Estimated Prevalence of Masked Asleep Hypertension in US Adults

Key Points

Question  What is the national prevalence of masked asleep hypertension among US adults?

Findings  This cohort study combined data from 3000 participants with 24-hour ambulatory blood pressure monitoring and data from 17 969 participants in the 2011-2016 National Health and Nutrition Examination Survey without ambulatory blood pressure monitoring. An estimated 22.7% (53.7 million) of US adults have masked asleep hypertension and 13.3% (31.5 million) have isolated masked asleep hypertension, using 2017 American College of Cardiology–American Heart Association guideline blood pressure thresholds.

Meaning  These findings suggest that the prevalence of masked asleep hypertension is high among US adults; data are needed on the cardiovascular risk reduction benefits of treating asleep hypertension.

Importance  High blood pressure (BP) during sleep (asleep blood pressure) is associated with an increased risk of cardiovascular disease, but a national prevalence estimate of masked asleep hypertension (high BP while sleeping but without high BP measured in the clinic [clinic BP]) for the United States is lacking.

Objectives  To estimate the prevalence of masked asleep hypertension among US adults by using BP thresholds from the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC7) and the 2017 American College of Cardiology–American Heart Association (ACC-AHA) BP guidelines.

Design, Setting, and Participants  This cohort analysis pooled data from 3000 participants in 4 US population-based studies that conducted 24-hour ambulatory BP monitoring (ABPM) and 17 969 participants in the 2011-2016 National Health and Nutrition Examination Survey (NHANES) without ABPM. Masked asleep hypertension status in NHANES was imputed using a 2-stage multiple imputation process. Data were collected from 2000 to 2016 and analyzed from March 4, 2019, to June 29, 2020.

Main Outcomes and Measures  High clinic BP was defined as clinic systolic BP (SBP)/diastolic BP (DBP) of at least 140/90 mm Hg using JNC7 and at least 130/80 mm Hg using 2017 ACC-AHA guidelines. High asleep BP was defined as mean asleep SBP/DBP of at least 120/70 mm Hg for JNC7 and at least 110/65 mm Hg for the 2017 ACC-AHA guidelines. Masked asleep hypertension was defined as high asleep BP without high clinic BP.

Results  For the 3000 pooled cohort participants, the mean (SD) age was 52.0 (12.0) years, and 62.6% were women. For the 17 969 NHANES participants, the mean (SD) age was 46.7 (17.5) years, and 51.8% (weighted) were women. The estimated prevalence of masked asleep hypertension among US adults was 18.8% (95% CI, 16.7%-20.8%; 44.4 million US adults) using the JNC7 guideline and 22.7% (95% CI, 20.6%-24.8%; 53.7 million US adults) using the 2017 ACC-AHA guideline criteria. The prevalence of masked asleep hypertension was higher among older adults (aged ≥65 years, 24.4% [95% CI, 20.7%-28.0%]), men (27.0% [95% CI, 24.1%-29.9%]), non-Hispanic Black individuals (28.7% [95% CI, 25.4%-32.0%]), those who were taking antihypertensives (24.4% [95% CI, 21.1%-27.8%]), those who had masked daytime hypertension (44.7% [95% CI, 40.1%-49.3%]), and those with diabetes (27.6% [95% CI, 23.5%-31.8%]), obesity (24.3% [95% CI, 21.8%-26.9%]), or chronic kidney disease (21.5% [95% CI, 17.3%-25.6%]) using the 2017 ACC-AHA guideline. An estimated 11.9% of US adults (28.2 million) had isolated masked asleep hypertension (masked asleep hypertension but without high awake BP) using JNC7 guideline criteria, as did an estimated 13.3% (31.5 million) using 2017 ACC-AHA guideline criteria.

Conclusions and Relevance  These findings suggest that the prevalence of masked asleep hypertension is high among US adults. Data are needed on the cardiovascular risk reduction benefits of treating asleep hypertension.

Ambulatory blood pressure monitoring (ABPM) is recommended by the 2017 American College of Cardiology–American Heart Association (ACC-AHA) blood pressure (BP) guideline to confirm a diagnosis of hypertension based on BP measurements obtained in the clinic (clinic BP).¹ Ambulatory BP monitoring can also identify BP phenotypes not detected using clinic measurements alone, including asleep hypertension.¹ Asleep hypertension, defined as high BP while sleeping, with or without high clinic BP, is associated with an increased risk of cardiovascular disease.^2-4 However, most large US population-based cohorts and US national health surveys have not performed ABPM, and a national estimate for asleep hypertension is currently lacking.⁵

The primary objective of this study was to estimate the national prevalence of masked asleep hypertension (high asleep BP without high clinic BP) among US adults, defined according to BP thresholds established by the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC7) and the 2017 ACC-AHA BP guidelines.^1,6 We focused on masked asleep hypertension because individuals with this phenotype do not have high clinic BP and will not have their high BP detected by measurements taken in the clinic. We assembled a pooled cohort of participants from the following 4 studies with 24-hour ABPM: (1) Coronary Artery Risk Development in Young Adults,⁷ (2) Jackson Heart Study,⁸ (3) Masked Hypertension Study,⁹ and (4) Improving the Detection of Hypertension Study.¹⁰ We then combined the pooled cohort with participants in the 2011-2016 National Health and Nutrition Examination Survey (NHANES), in whom 24-hour ABPM was not collected, to impute masked asleep hypertension status in US adults.

Details of the design of each study cohort are reported in the eMethods in the Supplement. For the current analysis, we only included individuals from the pooled cohort with 24-hour ABPM data (eFigure in the Supplement). We excluded participants with less than 10 awake or less than 5 asleep BP measurements because they did not meet the International Database of Ambulatory BP in relation to Cardiovascular Outcomes criteria for a complete ABPM.¹¹ The final sample size was 3000 participants. All study protocols were approved by the institutional review boards at participating institutions and all participants provided written informed consent. We also combined data from the 2011-2012, 2013-2014, and 2015-2016 NHANES and included 17 969 NHANES participants at least 18 years of age with clinic systolic BP (SBP) and diastolic BP (DBP) measurements for the analysis. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Details of data collection are reported in the eMethods in the Supplement; data for the present analyses were collected from 2000 to 2016. High clinic BP was defined by JNC7 as mean clinic SBP/DBP of at least 140/90 mm Hg and by the 2017 ACC-AHA BP guideline as mean clinic SBP/DBP of at least 130/80 mm Hg. High awake BP was defined by JNC7 as mean awake SBP/DBP of at least 135/85 mm Hg and by the 2017 ACC-AHA guideline as mean awake SBP/DBP of at least 130/80 mm Hg. The asleep BP threshold according to JNC7 was at least 120/75 mm Hg; however, guidelines and scientific statements have since adopted asleep BP thresholds of at least 120/70 mm Hg to correspond to clinic-measured BP of at least 140/90 mm Hg.^12,13 Therefore, in the present study, high asleep BP was defined as mean asleep SBP/DBP of at least 120/70 mm Hg in the JNC7 analysis and as mean asleep SBP/DBP of at least 110/65 mm Hg in the 2017 ACC-AHA guideline analysis. Masked asleep hypertension was defined as high asleep BP without high clinic BP. Isolated masked asleep hypertension was defined as high asleep BP without high clinic BP or high awake BP. Persistent high BP was defined as high asleep BP with high clinic BP.

Data were analyzed from March 4, 2019, to June 29, 2020. We used a 2-stage multiple imputation process to estimate the prevalence of asleep hypertension in US adults (eMethods in the Supplement). To validate our imputation approach, we performed 2 separate bootstrap cross-validations (eMethods in the Supplement). All analyses were performed using R, version 3.5.1 (R Project for Statistical Computing).

In the pooled cohort of 3000 participants with ABPM data, the mean (SD) age was 52.0 (12.0) years; 1122 (37.4%) were men and 1878 (62.6%) were women; 900 (30.0%) were non-Hispanic White (Table 1). Participant characteristics for each of the 4 studies in the pooled cohort are shown in eTable 1 in the Supplement. Among the 17 969 participants in the NHANES cohort, mean (SD) age was 46.7 (17.5) years; 48.2% were men and 51.8% were women (weighted); and 64.9% were non-Hispanic White. Compared with the pooled cohort participants, NHANES participants were younger and more likely to be male and non-Hispanic White (Table 1).

The estimated prevalence of masked asleep hypertension among US adults was 18.8% (95% CI, 16.7%-20.8%) using the JNC7 BP thresholds and 22.7% (95% CI, 20.6%-24.8%) using the 2017 ACC-AHA BP thresholds (Table 2 and Figure). This corresponds to an estimated 44.4 (95% CI, 39.1-49.7) million and 53.7 (95% CI, 47.9-59.6) million US adults with masked asleep hypertension using the respective guidelines. The prevalence of masked asleep hypertension was higher among older adults (aged ≥65 years, 24.4% [95% CI, 20.7%-28.0%]), men (27.0% [95% CI, 24.1%-29.9%]), non-Hispanic Black individuals (28.7% [95% CI, 25.4%-32.0%]), those who were taking antihypertensives (24.4% [95% CI, 21.1%-27.8%]), those who had masked daytime hypertension (44.7% [95% CI, 40.1%-49.3%]), and those with diabetes (27.6% [95% CI, 23.5%-31.8%]), obesity (24.3% [95% CI, 21.8%-26.9%]), or chronic kidney disease (21.5% [95% CI, 17.3%-25.6%]) using the 2017 ACC-AHA guideline. In addition, we estimated that 11.9% (28.2 million) of US adults had isolated masked asleep hypertension according to JNC7 guidelines, and 13.3% (31.5 million) had it according to 2017 ACC-AHA BP guideline criteria.

In the 2 validation analyses, the mean difference between the imputed and observed prevalence of masked asleep hypertension ranged from −0.8% (95% CI, −6.8% to 5.3%) to 0.9% (95% CI, −4.3% to 6.1%) (eTable 2 in the Supplement). In addition, the estimated prevalence of persistent high BP among US adults was 10.6% (95% CI, 9.7%-11.6%; 25.2 [95% CI, 22.6-27.7] million US adults) and 25.6% (95% CI, 24.2%-27.0%; 60.5 [95% CI, 55.5-65.6] million US adults) using the JNC7 and 2017 ACC-AHA BP guideline thresholds, respectively (eTable 3 in the Supplement).

In this study, we estimated that the prevalence of masked asleep hypertension was 18.8% (44.4 million US adults) using the JNC7 guideline BP thresholds and 22.7% (53.7 million US adults) using the 2017 ACC-AHA guideline BP thresholds. We also estimated that 11.9% (28.2 million) and 13.3% (31.5 million) of US adults had isolated masked asleep hypertension according to JNC7 and 2017 ACC-AHA criteria, respectively. These results underscore the potential importance of using ABPM not only to confirm a diagnosis of hypertension but to identify other high-risk BP phenotypes.

Ambulatory BP measurement allows clinicians to determine an individual’s BP phenotype and provide better estimates of cardiovascular disease risk than clinic BP alone.¹ Both high 24-hour and asleep BPs are associated with an increased risk for all-cause mortality and cardiovascular disease events, even after adjusting for clinic BP.^2,3 Regardless of the BP thresholds used to define it, the high prevalence of masked asleep hypertension suggests it may account for a substantial proportion of cardiovascular disease risk in the United States. Although the US Preventive Services Task Force guideline, 2017 ACC-AHA BP guideline, and other international guidelines recommend using ABPM to diagnose hypertension, screening for asleep hypertension is not currently recommended by the US guidelines.^1,13,14 Evidence regarding the use of antihypertensives to reduce cardiovascular disease events in asleep hypertension is also lacking.

Several limitations of our study should be considered. We assume the association between asleep hypertension and the variables that were present in the smaller pooled cohort was generalizable to the larger NHANES population. The total sample size of participants with 24-hour ABPM data was relatively small to allow reliable estimation of prevalence in some subgroups of interest. Although we included a racially and geographically diverse population, ABPM has yet to be collected in nationally representative US samples.

We estimated that the prevalence of masked asleep hypertension among US adults was 18.8% (44.4 million) using the JNC7 BP thresholds and 22.7% (53.7 million) using the 2017 ACC-AHA BP thresholds. The estimated prevalence of isolated masked asleep hypertension was 11.9% (28.2 million US adults) and 13.3% (31.5 million US adults) using JNC7 and 2017 ACC-AHA criteria, respectively. Future research should investigate the possible cardiovascular risk reduction and economic benefits of treating asleep hypertension.

Accepted for Publication: August 14, 2020.

Corresponding Author: Yiyi Zhang, PhD, Division of General Medicine, Columbia University, 622 W 168th St, Room PH9-109A, New York, NY 10032 (yz3160@cumc.columbia.edu).

Published Online: October 28, 2020. doi:10.1001/jamacardio.2020.5212

Author Contributions: Drs Bellows and Zhang contributed equally as senior authors. Drs Bellows and Zhang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Schwartz, Shimbo, Muntner, Booth, Bress, King, Correa, Moran, Bellows, Zhang.

Acquisition, analysis, or interpretation of data: Li, Schwartz, Shimbo, Muntner, Shikany, Allen, Jaeger, Bress, Clark, Butler, Correa, Bellows, Zhang.

Drafting of the manuscript: Li, Bellows, Zhang.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Li, Schwartz, Jaeger, Bress, Bellows, Zhang.

Obtained funding: Schwartz, Shimbo, Muntner, Correa, Moran.

Administrative, technical, or material support: Li, Shimbo, Muntner, Bress, Butler, Correa, Moran, Bellows.

Supervision: Clark, Moran, Bellows, Zhang.

Conflict of Interest Disclosures: Dr Schwartz reported receiving grants from the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH), during the conduct of the study. Dr Muntner reported receiving research support to his institution from Amgen, Inc, outside the submitted work. Dr Booth reported receiving salary support from CTI Clinical Trials and Consulting Services, Inc. outside the submitted work. Dr Bress reported receiving grants from NHLBI, NIH, during the conduct of the study and grants from Amgen, Inc, and Amarin Corporation outside the submitted work. Dr King reported receiving personal fees from MedStar Health and grants from Novartis International AG and Amgen, Inc, outside the submitted work. Dr Butler reported receiving grants from the NHLBI, NIH, during the conduct of the study. Dr Moran reported receiving grants from the NHLBI, NIH, during the conduct of the study. No other disclosures were reported.

Funding/Support: This study was supported by grant SFRN 15SFRN2390002 from the AHA (principal investigator [PI], Dr Muntner); grants K01HL140170 (Dr Bellows), K01HL133468 (Dr Bress), and R01HL117323 (Dr Muntner) from the NHLBI; grant P01HL047540 from the NHLBI (Masked Hypertension Study and Improving the Detection of Hypertension; PI, Dr Schwartz); grant 19CDA34760232 from the AHA (Dr Clark); and grant 1U54GM115428 from the National Institute of General Medical Sciences (Dr Clark). The Coronary Artery Risk Development in Young Adults (CARDIA) Study is conducted and supported by the NHLBI in collaboration with the University of Alabama at Birmingham (grants HHSN268201800005I and HHSN268201800007I), Northwestern University (grant HHSN268201800003I), University of Minnesota (grant HHSN268201800006I), and the Kaiser Foundation Research Institute (grant HHSN268201800004I). The Jackson Heart Study is supported and conducted in collaboration with Jackson State University (grant HHSN268201800013I), Tougaloo College (grant HHSN268201800014I), the Mississippi State Department of Health (grant HHSN268201800015I), and the University of Mississippi Medical Center (grants HHSN268201800010I, HHSN268201800011I, and HHSN268201800012I) from the NHLBI and the National Institute on Minority Health and Health Disparities.

Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the NHLBI, the NIH, or the US Department of Health and Human Services.

Additional Contributions: This manuscript was reviewed by the CARDIA investigators for scientific content. We thank the investigators, staff, and participants of all 5 cohorts for their valuable contributions.