPM-period at baseline and follow-up (n = 864).