Prevalence, Awareness, Treatment, and Control of Hypertension Among United States Adults 1999–2004

NHANES was conducted by the National Center for Health Statistics of the Centers for Disease Control and Prevention.² A stratified, multistage probability sampling design was used in the survey. The detailed measurement procedures and protocols have been described in previous publications and also on its website.^6–8 Since 1999, NHANES has become a continuous survey program. The 1999–2004 results are available online. All of the participants gave informed consent, and the study received approval from the Centers for Disease Control and Prevention Institutional Review Board.

BP was measured 3 or 4 times manually by a trained operator using a mercury sphygmomanometer according to a standard protocol and calculated as the average after excluding the first measurement.⁹ The BP measurement techniques were identical over the 3 periods (1999–2000, 2001–2002, and 2003–2004). Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Individuals aged ≥18 years were included in the analysis (n=17 061). Participants who were interviewed but not examined were excluded from analysis (n=1190). Those with missing data on BP or BMI were also excluded (n=1218).

Hypertension was defined as an average BP ≥140/90 mm Hg or if the participant was taking antihypertensive medications. The same BP criteria applied to diabetic participants. Participants who had been diagnosed to have hypertension were considered to be aware of their hypertension. Participants were considered to be treated if they were taking antihypertensive drugs. Hypertension was considered controlled in those on treatment if the average BP was <140/90 mm Hg in nondiabetic patients. Participants were considered to have diabetes if it had been diagnosed previously by a doctor or if they were receiving insulin or oral diabetic medications. For diabetic participants, hypertension was considered to be controlled if the average BP was <130/80 mm Hg.⁵

Data were analyzed using the complex sample function of SPSS (version 13.0). Sampling errors were estimated using the primary sampling units and strata provided in the data set. Sampling weights were used to adjust for nonresponse bias and the oversampling of blacks, Mexican Americans, and the elderly in NHANES. The prevalence of hypertension, as well as the awareness, treatment, and control rates, were age adjusted by direct standardization to the US 2000 standard population.¹⁰ To analyze differences over time, the 2003–2004 data were compared with the 1999–2000 data. Estimates with a coefficient of variation >0.3 were considered unreliable. A 2-tailed P value <0.05 was considered statistically significant.

In this analysis, data on 14 653 individuals (4749 in 1999–2000, 5032 in 2001–2002, and 4872 in 2003–2004) were included. Table 1 shows the age, race/ethnicity, BMI, BP, and diabetic status of the US population estimated from NHANES 1999–2004. There was no overall significant difference in the mean age between 1999–2000 and 2003–2004. However, in 2003–2004, there were fewer people in the 18 to 39 age group and more people in the 40 to 59 age group. In 2003–2004, fewer US adults had not completed high school. There was a trend of decreasing BP levels, especially in the elderly. Among the ≥60 age group, the mean systolic and diastolic BPs had decreased significantly (P=0.006 and 0.003, respectively). It is interesting that weight and BMI did not increase significantly from 1999–2000 to 2003–2004. However, the waist circumference had increased, from 94.95±0.64 cm (37.4±0.3 in) in 1999–2000 to 95.43±0.34 cm (37.6±0.1 in) in 2001–2002 to 96.74±0.35 cm (38.1±0.1 in) in 2003–2004 (P=0.014).

Table 2 shows the prevalence of hypertension in the United States. There was no significant change in the prevalence of hypertension, even in different age, sex, race/ethnicity, and BMI groups (P>0.05). The prevalence of hypertension was also analyzed by sex, race/ethnicity, and BMI in 18 to 39, 40 to 59, and ≥60 age groups, but the age-specific changes in sex, race/ethnicity, and BMI were not significant. In all 3 of the periods, the prevalence of hypertension increased with increasing age and BMI (P<0.001) but did not vary significantly with sex. Non-Hispanic blacks had the highest prevalence. To measure the independent association of these variables and education with hypertension, a multiple logistic regression using the data from NHANES 2003–2004 was performed with hypertension as the dependent variable. Increasing age, increasing BMI, being non-Hispanic black, and having less education were significantly associated with hypertension, whereas sex was not (Table 3).

Table 4 shows the rates of awareness, treatment, and BP control. In 2003–2004, 75.7±2.1% of people with hypertension were aware of their diagnosis, 65.1±2.4% were treated, and 36.8±2.3% had BP controlled (56.6±2.4% among those treated). The age-adjusted awareness and treatment rates increased but not significantly since 1999–2000. There was also a nonsignificant increase in the control rate among treated hypertensive people. The proportion of treated hypertensive people with diabetes reaching the contemporary target of <130/80 mm Hg increased, although this was not statistically significant after age adjustment. However, the BP control rate among all people with hypertension increased significantly from 29.2±2.3% in 1999–2000 to 36.8±2.3% in 2003–2004 (P=0.02 and after age adjustment, P=0.006). The increase in age-adjusted BP control rate in this 6-year period was 8.1% (95% CI: 2.4 to 13.8%) in absolute terms.

The Figure shows the age-specific prevalence of hypertension and BP control rates in different racial/ethnic groups. It shows the marked increase in the prevalence of hypertension with age and the significantly higher prevalence of hypertension in non-Hispanic blacks. The control rates were the lowest in middle-aged Mexican American women (27.8%) among all of the hypertensive people and in old non-Hispanic black women (39.8%) among those treated.

Table 5 shows the effects of sex and race/ethnicity on the awareness, treatment, and control of hypertension. Between 1999 and 2004, there were no significant changes in the awareness and treatment rates of hypertension by sex and race/ethnicity. There was also no significant change in the control rate among treated hypertensive people by sex and race/ethnicity. However, the control rate among all of the hypertensive people increased in both men and women (P=0.03 and 0.05, respectively). It improved significantly in non-Hispanic blacks and Mexican Americans (P=0.02 and 0.03, respectively). In Mexican American men, it increased from 8.7±2.2% in 1999–2000 to 31.1±7.0% in 2003–2004 (P=0.002), which was mainly because of the age-adjusted proportion on treatment increasing from 28.9±4.9% to 49.8±8.8% (P=0.04). Table 6 shows the effects of age and BMI on the awareness, treatment, and control of hypertension in 1999–2004. The awareness, treatment, and control rates of hypertension increased in the ≥60 age group (P≤0.01). In this elderly age group in 2003–2004, 81.0±2.0% were aware of their diagnosis, 73.4±2.1% received treatment, and BP was controlled in 36.7±1.9% (50.0±2.5% among those treated). In the 18- to 39-year age group, there was an increase in the control rate (P=0.04) but not in the awareness and treatment rates. In obese people with a BMI ≥30 kg/m², there was an increase in the control rate (P=0.007). The unadjusted rates of awareness, treatment, and control of hypertension in different sex, race/ethnicity, age, and BMI groups are shown in Supplementary Tables I and II (available online at http://hyper.ahajournals.org).

The NHANES database has been valuable for the study of the trends in the health status of a population because of its large sample size, complex sampling design, good quality control, and comprehensive content.² It has been used to study the prevalence, awareness, treatment, and control of hypertension in the United States.^3,4,11–15 The data for 2003–2004 have just been released, and our analysis confirms previous observations and extends previous conclusions. The prevalence of hypertension has not increased significantly since 1999. At the same time, there has been increasing control rate of hypertension, especially in Mexican American men, elderly, and obese people. The lack of increased prevalence of hypertension may be because of the lack of a further increase in BMI in the population, although there was an increase in waist circumference. Better publicity and education and effort from health professionals may be another reason for this. The increase in BP control rates may be the result of clinical guidelines on the management of hypertension⁵ rather than improved antihypertensive drugs, because there were no new major antihypertensive drugs introduced in this period. Public health measures and intensified antihypertensive treatment seem to be effective in improving BP control rates. The latest hypertension management guidelines recommended the use of diuretics as first-line antihypertensive medications for uncomplicated hypertension and the use of multiple antihypertensive drugs to control BP.^5,16 A recent report revealed an increase in the use of antihypertensive medications in men from NHANES III (1988–1994) and NHANES 1999–2002.¹⁷ Moreover, there was increasing use of angiotensin-converting enzyme inhibitor among hypertensive patients with diabetes, congestive heart failure, and heart attack.¹⁷ This was consistent with current guidelines.^5,16 The use of diuretics also increased shortly after the release of new evidence on the clinical equivalence of diuretics to calcium channel blockers and angiotensin-converting enzyme inhibitors in December 2002.^18,19 In recent years, it is recognized that most hypertensive patients require multiple antihypertensive drugs to control BP.^20–23 The readiness to use multiple drugs may explain the increase in BP control rates in 1999–2004.

The awareness and control rates of hypertension were better in older people and people with higher BMI, whereas there was relative undertreatment of hypertensive people of younger age and lower BMI. This may be because of greater concerns in health status among older and obese people. The significant increase in waist circumference but not weight and BMI suggests an increasing trend of abdominal obesity. In fact, a detailed study of obesity in NHANES 1999–2004 revealed a significant increase in the prevalence of obesity among men but not women.²⁴

Large clinical trials in hypertensive patients showed that BP reduction is crucial in reducing adverse cardiovascular outcomes.^25–28 A small reduction in BP could reduce the risk of heart failure, stroke, and myocardial infarction markedly.^5,29 The increased control rates and decreased mean BPs, especially among the elderly, may help to decrease the incidence of strokes and heart attacks, which is highly encouraging. Clinicians need to overcome clinical inertia and step up treatment to reach the BP target. Other than antihypertensive medication, lifestyle interventions should also be encouraged in hypertensive patients.³⁰

According to the national goal of Healthy People 2010, the proportion of adults with high BP should be reduced to 16%, and the proportion of adults with high BP whose BP is under control should be reduced to 50%.³¹ It seems unlikely, because of the aging population, that the proportion of adults with hypertension can be reduced to 16%. However, the improvement in BP control rates in the last few years is highly encouraging, suggesting that, with concerted effort from health professionals and the government agencies, the target of 50% BP control rate may be reached by 2010. Recent clinical trials showed that a high BP control rate of 66% to 71% could be achieved in some settings and environments.^22,23,32,33 Although we did not find a significant increase in the age-adjusted rate of BP control in patients under treatment, the age-adjusted rate was 63.9% in 2003–2004, close to the control rates of 70% to 80% found in clinical trials.^22,23,32,33 The increase in age-adjusted control rates among treated hypertensive patients with diabetes from 15.7% to 33.2% is promising although not statistically significant, but this is still in need of substantial improvement. In 2003–2004, approximately one third of hypertensive patients were unaware of their hypertension, and our efforts to detect hypertension need to be improved. The treatment rate among hypertensive patients was only 53.7%. Therefore, approximately half of the hypertensive subjects were not being treated. To meet Healthy People 2010, the greatest attention must be paid to this group. In countries with limited resources like Cuba, very good BP control can be achieved through investment in the training of health professionals, availability of locally manufactured drugs, and more aggressive policy at the primary care level.³⁴ Therefore, there is room for further improvement in the control of hypertension in the United States.

Studies using a BP goal of <140/90 mm Hg in hypertensive people with diabetes might overestimate BP control rates.^{1,3,4,13–15,35} Therefore, in this study, different control goals were used for those with and without diabetes to produce a more accurate control rate. However, there are some limitations in our study. The BP goal for hypertensive patients with diabetes was changed from <130/85 mm Hg to <130/80 mm Hg in 2003.⁵ Therefore, the BP control rates in people with diabetes might seem poor before 2003. Inaccuracies in estimating BP control rates in the diabetic population might also arise from not classifying diabetic people with prehypertension as hypertensive, although they may require antihypertensive medications to bring their BP to <130/80 mm Hg. The limited sample sizes for each 2-year period made the estimates in some subgroups less reliable and made it difficult to study interactions between factors such as age, sex, and race/ethnicity. Estimates of prevalence are substantially affected by age adjustment. The age distribution of the US general population, used for the adjustment standard, is much younger than the hypertensive population, so the adjustment emphasizes the estimates among the younger population. A better standard population might be the age distribution of those with hypertension, such as for the entire 1999–2004 period.

None.