The Association Between Habitual Sleep Duration and Hypertension Control in the United States (US) Adults With Hypertension

Although sleep duration has been identied as a signicant factor in risk for hypertension, there is limited data on the relationship between sleep duration and hypertension control. This study examined the association between habitual sleep duration and hypertension control in United States of America (USA) adults with hypertension. A total of 5,163 adults from the National Health and Nutrition Examination Survey (2015 – 2018) were analyzed. Survey-weighted multivariable logistic regression models were t to examine the association between habitual sleep duration (coded as <6, 6 - <7, 7 – 9 (reference), and >9 hours) and hypertension control (BP <130/80mmHg versus ≥ 130/80mmHg), adjusted for sociodemographic, sleep and health characteristics. In the fully adjusted model, habitual sleep duration of <6 hours/main sleep period was associated with reduced odds of hypertension control (OR = 0.66, 95% CI: 0.46 – 0.95, P = 0.027) when compared to 7 – 9 hours. No signicant differences were noted in hypertension control between the reference group (7 - 9 hours) and the 6 - <7 or >9 hours groups. These ndings suggest that measures to support adequate habitual sleep duration may help improve hypertension control in adults who habitually sleep for <6 hours/day. NHANES, SE, error;


Introduction
In the 2015-2018 period, 47.3% of US adults had hypertension, de ned as systolic blood pressure (SBP) ≥ 130 mmHg or diastolic blood pressure (DBP) ≥ 80 mmHg, or current intake of BP-lowering medications 1 . Of those, only 20.6% had achieved hypertension control 1 . Hypertension is a major modi able risk factor for heart disease and stroke 2 . In 2018, heart disease and stroke were among the top 5 causes of death in the United States of America (USA) 3 . Insu cient sleep is also a signi cant health issue, with over one-third of US adults sleeping < 7 hours per day 4

. The American Academy of Sleep
Medicine and Sleep Research Society (AASM/SRS) recommends a minimum of 7 hours daily of sleep to support optimal function of body systems, including the systems involved in BP regulation 5 .
The 2017 American College of Cardiology and American Heart Association (ACC/AHA) guidelines on nonpharmacological interventions for prevention and management of hypertension focus on well-known modi able risk factors for hypertension such as obesity, unhealthy diet, and excessive alcohol intake 2,6,7 .
However, several observational studies have also found a link between insu cient sleep and the risk of hypertension [8][9][10][11][12][13][14] . These ndings are supported by experimental studies, which have demonstrated a link between sleep restriction and elevation of BP [15][16][17][18] . Other ndings indicate that the strength of the relationship between short sleep and hypertension risk decreases with age 9,11,19 , and is stronger in women than men 9,20 .
The growing evidence showing that sleep duration is a signi cant predictor of hypertension risk points to the need to examine the potential role of sleep duration in managing and controlling hypertension. Little is known, however, about the relationship between habitual sleep duration and hypertension control. A few studies that have examined the relationship between sleep duration and BP outcomes in adults with hypertension have yielded mixed results. One of the studies demonstrated a signi cant BP increase when sleep was restricted to < 5 hours a day 15 . The other, a cross-sectional study, found a positive correlation between self-reported long sleep duration (≥ 10 hours/day) and SBP 21 .
Further research is needed to explore if any differences noted in BP across various sleep duration categories translate to differences in hypertension control. We addressed this gap in knowledge by examining the association between habitual sleep duration and hypertension control in adults with hypertension. We also examined whether the relationship between sleep duration and hypertension control was modi ed by age or gender.

Data Source and study population
This cross-sectional study used data from the 2015-2016 and 2017-2018 cycles of the National Health and Nutrition Examination Survey (NHANES). NHANES employs complex multistage sampling procedures to survey the civilian, noninstitutionalized USA population. The participants are interviewed at home, followed by physical examinations and further interviews at a mobile examination center (MEC) 22 .
The NHANES protocol was approved by the National Center for Health Statistics Ethics Review Board 23 .
Informed consents were obtained for all eligible subjects before they participated in the NHANES interviews and physical examinations. All methods were performed in accordance with relevant guidelines and regulations of the NHANES. We used de-identi ed and publicly available data that did not meet human subjects research criteria and thus was exempt from institutional review board oversight.
The datasets used are available at https://wwwn.cdc.gov/nchs/nhanes/. Figure 1 illustrates the criteria used for identifying the study sample. Of the 19,225 adults (≥ 18 years old) in the 2015-2018 NHANES 24 , 5,712 participated in interviews and physical examinations and met at least one of the following criteria for hypertension: current use of BP-lowering medication (n = 3,706), average SBP ≥ 130mmHg (n = 3,744), or SBP ≥ 80mmHg (n = 2,323) 2 . We excluded participants who were pregnant or missing sleep duration or BP data. To minimize the potential for unmeasured confounding 25 , we excluded those with severe disease, including congestive heart failure and severe (stage 4-5) chronic kidney disease (de ned as a history of undergoing dialysis in the previous 12 months or a Chronic Kidney Disease Epidemiology Collaboration estimated glomerular ltration rate < 30 mL/min/1.73 m 2 ) 26, 27,28 .

Key Measures
Habitual Sleep Duration. The NHANES calculated the amount of sleep usually obtained in a night or main sleep period during weekdays or workdays from two survey questions: "What time do you usually fall asleep on weekdays or workdays?" and "What time do you usually wake up on weekdays or workdays?" We categorized habitual sleep duration into < 6, 6 -<7, 7-9, and > 9 hours/night or main sleep period 5,11,12 . Hypertension Control. A standardized protocol was used to obtain three BP readings taken one minute apart at the MEC 29 . Hypertension control was de ned as average SBP < 130mmHg and DBP < 80mmHg 2 .
History of sleep apnea symptoms was de ned as a history of; (1) snoring ≥ 3 times/week; (2) snorting, gasping, or stopping breathing while sleeping ≥ 3 times/week; or (3) being excessively sleepy during the day ≥ 16 times/month despite sleeping for ≥ 7 hours per night 37,38 . Help-seeking for sleeping di culty was de ned as a history of ever telling a health care professional that one had trouble sleeping. Depressive symptoms were screened using the 9-item Patient Health Questionnaire (PHQ-9, range of 0-27) and categorized as minimal or none (0-4), mild (5-9), moderate to severe (≥ 10-14) 39 .
The number of healthcare visits in the past 12 months (excluding home visits, phone consultations, overnight hospitalization, and emergency room visits) was grouped into none, 1-2, and > 2 visits. Body mass index (BMI) was analyzed as a continuous variable. Cardiovascular disease was de ned as a history of being told by a health professional that one had coronary heart disease, angina, a heart attack, or stroke. Diabetes was de ned based on either a history of being told by a health care professional that one has diabetes or having a blood glycohemoglobin level of 6.5% or higher 40 . Moderate chronic kidney disease was de ned as an eGFR of 30 -<60 mL/min/1.73 m 2 26 .
Cigarette smoking was categorized as never smoker (never smoked at least 100 cigarettes in their lifetime), former smoker (smoked at least 100 cigarettes in their lifetime but not smoking currently), and current smoker. Physical activity was self-reported and included leisure, work, and transportation (commuting by walking or bicycling) activities. Moderate-intensity and transportation-related physical activities were assigned four metabolic equivalents of task (MET) scores/minute, and vigorous-intensity physical activity eight MET scores/minute 41 . Weekly physical activity levels were categorized as none (0 MET-minutes), low (< 600 MET-minutes), su cient (600-1200 MET-minutes), and high (> 1200 METminutes) 42 . Alcohol intake was classi ed as none (never had at least 12 alcoholic drinks in a lifetime or any alcohol in the past year), moderate (not more than one drink for women and not more than two drinks for men in a day), and heavy (more than one or two drinks in a day for women and men, respectively) 43 .

Statistical Analysis & Missing Data
Data distribution across various variables was analyzed for the total study sample and across habitual sleep duration categories. The percentage of observations with complete data for all covariates was 85%. The covariates with missing data (unweighted number and weighted percentage) included education level (n = 6, 0.1%), annual household income (n = 413, 6.0%), health insurance (n = 10, 0.2%), alcohol intake (n = 333, 5.3%), BMI (n = 67, 1.0%), depressive symptoms (n = 357, 5.6%), and healthcare visits in past year (n = 10, 0.1%). Missing data were imputed using multiple imputation with chained equations 44 . A total of 20 imputation datasets were generated, and the results pooled to generate estimates of the multiply impute model using STATA IC's multiple imputation (MI) estimate procedures 45 .
The variables used in the imputation model included all the variables in the analysis model, the NHANES cluster, strata, and weights variables, and auxiliary variables signi cantly associated with missing data in some covariates (age and education level of the reference person, homeownership, number of rooms in house of residence, and household size).
We compared the observed data to the complete imputed data to check for differences in data distribution in the covariates with missing data. The distribution remained similar for depressive symptoms, BMI, healthcare visits in the past year, education level, health insurance, and alcohol intake. In the annual household income, the proportion of the < $55,000 group increased from 48.5-49.3% while the ≥ $55,000 group reduced from 51.5-50.7% after imputation (Appendix Table 1).
Models to analyze the association between habitual sleep duration and hypertension control were t using the complete imputed data. Logistic regression models (unadjusted and adjusted) were t to analyze the association between habitual sleep duration and hypertension control, with the crude and adjusted odds ratios for hypertension control and their corresponding 95% con dence intervals presented. Effect modi cation by age and gender was assessed separately by adding an interaction term (sleep duration × age or sleep duration × gender) to the adjusted logistic regression model. All data were analyzed using STATA 1C software (Version 15, StataCorp LLC, College Station, Texas, 2017). Survey commands were used to apply sample weights to account for the NHANES complex sampling design.
The level of signi cance for all analyses was set at a p-value < 0.05.

Results
The current study included 5,163 study participants (see Fig. 1). Table 1 presents the descriptive characteristics of the participants by habitual sleep duration based on the 2015-2018 NHANES data. The average age of the adults with hypertension was 55.4 years, and 52.2% were male. Most participants (64.3%) were non-Hispanic White. More than half (53.2%) had some college education or higher level of education. Most of the participants had health insurance and had visited a health care facility for care at least once in the past year. The mean BMI was 31.2 kg/m 2 , and 49.3% reported being highly physically active (> 1200 MET minutes/week) while 25.1% were physically inactive.
Over half (54.3%) had sleep apnea symptoms, and 36.2% had a history of telling a health care professional that they had trouble sleeping. The majority (66.0%) reported sleeping 7-9 hours in a night or main sleep period, while 23.7% slept < 7 hours. The proportion that had hypertension control (SBP < 130mmHg and DBP < 80mmHg) was 19.7% (Table 1).
In the unadjusted logistic regression model of hypertension control as a function of habitual sleep duration (Table 2), those with a sleep duration of < 6 hours were less likely to have hypertension control when compared to those whose sleep duration was 7-9 hours (OR = 0.65, 95% CI: 0.44-0.98, P = 0.041 ). No signi cant differences were noted in hypertension control between the sleep duration reference group (7-9 hours) and those whose sleep duration was 6 -<7 hours or > 9 hours. After adjusting for all covariates (sociodemographic characteristics and other sleep and health-related characteristics), the nding that those with a sleep duration of < 6 hours were less likely to have hypertension control than those with a sleep duration of 7-9 hours remained robust (OR = 0.66, 95% CI: 0.46-0.95, P = 0.027). In the fully adjusted model, no signi cant differences were noted in hypertension control between those with 6 - In other ndings from the adjusted model, those aged 40-79 years had higher odds of hypertension control than the 18-39 age group. Those with a history of cardiovascular disease, moderate chronic kidney disease, or diabetes mellitus had higher odds of hypertension control. Other factors positively associated with higher odds of hypertension control included BMI, current smoking (compared to no smoking), having a history of seeking help for sleeping di culties, visiting a healthcare facility for care at least once in the past year, and having a household income ≥ $55,000 (compared to < $55,000) ( Table 2).
In the fully adjusted logistic model with the interaction term, habitual sleep duration × age, no signi cant interactions were noted between habitual sleep duration and age on the odds of hypertension control. A separate fully adjusted model with the interaction term habitual sleep duration × gender also yielded no signi cant interactions.
In sensitivity analyses, the multivariable logistic regression model results from the complete imputed data (n = 5,163) were compared to those from the observed data, n = 4,384) to note any differences. The ndings on odds for hypertension control in those with a habitual sleep duration of < 6 hours when using complete imputed data (OR = 0.66, 95% CI: 0.46-0.95) were comparable to the observed data results (OR = 0.65, 95% CI: 0.44-0.98). In both models, no signi cant differences were noted in hypertension control between the reference group (7-9 hours) and the 6-<7 hours or > 9 hours group (Appendix Table 2).

Discussion
In this nationally representative study of US adults with hypertension, we nd a negative association between short sleep duration and hypertension control. This nding builds upon evidence from previous studies showing short sleep duration to be a signi cant risk factor for hypertension 8-13 . Importantly, the association was robust, with no age or gender differences in magnitude or signi cance in the relationship between sleep duration and hypertension control.
Our ndings of an association between short sleep duration and hypertension control at < 6 hours of sleep but not 6-<7 hours support prior ndings that noted a stronger association between short sleep and hypertension at lower hours of sleep. In a cross-sectional study that used 7-8 hours/night as the normal sleep duration, self-reported sleep duration of 5-6 hours was associated with 22% increased risk for hypertension, and the risk substantially increased by twofold in those with a sleep duration of <5 hours/night 46 . These ndings are consistent with another study, which reported a 86%, 56%, and 27% increase in the risk of hypertension in adults whose sleep duration was ≤4, 5, and 6 hours, respectively, compared to 7 hours of sleep 9 . A dose-response relationship has been demonstrated in a meta-analysis of longitudinal studies, which reported a 0.32% reduction in risk for hypertension with each hour increase in sleep duration. The meta-analysis ndings also showed that a sleep duration of ≤5 hours was associated with a higher risk for hypertension than a sleep duration of 6 hours (p <0.05) compared to 7 hours of sleep 8 . These ndings, taken together, point to a higher risk of poor BP outcomes in adults habitually sleeping for less than 6 hours a day.
There are several ways through which inadequate sleep may negatively impact BP and hypertension control. Sleeping for less than the recommended hours can cause alterations in physiological functions, leading to many adverse effects, including an increase in BP. Examples of these alterations include increased sympathetic nervous system (SNS) activity, reduced insulin sensitivity, endothelial dysfunction, and hormonal alterations that increase the risk for obesity [47][48][49][50][51] . For instance, experimental studies in adults with normal BP and those with hypertension have shown that restricting sleep to less than 5 hours/day leads to a signi cant increase in SNS activity and elevation in BP [15][16][17][18] . Insu cient sleep is thought to lead to a sustained elevation in SNS activity throughout the time one remains awake during the night and throughout the next day following sleep loss 49 . Inadequate sleep can also interfere with speci c neurocognitive functions and impair attention to health-related cues such as being physically active and eating healthy food 52 . All these factors can contribute to di culty achieving hypertension control among individuals with hypertension.
The 2017 ACC/AHA guidelines outline pharmacological and non-pharmacological interventions for hypertension management to improve overall health status and reduce cardiovascular disease risk 2 . The main non-pharmacological interventions recommended are based on ndings from clinical trials. These interventions include weight control, physical activity, a healthy diet, increased dietary potassium, reduced dietary sodium, and reduced alcohol intake for those who consume alcohol 2 . These interventions are focused on well-known modi able risk factors for hypertension such as obesity, unhealthy diet, and excessive alcohol intake 2,6,7 , Based on the ndings from our study and others [8][9][10][11][12][13]15,16,18 , habitual sleep duration may play a crucial role in the prevention, management, and control of hypertension. Consequently, it should be one of the recommended basic lifestyle interventions for BP control in future iterations of the hypertension guidelines.
There are a few limitations related to this study. We cannot infer a causal link between habitual sleep duration and hypertension control because of the cross-sectional nature of the NHANES data. There may be response bias because data for several variables, including habitual sleep duration, other sleep characteristics, some chronic health conditions, and health behaviors, were self-reported. Although the measurement of BP was done by specially trained health professionals that followed a standardized protocol, the BP readings used to calculate the average BP were all obtained in a single visit to the mobile examination center. Using BP readings obtained only in a single visit can lead to misclassi cation of hypertension status or hypertension control status because of the di culty of identifying whitecoat and

Competing interests
The authors declare that they have no competing interests.