Abstract
Background
Beyond average blood pressure (BP) levels, short‐term BP fluctuations may carry important prognostic information. However, the clinical significance of within‐visit BP changes remains limited. This study aimed to evaluate the association between within‐visit BP changes and clinical outcomes.
Methods
We analyzed data from the UK Biobank. BP was measured twice, with at least 1 minute between measurements, while participants were seated. Within‐visit BP changes were defined as the absolute difference between 2 consecutive systolic BP measurements, and participants were categorized into within‐visit BP changes quartiles. The primary outcome was all‐cause mortality, and secondary outcomes included new‐onset hypertension and major adverse cardiovascular events (nonfatal myocardial infarction, nonfatal stroke, and heart failure). Multivariable Cox proportional hazards models, adjusted for demographics, cardiovascular risk factors, comorbidities, and baseline BP, were used to estimate hazard ratios (HRs).
Results
Higher within‐visit BP changes were associated with elevated systolic and diastolic BP and a higher incidence of hypertension. All‐cause mortality and major adverse cardiovascular events increased progressively across BP changes quartiles, and these results remained consistent in fully adjusted models. Compared with the lowest quartile, the highest quartile showed an elevated risk of all‐cause mortality (HR, 1.070 [95% CI, 1.041–1.100]), nonfatal myocardial infarction (HR,1.070 [95% CI, 1.021–1.121]), stroke (HR,1.086 [95% CI, 1.027–1.148]), and heart failure (HR,1.069 [95% CI, 1.015–1.127]). BP changes as a continuous variable demonstrated a significant linear association without evidence of nonlinearity.
Conclusions
Within‐visit BP changes are independently associated with the development of hypertension, mortality, and major adverse cardiovascular events, supporting its potential role as a marker of long‐term cardiovascular risk.
Keywords: blood pressure, cardiovascular diseases, cohort studies, systolic blood pressure
Subject Categories: Hypertension
Clinical Perspective.
What Is New?
In this large‐scale analysis of 416 922 participants from the UK Biobank without a history of hypertension, within‐visit blood pressure change was independently associated with increased risks of new‐onset hypertension, all‐cause mortality, and major adverse cardiovascular events.
What Are the Clinical Implications?
These findings suggest that within‐visit blood pressure change, a readily obtainable metric during routine clinical visits, may serve as an additional marker for identifying individuals at higher long‐term cardiovascular risk even in the absence of established hypertension.
Clinicians should consider documenting and monitoring blood pressure fluctuations within a single visit, as elevated variability may warrant closer surveillance for future hypertension development and adverse cardiovascular events.
Hypertension affects >1 billion adults worldwide; however, approximately 50% remain undiagnosed, and less than half receive appropriate diagnosis and treatment. 1 Despite these low levels of awareness and control, elevated blood pressure (BP) accounts for approximately one fifth of all‐cause mortality and is a leading contributor to cardiovascular mortality, irrespective of age and sex. 2 In addition to these global burden estimates, several large observational studies have consistently demonstrated that elevated BP is a well‐established risk factor for major adverse cardiovascular outcomes. 3 , 4 Given that hypertension is both a highly consequential and modifiable risk factor, concerted global efforts have been directed at improving its control. 5 , 6 Supporting these efforts, contemporary guidelines recommend optimal BP‐lowering treatment to reduce the risk of cardiovascular disease. 7 , 8
BP change, defined as dynamic physiological fluctuations in BP over time, has emerged as a significant cardiovascular risk factor independent of mean BP levels. 7 Multiple large‐scale cohort studies and meta‐analyses have consistently demonstrated that increased BP change is associated with a higher incidence of stroke, myocardial infarction, heart failure, and all‐cause mortality. 9 , 10 , 11 BP change can be categorized into several types based on the measurement time frame: very short‐term (from beat‐to‐beat to within minutes), short‐term (within 24 hours), and mid‐ to long‐term variability assessed by visit‐to‐visit differences. 12 Among these, both visit‐to‐visit 9 , 10 , 11 and short‐term BP changes 13 within a 24‐hour period have been consistently associated with adverse cardiovascular outcomes. In particular, the daytime and nighttime BP changes have been identified as a significant predictor of mortality, coronary artery disease, and stroke. 14 , 15 Furthermore, elevated morning BP has recently been recognized as an early and strong risk factor for stroke. 16 However, the clinical significance of within‐visit BP changes, defined as BP fluctuations occurring over a few minutes during a single visit, remains less well established. 17 , 18 Although previous study has suggested that higher within‐visit BP change is associated with increased cardiovascular risk, 19 further evidence from large‐scale cohort is warranted, as data specifically addressing within‐visit BP changes remain limited. Therefore, this study aimed to investigate the association between within‐visit BP changes and clinical outcomes in a large, community‐based prospective cohort.
METHODS
Data Transparency
All data and materials have been made publicly available at the public UK Biobank Resource and can be accessed at www.ukbiobank.ac.uk/.
Study Design and Patient Population
The UK Biobank is a community‐based prospective cohort comprising >500 000 middle‐to‐old‐aged adults recruited between 2006 and 2010 at 22 assessment centers across England, Wales, and Scotland. Comprehensive information on sociodemographic and lifestyle factors, physical and health assessments, genetic data, and imaging findings was collected through online questionnaires, in‐person interviews, physical measurements, linkage to electronic health records, biological sampling, and imaging scans. 20 , 21 The study flow chart is presented in Figure 1. Among the 502 412 participants in the UK Biobank, individuals were excluded if they had missing BP data (n=1326), only a single BP measurement (n=508), other missing baseline covariates (n=36 687), or a history of hypertension (n=36 250) or cardiovascular disease (n=17 772). History of hypertension was defined using the International Classification of Diseases, Tenth Revision (ICD‐10) codes I10–I13 and I15. Cardiovascular disease was defined as nonfatal myocardial infarction (MI), nonfatal stroke, or heart failure, consistent with the definitions used for clinical outcomes. A total of 416 922 participants were included in the final analysis and were categorized into quartiles according to within‐visit BP changes.
Figure 1. Study flow chart.
BP indicates blood pressure; DBP, diastolic blood pressure; and SBP, systolic blood pressure.
The UK Biobank was established with ethical approval from the North West Multi‐Centre Research Ethics Committee, and all participants provided written informed consent at recruitment. The current analysis was conducted under UK Biobank application number 85037.
Data Collection
Baseline information on the following potential confounders was obtained through the touchscreen questionnaire: age, sex, race, ethnicity, education level, physical activity (total weekly minutes of moderate physical activity 22 ), current smoking, medication use at baseline (aspirin and cholesterol‐lowering medication). Body mass index (BMI) was derived from anthropometric measurements by dividing weight (kg) by height squared (m2). Peripheral blood samples were obtained from all participants at recruitment. A validated protocol was employed for sample collection and processing. 23
BP was measured twice, with at least 1 minute between measurements, using the Omron 705 IT electronic BP monitor (OMRON Healthcare), after participants had been seated at rest for 5 minutes. 24 Within‐visit BP changes were defined as the absolute difference between the 2 systolic BP measurements. Quartiles of within‐visit BP changes were defined based on the distribution of the absolute difference: for systolic BP changes, Q1 (≤3 mm Hg), Q2 (4–6 mm Hg), Q3 (7–11 mm Hg), and Q4 (≥12 mm Hg); and for diastolic BP changes, Q1 (≤1 mm Hg), Q2 (2–3 mm Hg), Q3 (4–5 mm Hg), and Q4 (≥6 mm Hg).
Outcomes and Covariate Definitions
The primary outcome was all‐cause mortality, and secondary outcomes included new‐onset hypertension, nonfatal MI, nonfatal stroke, and heart failure. Follow‐up commenced at the time of enrolment in the UK Biobank and continued until the earliest occurrence of any predefined outcome or December 31, 2024. The date of the first occurrence of each outcome was identified through linkage with hospital admission records and national death registries across England, Scotland, and Wales. All‐cause mortality was defined using the national death registry, with causes of death coded according to the ICD‐10. 25 Nonfatal MI and nonfatal stroke were identified using an algorithm developed by the UK Biobank Outcome Adjudication Group, which integrates ICD‐coded data from hospital admissions, death registries, and baseline self‐reported data. To ensure high diagnostic accuracy, MI and stroke were defined based on hospital admission records listing the condition as the primary diagnosis, which has been shown to have the highest positive predictive value. Events identified solely from death registry and self‐reported data were excluded because of lower reliability. 26 Hypertension was defined using ICD‐10 codes I10–I13 and I15. New‐onset heart failure was identified using codes I11.0, I11.3, I13.2, I25.5, I42, and I50.
Baseline comorbidities, including diabetes (ICD‐10 codes E10–14), dyslipidemia (ICD‐10 code E78), coronary artery disease (defined by ICD‐10 codes I21–25, excluding codes used for algorithm‐defined MI), peripheral artery disease (ICD‐10 codes I70, I78.9, and I73.9), and atrial fibrillation or flutter (ICD‐10 code I48), were obtained from health records (Table S1).
Statistical Analysis
Continuous variables are summarized as mean±SD, and categorical variables are presented as number (%). Group differences in baseline characteristics across within‐visit BP changes quartiles were assessed using 1‐way ANOVA for continuous variables and the chi‐square test for categorical variables.
Incidence rates were calculated as events per 1000 person‐years with 95% CIs. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% CIs for clinical outcomes across quartiles of within‐visit BP changes, with the lowest quartile serving as the reference. Linear trends were assessed by including systolic and diastolic BP changes as continuous variables in the Cox proportional hazards models, with HRs estimated per 5 mm Hg increase. Three models were constructed with increasing levels of covariate adjustment to account for potential confounders: Model 1 was unadjusted; Model 2 was adjusted for age, sex, race, and ethnicity; and Model 3 was adjusted for the same variables as Model 2 and also for education level (categorized as low, intermediate, high, or uncategorized), exercise (categorized as inactive, low, moderate, or high), smoking status (yes or no), BMI (kg/m2), total cholesterol level (mg/dL), diabetes (yes or no), coronary artery disease (yes or no), atrial fibrillation (yes or no), peripheral arterial disease (yes or no), and baseline systolic BP and diastolic BP (mm Hg).
Kaplan–Meier survival curves and log‐rank tests were used to assess differences in cumulative incidence across groups. To analyze the potential nonlinear associations between within‐visit BP changes and clinical outcomes, restricted spline curve models were fitted within Cox proportional hazards models. Spline terms were based on within‐visit BP changes, with knots placed at the minimum, 33.3rd, 66.7th, and maximum percentiles of its distribution. HRs and 95% CIs were estimated across the BP changes spectrum, with the median value set as the reference point (HR=1).
Sensitivity analyses were conducted to evaluate the robustness of the association between within‐visit BP changes and clinical outcomes, including analyses stratified by baseline systolic and diastolic BP and prespecified subgroups. For the BP stratified analysis, participants were categorized into 3 groups based on baseline systolic BP (<120, 120–139, and ≥140 mm Hg) and diastolic BP (<70, 70–89, and ≥90 mm Hg). Associations between within‐visit BP changes and clinical outcomes were examined within each systolic BP stratum using fully adjusted Cox models to assess consistency across different BP levels. For subgroup analysis, HRs for the highest versus lowest quartile of within‐visit BP changes were estimated within each subgroup, using fully adjusted (model 3) Cox regression models. Subgroups were defined by age (≥65 versus <65 years), sex, race (White versus other ethnicities), BMI (≥30 versus <30 kg/m2), smoking status, diabetes status, dyslipidemia status, and physical activity (≥150 versus <150 min/wk). Potential effect modification across subgroups was explored by including multiplicative interaction terms in the Cox regression models. In addition, we performed a sensitivity analysis stratified by the direction of BP change to examine whether increases versus decreases between the 2 readings had different prognostic implications. Clinical outcomes were analyzed separately for participants with increased (1st < 2nd measurement) and decreased (1st > 2nd measurement) changes.
Data manipulation and statistical analyses were conducted in R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). All tests were 2 sided, and statistical significance was set at P<0.05.
RESULTS
Baseline Characteristics of Participants
The baseline characteristics of participants, stratified by within‐visit systolic BP changes quartiles, are summarized in Table 1. Among the 416 922 participants, 44.4% were male and 94.9% were White. The mean age of the cohort was 56.0 years. The mean systolic and diastolic BP were 137.3 and 82.2 mm Hg, respectively. The prevalence of diabetes, dyslipidemia, and coronary artery disease was 1.8%, 2.3%, and 1.4%, respectively. Use of cholesterol‐lowering medications was reported in 12.6% of participants.
Table 1.
Baseline Characteristics of Participants, Stratified by Quartiles of Systolic BP Changes
| Characteristic | Quartile of systolic BP changes | |||||
|---|---|---|---|---|---|---|
| Overall | Quartile 1 (≤3 mm Hg) | Quartile 2 (4–6 mm Hg) | Quartile 3 (7–11 mm Hg) | Quartile 4 (>12 mm Hg) | P value | |
| No.=416 922 | No.=126 113 | No.=92 859 | No.=106 423 | No.=91 527 | ||
| Age, y | 56.02±8.10 | 55.23±8.11 | 55.48±8.13 | 56.11±8.07 | 57.54±7.86 | <0.001 |
| Male sex | 185 246 (44.4%) | 57 094 (45.3%) | 41 932 (45.2%) | 47 391 (44.5%) | 38 829 (42.4%) | <0.001 |
| Race | <0.001 | |||||
| White | 395 537 (94.9%) | 119 133 (94.5%) | 87 807 (94.6%) | 101 107 (95.0%) | 87 490 (95.6%) | |
| Asian or Asian British | 8956 (2.1%) | 2861 (2.3%) | 2171 (2.3%) | 2195 (2.1%) | 1729 (1.9%) | |
| Black or Black British | 7082 (1.7%) | 2355 (1.9%) | 1678 (1.8%) | 1757 (1.7%) | 1292 (1.4%) | |
| Other* | 5347 (1.3%) | 1764 (1.4%) | 1203 (1.3%) | 1364 (1.3%) | 1016 (1.1%) | |
| Education level | <0.001 | |||||
| Lower education levels | 70 184 (16.8%) | 21 556 (17.1%) | 15 727 (16.9%) | 17 726 (16.7%) | 15 175 (16.6%) | |
| Intermediate education levels | 76 213 (18.3%) | 23 526 (18.7%) | 17 151 (18.5%) | 19 408 (18.2%) | 16 128 (17.6%) | |
| Higher education and professional qualifications | 201 357 (48.3%) | 61 120 (48.5%) | 44 928 (48.4%) | 51 655 (48.5%) | 43 654 (47.7%) | |
| Unknown | 69 168 (16.6%) | 19 911 (15.8%) | 15 053 (16.2%) | 17 634 (16.6%) | 16 570 (18.1%) | |
| Exercise | <0.001 | |||||
| Inactive (0–10 min/wk) | 63 462 (15.2%) | 19 332 (15.3%) | 14 186 (15.3%) | 16 061 (15.1%) | 13 883 (15.2%) | |
| Low (10–149 min/wk) | 97 709 (23.4%) | 29 720 (23.6%) | 21 953 (23.6%) | 24 996 (23.5%) | 21 040 (23.0%) | |
| Moderate (150–299 min/wk) | 63 525 (15.2%) | 19 161 (15.2%) | 14 452 (15.6%) | 16 103 (15.1%) | 13 809 (15.1%) | |
| High (≥300 min/wk) | 160 334 (38.5%) | 48 512 (38.5%) | 35 308 (38.0%) | 41 069 (38.6%) | 35 445 (38.7%) | |
| Unknown | 31 892 (7.6%) | 9388 (7.4%) | 6960 (7.5%) | 8194 (7.7%) | 7350 (8.0%) | |
| Current smoking | 43 431 (10.4%) | 13 931 (11.0%) | 9936 (10.7%) | 10 989 (10.3%) | 8575 (9.4%) | <0.001 |
| Mean body mass index (kg/m2) | 27.14±4.60 | 27.03±4.54 | 27.05±4.55 | 27.15±4.60 | 27.34±4.75 | <0.001 |
| Mean heart rate, bpm | 69.38±11.03 | 68.89±10.78 | 69.09±10.87 | 69.38±10.95 | 70.37±11.54 | <0.001 |
| Mean systolic BP, mm Hg | 137.33±18.51 | 134.44±18.06 | 135.43±18.10 | 137.53±18.06 | 143.01±18.76 | <0.001 |
| Mean diastolic BP, mm Hg | 82.21±10.07 | 81.37±10.00 | 81.70±9.98 | 82.28±9.99 | 83.79±10.19 | <0.001 |
| Mean pulse pressure, mm Hg | 55.12±13.39 | 53.07±12.78 | 53.73±12.88 | 55.25±13.09 | 59.22±14.11 | <0.001 |
| Total cholesterol level | 223.14±43.04 | 222.21±42.63 | 222.60±42.74 | 223.07±43.18 | 225.04±43.69 | <0.001 |
| Comorbidities | ||||||
| Dyslipidemia | 3916 (0.9%) | 1134 (0.9%) | 841 (0.9%) | 1017 (1.0%) | 924 (1.0%) | 0.037 |
| Diabetes | 3812 (0.9%) | 1074 (0.9%) | 772 (0.8%) | 994 (0.9%) | 972 (1.1%) | <0.001 |
| Coronary artery disease | 2969 (0.7%) | 863 (0.7%) | 628 (0.7%) | 772 (0.7%) | 706 (0.8%) | 0.049 |
| Atrial fibrillation or flutter | 2578 (0.6%) | 716 (0.6%) | 502 (0.5%) | 689 (0.6%) | 671 (0.7%) | <0.001 |
| Peripheral arterial disease | 178 (0.0%) | 51 (0.0%) | 20 (0.0%) | 43 (0.0%) | 64 (0.1%) | <0.001 |
| Medication | ||||||
| Cholesterol lowering agents | 52 416 (12.6%) | 14 403 (11.4%) | 10 978 (11.8%) | 13 548 (12.7%) | 13 487 (14.7%) | <0.001 |
| Aspirin | 42 190 (10.1%) | 12 058 (9.6%) | 8860 (9.5%) | 10 896 (10.2%) | 10 376 (11.3%) | <0.001 |
Data are presented as the mean±SD or n (%). BP indicates blood pressure.
Other includes Chinese, Mixed, and other ethnic groups based on UK Biobank self‐reported ethnicity classifications.
Higher within‐visit BP changes were associated with female sex, older age, White race, higher BMI, lower educational attainment, higher systolic and diastolic BP with wider pulse pressure, a higher burden of comorbidities, and increased medication use. Physical activity varied significantly across quartiles; this variation was mainly because of the proportion of unknown activity data rather than differences in reported activity levels.
The baseline characteristics of participants, stratified by within‐visit diastolic BP changes quartiles, are presented in Table S2, showing trends similar to those observed for systolic BP changes.
Risk of Hypertension Across Within‐Visit BP Changes
The incidence of hypertension among participants is presented in Table 2. Higher within‐visit BP changes were associated with a stepwise increase in both the incidence and HR of hypertension, with the highest quartile (Q4) showing the greatest risk. These results persisted after adjusting for several potential confounders in Models 2 and 3. Kaplan–Meier curves demonstrated a clear separation among quartiles in both systolic and diastolic BP changes, with the highest cumulative incidence of hypertension in Q4 (log rank P<0.001) (Figure 2).
Table 2.
Incidence of Hypertension Stratified by Quartiles of Systolic BP Changes
| Outcome | Quartile 1 (≤3 mm Hg) | Quartile 2 (4–6 mm Hg) | Quartile 3 (7–11 mm Hg) | Quartile 4 (>12 mm Hg) |
|---|---|---|---|---|
| Hypertension | ||||
| Cases, N | 29 108 (23.08%) | 22 194 (23.90%) | 27 574 (25.91%) | 27 848 (30.43%) |
| Incidence rate* (95% CI) | 16.996 (16.803–17.191) | 17.735 (17.504–17.968) | 19.544 (19.316–19.774) | 23.903 (23.626–24.182) |
| Hazard ratio (95% CI) | ||||
| Model 1† | 1.000 (Reference) | 1.042 (1.024–1.060) | 1.146 (1.127–1.165) | 1.395 (1.373–1.419) |
| Model 2‡ | 1.000 (Reference) | 1.024 (1.006–1.042) | 1.087 (1.070–1.105) | 1.220 (1.200–1.240) |
| Model 3§ | 1.000 (Reference) | 1.040 (1.022–1.058) | 1.095 (1.077–1.113) | 1.191 (1.171–1.211) |
BP indicates blood pressure.
Per 1000 person‐years.
Unadjusted model.
Includes adjustment for age, sex, race, and ethnicity.
Includes variables in Model 2 and additional adjustment for education level, smoking status, exercise, body mass index, total cholesterol level, diabetes, and coronary artery disease, atrial fibrillation, peripheral arterial disease, baseline systolic BP, and diastolic BP.
Figure 2. Time‐to‐event curves for hypertension, stratified by quartiles of systolic and diastolic blood pressure changes.
Kaplan–Meier estimate for hypertension across quartiles of (A) SBP and (B) DBP changes. DBP indicates diastolic blood pressure; and SBP, systolic blood pressure.
Cardiovascular Outcomes and Mortality Across Within‐Visit BP Changes
The median follow‐up duration was 15.8 years (interquartile range, 15.1–16.5). Cardiovascular outcomes and mortality among participants are presented in Table 3. The cumulative incidence of all‐cause mortality, nonfatal MI, stroke, and heart failure increased progressively across within‐visit BP changes quartiles. This association remained consistent after adjustment for several potential confounders in Models 2 and 3. In the fully adjusted model (Model 3), participants in Q4 had significantly higher risks of all‐cause mortality (HR, 1.070 [95% CI, 1.041–1.100]), nonfatal MI (HR,1.070 [95% CI, 1.021–1.121]), stroke (HR,1.086 [95% CI, 1.027–1.148]), and heart failure (HR,1.069 [95% CI, 1.015–1.127]) than did those in the lowest quartile (Q1). These findings are depicted in Figures S1 and S2, which show Kaplan–Meier survival curves for each outcome according to systolic and diastolic BP changes quartiles. In addition, the detailed Cox proportional HRs with fully adjusted model (Model 3) are presented in Tables S3 and S4. These results demonstrated that the traditional cardiovascular risk factors were consistently associated with all clinical outcomes, supporting the robustness of our findings.
Table 3.
Clinical Outcomes Stratified by Quartiles of Systolic and Diastolic BP Changes
| Systolic BP changes | Diastolic BP changes | |||||
|---|---|---|---|---|---|---|
| Model 1* | Model 2† | Model 3‡ | Model 1* | Model 2† | Model 3‡ | |
| All‐cause mortality | ||||||
| Quartile 1 | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) |
| Quartile 2 | 1.044 (1.015–1.075) | 1.019 (0.990–1.049) | 1.026 (0.997–1.056) | 1.005 (0.978–1.034) | 0.996 (0.969–1.024) | 0.998 (0.971–1.026) |
| Quartile 3 | 1.074 (1.045–1.104) | 0.996 (0.969–1.024) | 1.002 (0.974–1.030) | 1.071 (1.039–1.104) | 1.034 (1.004–1.066) | 1.031 (1.000–1.062) |
| Quartile 4 | 1.280 (1.245–1.316) | 1.064 (1.035–1.093) | 1.070 (1.041–1.100) | 1.186 (1.153–1.219) | 1.086 (1.056–1.117) | 1.064 (1.035–1.094) |
| Nonfatal myocardial infarction | ||||||
| Quartile 1 | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) |
| Quartile 2 | 1.018 (0.970–1.068) | 1.000 (0.953–1.049) | 1.008 (0.961–1.058) | 0.993 (0.948–1.040) | 0.986 (0.942–1.033) | 0.987 (0.942–1.033) |
| Quartile 3 | 1.060 (1.013–1.110) | 1.014 (0.968–1.061) | 1.020 (0.975–1.068) | 1.049 (0.997–1.104) | 1.021 (0.970–1.074) | 1.011 (0.961–1.064) |
| Quartile 4 | 1.196 (1.142–1.253) | 1.075 (1.026–1.126) | 1.070 (1.021–1.121) | 1.176 (1.122–1.232) | 1.100 (1.050–1.153) | 1.062 (1.013–1.113) |
| Stroke | ||||||
| Quartile 1 | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) |
| Quartile 2 | 1.021 (0.963–1.082) | 0.997 (0.941–1.057) | 1.008 (0.951–1.069) | 1.045 (0.988–1.105) | 1.037 (0.980–1.096) | 1.040 (0.984–1.100) |
| Quartile 3 | 1.082 (1.024–1.144) | 1.011 (0.956–1.068) | 1.019 (0.964–1.078) | 1.105 (1.039–1.175) | 1.071 (1.007–1.139) | 1.070 (1.006–1.137) |
| Quartile 4 | 1.282 (1.213–1.355) | 1.083 (1.025–1.145) | 1.086 (1.027–1.148) | 1.239 (1.172–1.311) | 1.146 (1.083–1.212) | 1.118 (1.057–1.183) |
| Heart failure | ||||||
| Quartile 1 | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) | 1.000 (Reference) |
| Quartile 2 | 1.053 (0.996–1.112) | 1.025 (0.971–1.083) | 1.041 (0.985–1.099) | 0.989 (0.937–1.043) | 0.980 (0.929–1.034) | 0.976 (0.925–1.030) |
| Quartile 3 | 1.099 (1.043–1.158) | 1.018 (0.966–1.073) | 1.021 (0.969–1.076) | 1.153 (1.089–1.221) | 1.113 (1.051–1.178) | 1.095 (1.034–1.159) |
| Quartile 4 | 1.306 (1.239–1.376) | 1.081 (1.026–1.140) | 1.069 (1.015–1.127) | 1.297 (1.230–1.366) | 1.183 (1.122–1.247) | 1.114 (1.056–1.174) |
BP indicates blood pressure.and HR, hazard ratio.
Unadjusted model.
Includes adjustment for age, sex, race, and ethnicity.
Includes variables in Model 2 and additional adjustment for education level, smoking status, exercise, body mass index, total cholesterol level, diabetes, and coronary artery disease, atrial fibrillation, peripheral arterial disease, baseline systolic BP, and diastolic BP.
The association between within‐visit BP changes and clinical outcomes was further analyzed by modeling BP changes as a continuous variable. The HRs for each 5 mm Hg in systolic or diastolic BP changes are presented in Table S5. These analyses indicated that the diastolic BP changes was consistently associated with increased risks of all clinical outcomes, whereas the associations for systolic BP changes were attenuated after full adjustment, remaining significant only for all‐cause mortality. To explore potential nonlinear relationships, restricted cubic spline modeling based on the fully adjusted model (Model 3) was performed (Figures 3 and 4). This analysis revealed significant linear associations between within‐visit BP changes and all‐cause mortality, nonfatal MI, stroke, and heart failure (P for linearity: <0.001 for all clinical outcomes), without evidence of nonlinearity (P for nonlinearity; 0.108, 0.951, 0.307, and 0.952, respectively). A similar linear pattern was also observed for diastolic BP changes.
Figure 3. Restrict spline curves for all‐cause mortality, nonfatal myocardial infarction, stroke, and heart failure according to continuous within‐visit systolic blood pressure changes.
Adjusted hazard ratios with 95% CIs were estimated using Cox proportional hazards models with restricted cubic splines to model the continuous relationship between SBP changes and (A) all‐cause mortality, (B) nonfatal MI, (C) stroke, and (D) heart failure. Histograms indicate the distribution of systolic BP changes in the study population. MI indicates myocardial infarction; and SBP, systolic blood pressure.
Figure 4. Restrict spline curves for all‐cause death, nonfatal myocardial infarction, stroke, and heart failure according to continuous within‐visit diastolic blood pressure changes.
Adjusted hazard ratios with 95% CIs were estimated using Cox proportional hazards models with restricted cubic splines to model the continuous relationship between systolic BP changes and (A) all‐cause mortality, (B) nonfatal MI, (C) stroke, and (D) heart failure. Histograms indicate the distribution of DBP changes in the study population. DBP indicates diastolic blood pressure; and MI, myocardial infarction.
Sensitivity Analysis
BP‐stratified analysis revealed that higher within‐visit BP changes were consistently associated with an increased risk of all‐cause mortality across all systolic and diastolic BP categories. Although HRs were highest among participants with systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, the relative increase across BP changes categories was more pronounced in individuals with borderline systolic and diastolic BP levels (Figure 5 and Figure S3). A similar directional pattern was observed for secondary cardiovascular outcomes across BP strata (Figures S4 and S5).
Figure 5. Risk of all‐cause mortality, stratified by within‐visit blood pressure changes, across systolic and diastolic blood pressure strata.
A, Systolic BP categories and B, Diastolic BP categories. BP indicates blood pressure.
Subgroup analyses demonstrated generally consistent associations between within‐visit BP changes and clinical outcomes for most participant characteristics. The association between BP changes and all‐cause mortality and stroke was more pronounced in participants younger than 65 years (P for interaction: <0.001 and =0.003, respectively), and stroke risk was also higher in other ethnicities individuals (P for interaction=0.004). Risk was notably higher among current smokers for all‐cause mortality (P for interaction=0.039). In addition, significant effect modification by BP change direction was observed. The associations of BP changes with all‐cause mortality (for systolic BP changes) and nonfatal MI for (diastolic BP changes) were stronger among individuals whose BP increased between measurements compared with those whose BP decreased (P for interaction=0.006 and =0.015, respectively). No significant interactions were observed for sex, BMI, diabetes, dyslipidemia, or physical activity (all P for interaction >0.05) (Figure S6).
In participants with increased BP (1st < 2nd measurement) during the visit (Table S6), Q4 demonstrated significantly higher risks for all outcomes compared with Q1 across all models. Similarly, in participants with decreased BP (first < second measurement) (Table S7), higher BP changes quartiles remained associated with increased risks for adverse outcomes, though the magnitude of associations was somewhat attenuated after full adjustment.
DISCUSSION
This retrospective, observational post hoc analysis of the large UK Biobank cohort demonstrated that higher within‐visit BP changes were associated with a greater incidence of hypertension, which was subsequently linked to increased risks of all‐cause mortality and adverse cardiovascular outcomes. These results remained consistent in fully adjusted models and demonstrated a linear trend in restricted cubic spline analysis. Similar patterns were observed with diastolic BP changes analysis, supporting the robustness of our findings across both systolic and diastolic measurements. These findings build upon previous observational data indicating that greater within‐visit BP changes, typically assessed using 3 measurements, is associated with worse cardiovascular risk profiles and may have clinical significance. 17 Our results extend this concept by demonstrating that even a simple measure of within‐visit BP changes, derived from 2 consecutive measurements during a single visit, was associated not only with unfavorable cardiovascular risk profiles but also with clinical cardiovascular events and mortality.
Growing evidence has linked long‐term BP change, assessed by visit‐to‐visit differences, to increased cardiovascular risk in both the general population and patients with hypertension. 9 , 10 Several studies have also suggested that short‐term BP change, assessed within a 24‐hour period, is associated with adverse cardiovascular outcomes. Previous studies have shown that all types of BP change—including daytime, night‐time, and visit‐to‐visit change—are associated with cardiovascular outcomes. For instance, the Ohasama study 15 demonstrated that daytime BP change was significantly associated with cardiovascular mortality. However, other studies 13 , 27 have indicated that the prognostic impact of short‐term BP change appears to be primarily driven by night‐time BP change. These discrepancies may reflect differences in sample sizes, study designs, and populations, leaving the relative importance of different BP change types uncertain.
In comparison, only a limited number of studies have investigated within‐visit BP change, and the evidence regarding its association with cardiovascular outcomes remains inconclusive. Schutte et al. 18 failed to identify a significant relationship between within‐visit BP change and cardiovascular outcomes or mortality. Subsequently, Grassi et al. 17 demonstrated that higher within‐visit BP change was associated with worse cardiovascular risk profiles, including unfavorable glycolipid profiles, metabolic syndrome, and resistant hypertension. However, they did not assess direct associations with clinical cardiovascular outcomes. More recently, Zarife et al. 19 showed that within‐visit BP change was independently associated with increased cardiovascular risk, suggesting the potential clinical relevance of within‐visit BP change. Another study from Poland reported an association between within‐visit BP change and cardiovascular outcomes, although the results may have been confounded by other clinical factors. 27 These discrepancies may be attributable to differences in study design, baseline characteristics of study populations, and statistical power. For instance, the absence of significant associations in Schutte et al. 18 may reflect the relatively young and low‐risk study populations with low event rates, whereas the positive associations in Zarife et al. 19 and Polish study 27 could have been influenced by smaller sample sizes, single‐country populations, or residual confounding factors. In addition, whereas most prior studies have quantified BP variability using indices derived from ≥3 measurements, within‐visit BP changes in the present study were defined as the absolute difference between 2 consecutive office BP measurements obtained during a single visit. Although previous studies also focused on short‐term BP fluctuations, they generally relied on multiple readings to derive conventional variability indices, whereas our approach operationalizes BP change using a difference‐based metric obtained during a single visit. This methodological distinction reflects the heterogeneity of approaches used in prior studies to characterize short‐term BP fluctuations and highlights a need for greater conceptual clarity when interpreting such measures. Against this background, the present study used the UK Biobank, a validated, large, and well‐characterized prospective cohort with long‐term follow‐up and adjudicated clinical outcomes, which may strengthen the generalizability and reliability of our findings. This robust data set provided sufficient statistical power to assess the prognostic impact of within‐visit BP changes not only on the incidence of hypertension but also on hard clinical outcomes, including cardiovascular events and mortality.
The mechanisms underlying BP change are multifactorial and involve both physiological and behavioral components. Key physiological contributors include autonomic dysregulation, arterial stiffness, and endothelial dysfunction, all of which have been implicated in impaired BP homeostasis and increased cardiovascular risk. 28 In addition, environmental and behavioral factors, such as physical activity, emotional stress, and poor adherence to antihypertensive treatment, can further exacerbate BP fluctuations during clinical encounters. 29 Although beat‐to‐beat variability predominantly reflects physiological mechanisms, such as baroreflex sensitivity and cardiorespiratory interactions, within‐visit BP changes are more strongly influenced by behavioral and environmental factors. 29 Despite being less physiologically specific, within‐visit BP changes may provide clinical insights by capturing real‐world cardiovascular instability and reflecting dynamic interactions between physiological and behavioral stressors. 12 , 29 Compared with short‐term BP change within 24 hours, which reflects circadian and activity‐related fluctuations, within‐visit BP changes operate on a different temporal scale but shares overlapping mechanisms involving autonomic regulation. 28 , 29 , 30 This physiological overlap across time scales may explain why associations between within‐visit BP changes and adverse outcomes have been observed in previous studies, consistent with the associations observed in our study.
Within‐visit BP changes may reflect physiological mechanisms, such as underlying autonomic dysregulation, arterial stiffness, or endothelial dysfunction, although it is largely influenced by behavioral and environmental factors. 12 Given these contributing mechanisms, greater within‐visit BP changes may be associated with the future development of hypertension. However, few studies have directly investigated the relationship between BP change and the incidence of hypertension. Consistent with these observations, this study demonstrated that higher within‐visit BP changes were significantly associated with an increased risk of developing hypertension. These findings suggest that within‐visit BP changes may represent a simple and practical surrogate marker for identifying individuals at higher risk of progressing to sustained hypertension. Because hypertension is a major modifiable risk factor for cardiovascular disease and mortality, 2 , 31 individuals with greater BP changes may also face an increased long‐term cardiovascular risk. From a clinical and preventive perspective, within‐visit BP changes may help identify high‐risk individuals earlier and facilitate timely strategies to prevent hypertension and future adverse cardiovascular events.
The recently revised European guidelines for hypertension introduced a new BP category termed elevated BP, beyond the conventional definition of hypertension. This reflects growing recognition that individuals with systolic BP in the range of 120 to 139 mm Hg may still carry substantial cardiovascular risk. The guideline suggests pharmacological treatment in individuals with elevated BP and high cardiovascular risk, based on risk stratification. However, from a screening perspective, only annual opportunistic BP screening is recommended for those with nonelevated BP or elevated BP without cardiovascular risk. 7 In our BP stratified analysis, we found that even among individuals with nonelevated BP, those with elevated within‐visit BP changes were at an increased risk of incident hypertension, cardiovascular outcomes, and all‐cause mortality. These findings suggest that such individuals may warrant consideration for more intensive surveillance than the current guideline recommendation of annual screening. Incorporating BP changes into risk stratification may help identify high‐risk individuals who could potentially benefit from closer follow‐up or earlier intervention, thereby reducing long‐term adverse clinical outcomes. Further research is needed to determine whether early intervention in patients with nonelevated BP but elevated within‐visit BP changes can reduce long‐term cardiovascular risk.
Limitations
This study has several limitations. First, within‐visit BP changes were defined as the absolute difference between 2 consecutive measurements rather than a conventional variability index (such as SD or coefficient of variation) derived from multiple readings, including the 3 measurements recommended by current guidelines. 7 , 32 Although this approach may be less accurate in terms of variability, it offers practical advantages in clinical implementation, including simplicity, accessibility, and the potential to reduce dilutional bias. Additionally, in such a large‐scale epidemiological study, BP measurements may not have always followed fully standardized procedures regarding rest time, posture, cuff placement, and environmental control. As a result, our measure of BP changes may capture both true BP instability and transient fluctuations including white‐coat effects or alarm reactions during the visit. These methodological considerations should be acknowledged when interpreting the clinical significance of within‐visit BP changes and comparing our results with studies using different BP variability metrics or measurement protocols. Second, given the observational nature of this study, residual confounding cannot be fully excluded despite adjustment for multiple known confounders. Third, the UK Biobank cohort may be subject to selection bias, such as the healthy volunteer effect, and therefore may not fully represent the national population. Although this cohort has limited generalizability, the relationships observed between exposures and outcomes may still be applicable to broader populations. 32 Fourth, some variables in the UK Biobank data set were self‐reported, which may have led to underreporting or misclassification. Lastly, data on out‐of‐office BP and BP changes (eg, home BP monitoring or 24‐hour ambulatory BP monitoring) were not available in the UK Biobank data set. Therefore, it was not possible to determine whether increased within‐visit BP changes reflects true daily‐life BP variability or is substantially influenced by transient factors such as the white‐coat effect. Because only office BP was collected, we were unable to apply the diagnostic approach recommended by current guidelines, which require confirmation with out‐of‐office BP. Consequently, some misclassification, including the potential inclusion of individuals with white‐coat hypertension or undiagnosed hypertension at baseline, cannot be excluded.
Conclusions
In conclusion, in this large cohort study, higher within‐visit BP changes were independently associated with an increased incidence of all‐cause mortality, as well as an elevated risk of subsequent major cardiovascular events and new‐onset hypertension. These associations remained consistent across fully adjusted models and showed a clear linear trend in spline analysis. Similar findings with diastolic BP changes further reinforce the robustness and generalizability of our results. These findings highlight the clinical relevance of within‐visit BP changes, suggesting that even the difference between 2 consecutive BP measurements taken during a single visit may serve as an early and accessible marker for long‐term cardiovascular risk.
Sources of Funding
This research was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute, funded by the Ministry of Health and Welfare, Republic of Korea (grant number RS‐2023‐KH135442).
Disclosures
None.
Supporting information
Tables S1–S7
Figures S1–S6
Acknowledgments
Conceptualization: SungA Bae; Data curation: In Tae Jin, Seok‐Jae Heo, Jin Young Park, SungA Bae; Formal analysis: Seok‐Jae Heo; Funding acquisition: Jin Young Park, SungA Bae; Investigation: In Tae Jin, SungA Bae; Methodology: In Tae Jin, Seok‐Jae Heo, SungA Bae; Project administration: In Tae Jin, Seok‐Jae Heo, SungA Bae; Resources: Seok‐Jae Heo, Jin Young Park, SungA Bae; Software: SungA Bae; Supervision: SungA Bae; Validation: In‐Tae Jin, SungA Bae; Writing—original draft: In Tae Jin; Writing—review and editing: Seok‐Jae Heo, Jin Young Park, SungA Bae.
This article was sent to Michelle H. Leppert, MD, MBA, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.046365
For Sources of Funding and Disclosures, see page 11.
REFERENCES
- 1. World Health Organization . Hypertension. Geneva: World Health Organization; 2023. Updated September 7, 2023. https://www.who.int/news‐room/fact‐sheets/detail/hypertension [Google Scholar]
- 2. Collaborators GBDRF . Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease study 2019. Lancet. 2020;396:1223–1249. doi: 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kannel WB. Blood pressure as a cardiovascular risk factor: prevention and treatment. JAMA. 1996;275:1571–1576. doi: 10.1001/jama.1996.03530440051036 [DOI] [PubMed] [Google Scholar]
- 4. Flint AC, Conell C, Ren X, Banki NM, Chan SL, Rao VA, Melles RB, Bhatt DL. Effect of systolic and diastolic blood pressure on cardiovascular outcomes. N Engl J Med. 2019;381:243–251. doi: 10.1056/NEJMoa1803180 [DOI] [PubMed] [Google Scholar]
- 5. World Health Organization . HEARTS: Technical Package for Cardiovascular Disease Management in Primary Health Care. Geneva: World Health Organization; 2018. Updated 2018. https://www.who.int/publications/i/item/hearts‐technical‐package [Google Scholar]
- 6. World Health Organization . Global Action Plan for the Prevention and Control of Noncommunicable Diseases 2013–2020. Geneva: World Health Organization; 2013. [Google Scholar]
- 7. McEvoy JW, McCarthy CP, Bruno RM, Brouwers S, Canavan MD, Ceconi C, et al. 2024 ESC guidelines for the management of elevated blood pressure and hypertension. Eur Heart J. 2024;45:3912–4018. doi: 10.1093/eurheartj/ehae178 [DOI] [PubMed] [Google Scholar]
- 8. Unger T, Borghi C, Charchar F, Khan NA, Poulter NR, Prabhakaran D, Ramirez A, Schlaich M, Stergiou GS, Tomaszewski M, et al. 2020 International Society of Hypertension global hypertension practice guidelines. J Hypertens. 2020;38:982–1004. doi: 10.1097/HJH.0000000000002453 [DOI] [PubMed] [Google Scholar]
- 9. Muntner P, Whittle J, Lynch AI, Colantonio LD, Simpson LM, Einhorn PT, Levitan EB, Whelton PK, Cushman WC, Louis GT, et al. Visit‐to‐visit variability of blood pressure and coronary heart disease, stroke, heart failure, and mortality: a cohort study. Ann Intern Med. 2015;163:329–338. doi: 10.7326/M14-2803 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Gosmanova EO, Mikkelsen MK, Molnar MZ, Lu JL, Yessayan LT, Kalantar‐Zadeh K, Kovesdy CP. Association of systolic blood pressure variability with Mortality, coronary heart disease, stroke, and renal disease. J Am Coll Cardiol. 2016;68:1375–1386. doi: 10.1016/j.jacc.2016.06.054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Stevens SL, Wood S, Koshiaris C, Law K, Glasziou P, Stevens RJ, McManus RJ. Blood pressure variability and cardiovascular disease: systematic review and meta‐analysis. BMJ. 2016;354:i4098. doi: 10.1136/bmj.i4098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Parati G, Ochoa JE, Lombardi C, Bilo G. Assessment and management of blood‐pressure variability. Nat Rev Cardiol. 2013;10:143–155. doi: 10.1038/nrcardio.2013.1 [DOI] [PubMed] [Google Scholar]
- 13. Verdecchia P, Angeli F, Gattobigio R, Rapicetta C, Reboldi G. Impact of blood pressure variability on cardiac and cerebrovascular complications in hypertension. Am J Hypertens. 2007;20:154–161. doi: 10.1016/j.amjhyper.2006.07.017 [DOI] [PubMed] [Google Scholar]
- 14. Fagard RH, Celis H, Thijs L, Staessen JA, Clement DL, De Buyzere ML, et al. Daytime and nighttime blood pressure as predictors of death and cause‐specific cardiovascular events in hypertension. Hypertension. 2008;51:55–61. doi: 10.1161/HYPERTENSIONAHA.107.100727 [DOI] [PubMed] [Google Scholar]
- 15. Kikuya M, Hozawa A, Ohokubo T, Tsuji I, Michimata M, Matsubara M, Ota M, Nagai K, Araki T, Satoh H, et al. Prognostic significance of blood pressure and heart rate variabilities: the Ohasama study. Hypertension. 2000;36:901–906. doi: 10.1161/01.hyp.36.5.901 [DOI] [PubMed] [Google Scholar]
- 16. Kario K, Tomitani N, Fujiwara T, Okawara Y, Kanegae H, Hoshide S. Peak home blood pressure as an earlier and strong novel risk factor for stroke: the practitioner‐based nationwide J‐HOP study extended. Hypertens Res. 2023;46:2113–2123. doi: 10.1038/s41440-023-01297-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Grassi G, Seravalle G, Maloberti A, Facchetti R, Cuspidi C, Bombelli M, Laurent S, Redon J, Mancia G. Within‐visit BP variability, cardiovascular risk factors, and BP control in central and eastern Europe: findings from the BP‐CARE study. J Hypertens. 2015;33:2250–2256. doi: 10.1097/HJH.0000000000000700 [DOI] [PubMed] [Google Scholar]
- 18. Schutte R, Thijs L, Liu YP, Asayama K, Jin Y, Odili A, Gu YM, Kuznetsova T, Jacobs L, Staessen JA. Within‐subject blood pressure level—not variability—predicts fatal and nonfatal outcomes in a general population. Hypertension. 2012;60:1138–1147. doi: 10.1161/HYPERTENSIONAHA.112.202143 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Zarife AS, Fraga‐Maia H, Mill JG, Lotufo P, Griep RH, Fonseca M, Brito LL, Almeida MDC, Aras R, Matos SMA. Within‐visit blood pressure variability and cardiovascular risk in ELSA‐Brasil study participants. Arq Bras Cardiol. 2022;119:505–511. doi: 10.36660/abc.20210804 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Allen NE, Lacey B, Lawlor DA, Pell JP, Gallacher J, Smeeth L, et al. Prospective study design and data analysis in UK Biobank. Sci Transl Med. 2024;16:eadf4428. doi: 10.1126/scitranslmed.adf4428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, Downey P, Elliott P, Green J, Landray M, et al. UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12:e1001779. doi: 10.1371/journal.pmed.1001779 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Bull FC, Al‐Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54:1451–1462. doi: 10.1136/bjsports-2020-102955 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Elliott P, Peakman TC; Biobank UK . The UK biobank sample handling and storage protocol for the collection, processing and archiving of human blood and urine. Int J Epidemiol. 2008;37:234–244. doi: 10.1093/ije/dym276 [DOI] [PubMed] [Google Scholar]
- 24. Biobank UK . Blood Pressure, Version 1.0 . 2011. https://biobank.ctsu.ox.ac.uk/crystal/crystal/docs/Bloodpressure.pdf.
- 25. Biobank UK . Mortality Data: Linkage to Death Registries. Version 3.0 . 2023. https://biobank.ndph.ox.ac.uk/showcase/ukb/docs/DeathLinkage.pdf.
- 26. Biobank UK . Algorithmically Defined Outcomes (ADOs), Version 2.0 . 2022. https://biobank.ndph.ox.ac.uk/showcase/ukb/docs/alg_outcome_main.pdf.
- 27. Anisetti B, Youssef H, Elkhair AM, Lin MP. Association between within‐visit blood pressure variability, stroke, coronary heart disease, and cardiovascular mortality. Neurol Neurochir Pol. 2023;57:423–429. doi: 10.5603/pjnns.94839 [DOI] [PubMed] [Google Scholar]
- 28. Mancia G, Grassi G. Mechanisms and clinical implications of blood pressure variability. J Cardiovasc Pharmacol. 2000;35:S15–S19. doi: 10.1097/00005344-200000004-00003 [DOI] [PubMed] [Google Scholar]
- 29. Parati G, Torlasco C, Pengo M, Bilo G, Ochoa JE. Blood pressure variability: its relevance for cardiovascular homeostasis and cardiovascular diseases. Hypertens Res. 2020;43:609–620. doi: 10.1038/s41440-020-0421-5 [DOI] [PubMed] [Google Scholar]
- 30. Grassi G, Seravalle G, Quarti‐Trevano F, Dell'Oro R, Bombelli M, Cuspidi C, Facchetti R, Bolla G, Mancia G. Adrenergic, metabolic, and reflex abnormalities in reverse and extreme dipper hypertensives. Hypertension. 2008;52:925–931. doi: 10.1161/HYPERTENSIONAHA.108.116368 [DOI] [PubMed] [Google Scholar]
- 31. Mensah GA, Fuster V, Murray CJL, Roth GA; Global Burden of Cardiovascular D, Risks C . Global burden of cardiovascular diseases and risks, 1990–2022. J Am Coll Cardiol. 2023;82:2350–2473. doi: 10.1016/j.jacc.2023.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, Collins R, Allen NE. Comparison of sociodemographic and health‐related characteristics of UK Biobank participants with those of the general population. Am J Epidemiol. 2017;186:1026–1034. doi: 10.1093/aje/kwx246 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S7
Figures S1–S6