Factors Associated With Hypertension Control in the General Population of the United States

Background  Uncontrolled hypertension is the most common and important risk factor for cardiovascular and renal disease. We studied factors associated with hypertension control in the Third National Health and Nutrition Examination Survey.

Methods  A total of 3077 non-Hispanic whites, 1742 non-Hispanic blacks, and 1067 Mexican Americans 18 years or older with hypertension were included in the current analysis. Blood pressure was measured by trained observers by means of a standard mercury sphygmomanometer, and controlled hypertension was defined as a mean systolic/diastolic blood pressure less than 140/90 mm Hg.

Results  Percentages of persons with controlled hypertension differed significantly by ethnicity and sex: 19.2% and 28.7% for white men and women, 17.5% and 28.6% for black men and women, and 12.7% and 18.0% for Mexican American men and women, respectively. After adjustment for important covariables, percentages of persons with controlled hypertension were significantly higher among persons who were currently (odds ratio [OR] 2.39; 95% confidence interval [CI], 1.52-3.74) or formerly (OR, 1.81; 95% CI, 1.12-2.93) married, had private health insurance (OR, 1.59; 95% CI, 1.02-2.49), visited the same facility for their health care (OR, 2.77; 95% CI, 1.88-4.09) or saw the same provider for their health care (OR, 2.29; 95% CI, 1.74-3.02), had their blood pressure checked during the preceding 6 months (OR, 8.00; 95% CI, 3.75-17.1) or 6 to 11 months (OR, 5.31; 2.51-11.2), and reported using lifestyle modification to control their hypertension (OR, 6.02; 95% CI, 4.20-8.63).

Conclusion  These data strongly suggest that access to a regular source of health care and modification of lifestyle are important factors in the control of hypertension in the community.

HYPERTENSION IS an important public health challenge in the United States. As many as 50 million Americans have hypertension, defined as a systolic blood pressure (BP) of 140 mm Hg or more and/or diastolic BP of 90 mm Hg or more and/or taking antihypertensive medication.¹ More than $26.1 billion is spent annually for medications, office visits, and laboratory tests related to treatment of hypertension in the United States.² Observational epidemiologic studies have demonstrated that hypertension is associated with an increased risk of coronary heart disease (the leading cause of death in the United States), stroke (the third leading cause of death), congestive heart failure, end-stage renal disease, and peripheral vascular disease.^3-6 Clinical trials have shown that lowering BP reduces the incidence of and mortality from cardiovascular disease.^7,8

The proportion of persons with hypertension who have their BP controlled (defined as systolic/diastolic BP <160/95 mm Hg) has increased dramatically during the past several decades.⁹ However, percentages of persons with controlled hypertension defined as a systolic/diastolic BP less than 140/90 mm Hg are still only 19% in men and 28% in women in the US general population.¹⁰ More troublesome, the increase in percentages of persons with controlled hypertension has lessened in recent years, even though several new effective pharmacologic and nonpharmacologic interventions have been introduced for lowering BP.^11,12 Epidemiologic studies have examined the effects of socioeconomic status, health care, and lifestyle factors on hypertension control.^13-16 These studies have been conducted in hospitals, in managed care settings, and among inner city minority groups.^13-16 Data on factors associated with hypertension control in the general population are sparse, although this information is critically important as a scientific basis for developing strategies to enhance hypertension treatment and control in the community.

We took advantage of the large representative sample of the US general population studied in the Third National Health and Nutrition Examination Survey (NHANES III) to examine the association between socioeconomic status, health care, and lifestyle factors, and hypertension control. This article updates previous findings¹⁰ on hypertension control from NHANES III phase 1 participants and extends the focus to factors related to hypertension control.

The NHANES III was conducted by the National Center for Health Statistics from 1988 to 1994. Details of the study's design and methods have been published elsewhere.¹⁷ In brief, the NHANES III used a stratified multistage probability design to obtain a representative sample of the civilian noninstitutionalized US general population.¹⁷ The design included oversampling of the very young, the elderly, non-Hispanic blacks, and Mexican Americans to improve the precision of estimates in these groups. A total of 19 618 persons 18 years and older participated in an interview and 17 752 participated in a medical examination at a mobile examination center or their home. The overall response rate for interview was 85.6%, and for examination, 78.9%, in the NHANES III. Of this group, 6143 participants were identified as having hypertension at the examination on the basis of the presence of an average systolic BP of 140 mm Hg or more and/or diastolic BP of 90 mm Hg or more, or having a personal history of hypertension and taking antihypertensive medication. Of them, 164 participants were excluded from the current analysis because their self-reported race or ethnicity was not non-Hispanic white, non-Hispanic black, or Mexican American. In addition, 93 participants were excluded because of a lack of a BP measurement. Therefore, 5886 participants, consisting of 3077 non-Hispanic whites, 1742 non-Hispanic blacks, and 1067 Mexican Americans with hypertension, were included in the current analysis. A total of 1254 NHANES III participants who had a self-reported history of hypertension but did not meet the current definition of hypertension (BP ≥140/90 mm Hg and/or taking antihypertensive medication¹¹) were not included in this analysis. Of them, 653 currently used lifestyle modification intervention to control their BP.

The NHANES III data were obtained during a standardized home interview and a subsequent physical examination at a mobile examination center. Information on a wide variety of sociodemographic, medical history, nutritional history, and health service questions, such as age, race/ethnicity, sex, education, physical activity, history of smoking, hypertension, diabetes, and alcohol consumption, and health service, were obtained during the participant's home interview. A question on leisure-time physical activity, "Compared with most [men/women] your age, would you say that you are more active, less active, or about the same?" was asked. The interview included 5 questions related to the diagnosis and treatment of hypertension.^10,17

Blood pressure was measured 3 times during the home interview and another 3 times at the mobile examination center by trained observers according to a standard protocol.¹⁰ The BP for individual participants was calculated as the average of all available systolic and diastolic readings. Controlled hypertension was defined as systolic BP less than 140 mm Hg and diastolic BP less than 90 mm Hg. Body weight and height were measured according to a standard protocol, and body mass index (calculated as weight in kilograms divided by the square of height in meters) was used as an index for obesity.

Percentage of persons with controlled hypertension was calculated, with all study participants with hypertension (both pharmaceutically treated and untreated) used as the denominator. All estimates were weighted to represent the total civilian, noninstitutionalized US general population. Sampling weights were adjusted to reduce bias from nonresponse at the interview stage. Standard errors were calculated by a technique appropriate to the complex survey design and estimation procedure.¹⁸ The proportions (means) of hypertension control and socioeconomic, health care, and lifestyle factors were calculated by race/ethnicity and tested by χ² statistics or weighted analysis of variance. Multiple logistic regression analysis was used to explore the association between study factors and hypertension control rates. The odds ratios and 95% confidence intervals for controlled hypertension were calculated by race/ethnicity groups. Because of a relatively high percentage of persons with controlled hypertension, odds ratios may be an overestimate of relative risk in the present study. All statistical analyses were performed with Stata software (Stata Corp, College Station, Tex).¹⁸

Socioeconomic, health care, and lifestyle characteristics of the study participants by race/ethnicity are presented in Table 1. On average, non-Hispanic whites were more likely to have graduated from high school, to have a household income of $20 000/y or less, and to be more physically active, while non-Hispanic blacks were less likely to be married and more likely to be current smokers. A majority of the study participants had health insurance, visited the same health care facility, and saw the same health care provider, although the percentage for them was lower for Mexican Americans. Most study participants had their BP measured during the preceding 6 months and more than half of non-Hispanic whites and blacks were taking antihypertensive medication, although the proportions were lower in Mexican Americans. Approximately half of the study participants reported use of lifestyle modification as definitive or adjunctive therapy for management of their hypertension. The most common lifestyle modification interventions reported were dietary sodium reduction, weight loss, and exercise.

In each race/ethnicity group, percentages of persons with controlled hypertension were higher in women than in men (Figure 1). Overall, non-Hispanic white and black women had the highest percentage of persons with controlled hypertension, while Mexican American men had the lowest percentage. The pattern for hypertension control by age was similar for each of the race/ethnicity groups, being greatest for those 45 to 64 years old and lowest for those 75 years or older (Figure 2).

Table 2 presents the percentage of persons with controlled hypertension according to socioeconomic status, health care, and lifestyle factors. The percentage was statistically significantly higher in persons who were overweight, visited the same health care facility or saw the same health care provider, had a recent BP measurement, or reported using lifestyle modifications to control their hypertension among all 3 race/ethnicity groups. The percentage was higher in persons who were currently or had been formerly married in non-Hispanic whites and Mexican Americans. The percentage was lower in current smokers among non-Hispanic blacks and lower in heavy alcohol drinkers among non-Hispanic blacks and Mexican Americans. The percentage was higher in persons who reported being less physically active or had private health insurance among non-Hispanic whites and blacks.

Table 3 shows age- and sex-adjusted odds ratios of hypertension control associated with socioeconomic status, health care, and lifestyle factors by race/ethnicity group. The findings were consistent with univariate analysis presented in Table 2.

In a combined analysis that including all study participants, adjustment for education, income, cigarette smoking, alcohol consumption, physical activity, and body weight, in addition to age, sex, and race/ethnicity, did not change the results (Table 4). Participants who were currently or had been formerly married, had private health insurance, visited the same health care facility or saw the same health care provider, had their BP checked during the preceding 6 months or 6 to 11 months, or used a lifestyle modification intervention for hypertension management were more likely to have their hypertension controlled.

This population-based study extends our understanding of hypertension control in the US general population in several important ways. First, the findings confirm previous reports indicating that the overall rate of hypertension control is low in the US general population.^10,15,16 Furthermore, percentages of persons with controlled hypertension are lower in men than in women, and lower in those aged 75 years or older. Mexican Americans have the lowest rates of hypertension control in every age and sex group. This study also indicates that a regular source of health care is an important predictor of hypertension control. In addition, this study suggests that lifestyle modification is in common use by persons with hypertension and is associated with a higher proportion of hypertension control in the US general population.

Socioeconomic status has been associated with access to health care and hypertension control in several epidemiologic studies.^15,19-21 National Health Interview Surveys indicate that lower socioeconomic status is associated with fewer visits to physicians' offices and with a lower likelihood of having been screened for hypertension.^20,21 Kotchen and colleagues¹⁵ studied socioeconomic status and hypertension control in a random sample of African Americans living in inner-city Milwaukee, Wis. Socioeconomic status was divided into high, middle, and low categories on the basis of median residential rental of the census blocks. Lower socioeconomic status was associated with lower access to private health care and poor hypertension control in the study population.¹⁵ Lower education and household income was also associated with poor BP control in the Atherosclerosis Risk in Communities Study.²² In the current study, neither education level nor household income was associated with hypertension control. This may indicate that many federal, state, and local public health interventions have been effective in reaching the poor and groups with lower socioeconomic status.

Several studies reported that lack of health insurance is associated with poor hypertension control in inner-city minority groups.^13,23 The present study found that having government or private health care insurance was associated with a significantly higher rate of hypertension control in non-Hispanic blacks and that having private health care insurance was associated with a higher rate of hypertension control in an overall analysis.

The present study indicated that hypertensive persons who visited the same health care facility or had the same health care provider had a 2- to 5-fold higher odds of having their hypertension controlled. This finding was consistent with experience in a case-control study conducted in 93 black or Hispanic cases with severe, uncontrolled hypertension and 114 controls with controlled hypertension who were seen in the New York City Hospital's emergency department.¹³ In their study, Shea and colleagues¹³ found that severe, uncontrolled hypertension was 3.5-fold more common among persons who had no primary care physician after adjustment for age, sex, race/ethnicity, education, smoking status, alcohol-related problems, use of illicit drugs during the previous year, and lack of health insurance.¹³

The present study also identified a strong and consistent association between frequency of BP measurement and hypertension control. Participants who had had their BP measured during the preceding 6 months were 6- to 13-fold more likely to have their hypertension controlled than those whose BP had not been measured for at least 1 year before the study. These findings suggest that access to high-quality health care may be one of the most important factors for hypertension control in the general population.

The present study identified a strong association between lifestyle modification intervention and hypertension control. Study participants who used any lifestyle modification to lower their BP were 5- to 11-fold more likely to have their hypertension controlled. Clinical trials have documented that weight loss, sodium reduction, exercise, alcohol restriction, and potassium supplementation reduce BP in persons with hypertension.¹¹ Besides its BP-lowering effect, lifestyle modification may also lead to an improvement in compliance with antihypertensive drug treatment recommendations by actively involving persons in their health care.¹⁹ However, because of the cross-sectional design used in the current study, it is not possible to rule out the possibility that the association noted between lifestyle modification and control of hypertension resulted from selection bias, ie, patients with severe hypertension being less likely to receive lifestyle modification interventions. In addition, compliance with lifestyle modification may also be an indicator of overall compliance with all advice and treatment. Therefore, our findings do not provide evidence of a causal relationship between lifestyle modification and hypertension control.

The present study suggests that about 40% to 53% of patients with hypertension in the US general population were attempting to control their hypertension by reducing their sodium intake, 27% to 34% by controlling their weight, and 11% to 13% by exercising. These may be slight underestimates, as they do not include subjects who have successfully controlled their hypertension through lifestyle modifications. We excluded 653 participants who were currently using lifestyle modification intervention and whose BP was less than 140/90 mm Hg because they did not meet the current definition of hypertension.¹¹ These data indicated that lifestyle modification interventions are in common use for treatment of hypertension in the US general population.

In previous studies, current cigarette smoking and alcohol consumption were associated with poor control of hypertension.^19,22 In the present investigation, current cigarette smoking was associated with a slightly, but nonsignificantly, lower percentage of persons with controlled hypertension. Heavy alcohol consumption was associated with lower percentages of persons with controlled hypertension. The present study found that less self-reported physical activity was associated with higher percentages of persons with controlled hypertension in non-Hispanic blacks. However, this association was no longer statistically significant after adjustment for other important covariables. Our results also indicated that overweight was associated with higher percentages of persons with controlled hypertension in non-Hispanic whites and blacks. In the Atherosclerosis Risk in Communities Study, overweight was associated with enhanced hypertension awareness and treatment in white and African American participants.²² This finding may reflect the fact that patients who were overweight may be monitored and have their hypertension treated more aggressively because being overweight is known to increase the risk of having hypertension and developing cardiovascular disease.

The present study also indicated that marital status was associated with hypertension control. Men and women who were currently or had been formerly married had higher percentages of control of hypertension. Lack of social support has been identified as a risk factor for poor compliance with antihypertensive treatment and hypertension control in other studies.¹⁹

Although overall findings from this study are consistent across race/ethnicity groups, several race/ethnicity differences are worth mentioning. The proportion of married persons was much lower in non-Hispanic blacks compared with other groups, and marital status was not associated with hypertension control in this group. This might reflect the cultural differences among race/ethnicity groups. Percentage of hypertensive patients who were reducing their dietary intake of sodium was much higher in non-Hispanic blacks compared with other race/ethnicity groups in our study. This might reflect the fact that physicians were more likely to recommend sodium reduction in non-Hispanic blacks because of perceptions that they were more sensitive to sodium reduction.

Because of our study's cross-sectional design, causal associations cannot be established with certainty. Other limitations include the relative lack of information regarding the participants' health care providers, the providers' relationship to their patients, and the lack of information regarding patients' knowledge and their attitude toward hypertension. In addition, "white-coat" hypertension cannot be defined in the present study because ambulatory BP measurements were not obtained.

Important strengths of this study include the fact that it was conducted in a large representative sample of the US general population; that there were a large number of hypertension participants in each major race/ethnicity group, permitting race-specific analyses; that all the data were carefully collected by a specially trained study staff using standard instruments; and that 6 BP readings obtained at 2 separate visits were available for each of the participants.

Our findings have important public health implications. First, this study indicates that the overall rate of hypertension control is unacceptably low in the US general population. Improvement of hypertension control should be a public health priority to reduce the burden of BP-related morbidity and mortality in the United States. To achieve better rates of hypertension control in communities, patients should have a regular source and same provider of health care, and health care providers should check their patients' BP on a regular basis. Finally, lifestyle modification should be recommended as an important component of hypertension control in the general population.

Accepted for publication October 2, 2001.

This study was supported in part by grant R01HL60300 from the National Heart, Lung, and Blood Institute of the National Institutes of Health, Bethesda, Md.

Corresponding author and reprints: Jiang He, MD, PhD, Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, 1430 Tulane Ave SL18, New Orleans, LA 70112 (e-mail: jhe@tulane.edu).