Is calculation of a home blood pressure-based stability score a tool to improve risk stratification in clinical practice? (2025)

Emerging evidence suggests that the risk of cardiovascular complications is influenced not only by the extent of elevated average blood pressure levels but also by increased blood pressure variability [1] Numerous studies have supported this idea, indicating that higher blood pressure variability —whether in the short term (24 h), midterm (day-by-day), or long term (visit-to-visit)—can enhance cardiovascular risk prediction beyond the impact of average blood pressure [2]. However, the practical significance of blood pressure variability is still unclear. Currently, it is often overlooked in clinical practice, likely due to the complexity and lack of standardization in assessment methods and indices, as well as uncertain thresholds and clinical relevance.

Home blood pressure (HBP) monitoring allows for multiple readings in an individual’s usual environment, thereby avoiding the white-coat and masked hypertension phenomena [3]. HBP monitoring devices are widely used in many countries, are relatively low-cost, and are well-accepted for long-term use [3]. HBP monitoring is currently recommended as the best method for the long-term follow-up of patients undergoing hypertension treatment [3, 4]. HBP monitoring are primarily used to assess midterm BPV and provide a more accurate representation of blood pressure variability compared to in-office measurements, as they can offer a greater number of readings over time. However, both methods measure blood pressure under controlled conditions (such as specific settings, body positions, and protocols), which can buffer and thus underestimate the true blood pressure variability [5].

Numerous indices have been proposed to evaluate and quantify blood pressure variability. These indices aim to measure overall variability by considering the dispersion, sequence, and irregularity of blood pressure values. They also assess specific blood pressure variability patterns, such as nocturnal blood pressure decline, the ratio of night-to-day blood pressure levels, and morning blood pressure surge [6].

The standard deviation is commonly used to estimate the dispersion of raw blood pressure values due to its simplicity in calculation. However, it is highly dependent on the average blood pressure value [7]. The coefficient of variation, which is the standard deviation divided by the mean, is another widely used measure of blood pressure variability because it is independent of the average blood pressure levels, allowing for comparisons between data with different average values [7]. Since both the standard deviation and the coefficient of variation are influenced by diurnal blood pressure variation, these parameters should be adjusted for the duration of awake and asleep periods when assessing 24-h blood pressure variability [8].

Another measure of blood pressure variability is the average real variability (ARV). This metric evaluates the sequence of blood pressure readings by calculating the average of the absolute differences between consecutive measurements [9]. Recently, the blood pressure time in target range (TTR) has been introduced. This metric quantifies the proportion of time spent within the established therapeutic range, thereby reflecting both the mean blood pressure value and the degree of blood pressure variability [10].

In this issue of Hypertension Research, Kario et al. introduced a HBP stability score and evaluated its prognostic value in 4070 older Japanese residents from the Japan Morning Surge-Home Blood Pressure (J-HOP) extended study [11]. Over a 6-year observation period, 260 cardiovascular disease cases were recorded. The key finding was that individuals with a low HBP stability score had a progressively higher risk compared to those with a high score. Specifically, those in the very high-risk group (0 points) had a significantly increased risk of cardiovascular events (adjusted hazard ratio 3.97, 95% confidence interval 2.22–7.09; p < 0.001) compared to those in the optimal HBP stability score group (9–10 points), regardless of age, sex, medication, cardiovascular risk factors, and office blood pressure [11]. Methodologically, the baseline HBP collection period was 14 days, excluding readings from the first day. The HBP device was set to automatically take three separate blood pressure measurements at 15-s intervals for each assessment, both in the morning and evening.

The HBP stability score was calculated using the average of morning and evening systolic blood pressure, along with three HBP variability metrics: ARV, average peak (the average of the highest three systolic blood pressure values for each individual), and TTR (defined in this study as the proportion of time with an average systolic HBP of 100–135 mmHg). The thresholds allocated to each measure were selected based on the increased risk observed in previous publications using the same cohort [12, 13]. Therefore, these thresholds require validation in independent cohorts, including different ethnicities, other medical conditions, and in the general population.

The three measures of variability included in the stability score were somewhat correlated, suggesting the possibility of reducing the number of metrics in a simpler score. However, each measure also provides unique information. A more simplified score, potentially including just the average of morning and evening systolic blood pressure and one measure of HBP variability, could be an interesting area for future research, given that a prediction score that is easy to generate and interpret is more likely to be used in clinical practice.

No data on nighttime HBP were included in the stability score calculation. However, recent findings from the J-HOP nocturnal blood pressure study showed that nighttime peak HBP was significantly associated with stroke risk, independent of office blood pressure and mean morning, evening, and nighttime systolic blood pressure [14]. Therefore, nighttime HBP represents a residual risk factor for cardiovascular events that is not captured by the current version of the stability score. This could be an important consideration for future research.

The study is significant because it showed that a straightforward HBP-based stability score could predict the risk of cardiovascular events in individuals with hypertension. This finding offers valuable insights for an HBP-focused strategy in managing hypertension within healthcare and clinical settings.

The authors have recently developed a cardiovascular risk prediction model that incorporates HBP (systolic mean blood pressure and ARV) and found that the model performed best when HBP variability was included [15]. However, this model also considered other factors such as age, sex, high-density lipoprotein cholesterol, and history of cardiovascular disease and diabetes, which may not always be available in clinical settings. A key advantage of the HBP stability score presented in this paper [11] is that it relies solely on HBP readings, making it easier to use in clinical practice. Further research is needed to validate the HBP stability score in different populations before determining if it can improve risk stratification in clinical practice. Nonetheless, the paper by Kario et al. provides the initial evidence.