Uncontrolled and Apparent Treatment Resistant Hypertension in the United States, 1988 to 2008Clinical Perspective
Background—Despite progress, many hypertensive patients remain uncontrolled. Defining characteristics of uncontrolled hypertensives may facilitate efforts to improve blood pressure control.
Methods and Results—Subjects included 13 375 hypertensive adults from National Health and Nutrition Examination Surveys (NHANESs) subdivided into 1988 to 1994, 1999 to 2004, and 2005 to 2008. Uncontrolled hypertension was defined as blood pressure ≥140/≥90 mm Hg and apparent treatment-resistant hypertension (aTRH) when subjects reported taking ≥3 antihypertensive medications. Framingham 10-year coronary risk was calculated. Multivariable logistic regression was used to identify clinical characteristics associated with untreated, treated uncontrolled on 1 to 2 blood pressure medications, and aTRH across all 3 survey periods. More than half of uncontrolled hypertensives were untreated across surveys, including 52.2% in 2005 to 2008. Clinical factors linked with untreated hypertension included male sex, infrequent healthcare visits (0 to 1 per year), body mass index <25 kg/m2, absence of chronic kidney disease, and Framingham 10-year coronary risk <10% (P<0.01). Most treated uncontrolled patients reported taking 1 to 2 blood pressure medications, a proxy for therapeutic inertia. This group was older, had higher Framingham 10-year coronary risk than patients controlled on 1 to 2 medications (P<0.01), and comprised 34.4% of all uncontrolled and 72.0% of treated uncontrolled patients in 2005 to 2008. We found that aTRH increased from 15.9% (1998–2004) to 28.0% (2005–2008) of treated patients (P<0.001). Clinical characteristics associated with aTRH included ≥4 visits per year, obesity, chronic kidney disease, and Framingham 10-year coronary risk >20% (P<0.01).
Conclusion—Untreated, undertreated, and aTRH patients have consistent characteristics that could inform strategies to improve blood pressure control by decreasing untreated hypertension, reducing therapeutic inertia in undertreated patients, and enhancing therapeutic efficiency in aTRH.
Blood pressure (BP) control to <140/<90 mm Hg among a representative sample of hypertensive patients participating in the National Health and Nutrition Examination Surveys (NHANESs) improved from 27.3% in 1988 to 1994 to 50.1% in 2007 to 2008.1,2 On the basis of NHANES 2007 to 2008 hypertension prevalence data and estimates of the US population ≥18 years of age in 2008, there are ≈67.5 million hypertensive adults in the United States, with >33 million uncontrolled.2,3 Approximately 13 of the 33 million uncontrolled hypertensive patients are unaware and 20 million aware of their hypertension.
Clinical Perspective on p 1058
Data from NHANES 1988 to 2008 document that the majority of uncontrolled hypertensive patients remain untreated.2 Other evidence suggests that therapeutic inertia, ie, the failure to uptitrate or to add antihypertensive medications when BP is uncontrolled, is a significant contributor to uncontrolled hypertension.4–6 Moreover, the prevalence of treatment-resistant hypertension, ie, BP uncontrolled on a rational regimen including ≥3 or controlled on ≥4 antihypertensive medications, appears to be increasing. This may reflect the effects of an aging population with more obesity and related comorbidities, eg, diabetes mellitus and chronic kidney disease (CKD).7
Defining the clinical characteristics of hypertensive patients with untreated, inadequately treated, and treatment-resistant hypertension could inform strategies and interventions for improving BP management. Therefore, the NHANESs 1988 to 2008 were analyzed. Although medication dose and adherence and BP measurement artifacts are also significant factors in defining inadequately treated and treatment-resistant hypertension,7 NHANES does not capture these variables. Thus, the term apparent treatment-resistant hypertension (aTRH) is applied.
As described previously,2 NHANESs 1988 to 1994 and 1999 to 2008 were conducted by the Centers for Disease Control and Prevention National Center for Health Statistics (NCHS). Through the use of stratified, multistage probability sampling of the noninstitutionalized US population, NHANES volunteers were selected. All adults provided written informed consent approved by the NCHS Institutional/Ethics Review Board.8,9
Race/Ethnicity and BP Measurement
Race/ethnicity was determined by self-report and separated into non-Hispanic white (white), non-Hispanic black (black), and Hispanic ethnicity of any race as described.1,2,10 Although other race data were obtained (American Indian, Alaskan Native, Asian or Pacific Islander, and other race not specified), the number of individuals was insufficient for meaningful analyses.2
The BP measurement methods were consistent across 1988 to 1994 and 1999 to 2008.2 In brief, BP was measured by trained physicians using a mercury sphygmomanometer and appropriately sized arm cuff on subjects after 5 minutes of seated rest. Individuals without recorded BP were excluded. In the determination of mean systolic and diastolic BPs for individuals, the first BP value was used if only 1 measurement was obtained. The second BP was used if 2 readings were taken; the second and third values were averaged when available. More than 90% of subjects had ≥2 BP measurements in all survey periods.2
Hypertension was defined as mean systolic BP ≥140 mm Hg and/or mean diastolic BP ≥90 mm Hg and/or a positive response to the question “Are you currently taking medication to lower your BP?”1,2,10 Awareness of hypertension was determined by hypertensive patients responding affirmatively to the question, “Have you ever been told by a doctor or other healthcare professional that you had hypertension, also called high blood pressure?”1,2,10 Treatment of hypertension was established by participants responding “yes” to the question, “Because of your hypertension/high blood pressure, are you now taking prescribed medicine?”1,2,10
Control of hypertension was defined as BP <140/<90 mm Hg across all survey periods, although the BP goals for high-risk subgroups, including diabetics, were lower for 1999 to 2008.11–14 Recent evidence does not clearly support a goal systolic BP <140 mm Hg for diabetic patients or those with nonproteinuric hypertensive renal disease.15,16 For these reasons and to facilitate comparisons across study, this report focuses exclusively on goal BP <140/<90 mm Hg.
Concomitant risk factors, including diabetes mellitus and cigarette smoking, were defined as described.2 Definitions for cardiovascular disease, including stroke and coronary heart disease (CHD), were described.2 We defined CKD as an estimated glomerular filtration rate <60 mL·m−2·min−1 and/or urine albumin:creatinine ≥300 mg/g.17,18 Serum creatinine values were adjusted to facilitate comparisons of estimated glomerular filtration rate across surveys.19
Diabetes mellitus was defined by a positive response to the questions, “Have you ever been told by a doctor that you have diabetes?” and/or “Are you now taking insulin?” and/or “Are you now taking diabetes pills to lower your blood sugar?” The definition did not include patients with only fasting plasma glucose ≥126 mg/dL, ie, “undiagnosed diabetes mellitus.”20,21
Low-Density Lipoprotein Cholesterol and Framingham CHD Risk
The number of medical visits in the past year was assessed by the question, “How many times did you receive health care over the last year”? The responses were grouped into 3 categories: 0 to 1, 2 to 3, and ≥4 healthcare visits a year.
Insurance Status and Therapeutic Inertia
Insurance status is determined by a yes/no answer to the question, “Are you covered by health insurance or some other kind of healthcare plan?”
Therapeutic inertia was calculated for uncontrolled patients who reported ≥2 healthcare visits per year. Uncontrolled patients reporting 0 to 1 visits annually were excluded because infrequent care provides little opportunity to intensify treatment. Therapeutic inertia was arbitrarily defined as high, moderate, low, and none on the basis of the number of antihypertensive medications reported, BP, and 10-year Framingham CHD risk.
High-level therapeutic inertia included patients on 0 to 1 antihypertensive medications with BP ≥160/≥100 mm Hg regardless of 10-year CHD risk, ≥150/≥95 and <160/<100 mm Hg with 10-year CHD risk 10% to 20% and >20%, and ≥140/≥90 and <150/<95 mm Hg with 10-year CHD risk ≥20%.
Moderate-level therapeutic inertia included patients on 2 medications with BP ≥160/≥100 mm Hg regardless of 10-year CHD risk, ≥150/≥95 and <160/<100 mm Hg with 10-year CHD risk 10% to 20% and >20%, and ≥140/≥90 and <150/<95 mm Hg with 10-year CHD risk ≥20%, as well as patients on 0 to 1 medications with BP ≥150/≥95 and <160/<100 mm Hg with 10-year CHD risk <10% or ≥140/≥90 and <150/<95 mm Hg with 10-year CHD risk 10% to 20%.
Low-level therapeutic inertia included patients on 2 meds with BP ≥150/≥95 and <160/<100 mm Hg with 10-year CHD risk <10% and ≥140/≥90 and <150/<95 mm Hg with 10-year CHD risk 10% to 20%, as well as patients with BP ≥140/≥90 and <150/<95 mm Hg on 0 to 1 medications and 10-year risk <10%. No therapeutic inertia included patients with BP ≥140/≥90 and <150/<95 mm Hg on 2 medications and 10-year risk <10% and all uncontrolled patients on ≥3 BP medications.
Participants were asked if they had taken any prescription medications in the past month. During the household surveys in 1999 to 2004 and 2005 to 2008, participants were requested to provide prescription containers, and 88.8% and 88.3%, respectively, did so. Each medication identified from medications provided or described was recorded and matched to a prescription drug database. Each medication identified was assigned its generic equivalent. Antihypertensive medications were classified into a single category according to the seventh report of the Joint National Committee on Hypertension with the addition of proprietary medications not marketed when the document was published.14 Single-pill combinations were separated into their generic components. Each medication was classified into only 1 category. The sum of BP medication categories defined the number of antihypertensive medications taken by each patient.
The NHANES Analytic and Reporting Guidelines were followed.23 SAS callable SUDAAN version 9.0.1 (SAS Institute Inc, Cary, NC) was used for all analyses to account for the complex NHANES sampling design. Standard errors were estimated with Taylor series linearization. All NHANES periods were age adjusted to the US 2000 Census data as described.2
To test for significant differences in variables between/among groups within each survey, the χ2 test in the CROSSTAB procedure was used for categorical variables, the WALD F test in the REGRESS was used for continuous variables, and LOGLINK was used for medications. When >2 groups were compared within an NHANES survey period, analysis was limited to assessing differences across the groups and not between pairs. Pairwise comparisons between the 3 NHANES periods were conducted with t tests of weighted means1 between pairs of the 3 groups matched for medication number. The effects of time were not assessed before pairwise comparisons because SUDAAN would not allow this function when only 3 time periods were analyzed. Because multiple statistical comparisons were performed within and between the 3 NHANES time periods, 2-sided values of P<0.01 were accepted as statistically significant.
Odds ratios and 95% confidence intervals were calculated to assess the relationship of selected independent clinical covariables to 3 dependent variables: untreated uncontrolled hypertension, uncontrolled hypertension treated with 1 to 2 BP medications, and uncontrolled aTRH treated with ≥3 BP medications. Univariable and fully adjusted multivariable logistic regressions were performed with RLOGIST.
Uncontrolled and Controlled Hypertensive Patients
The proportion of patients with uncontrolled hypertension declined from 73.2% in 1988 to 1994 to 52.5% in 2005 to 2008 (Table 1). Men were more likely than women to have uncontrolled hypertension in 1988 to 1994 and 2005 to 2008. Across surveys, a lower proportion of whites and a higher proportion of blacks and Hispanics of any race were represented among uncontrolled patients. As expected, systolic and diastolic BP values were lower in controlled than uncontrolled patients. Systolic BP did not change over time in either group, whereas diastolic BP declined in both.
The number of BP medications increased over time in both treated uncontrolled and controlled patients but did not differ between groups. Uncontrolled patients were more likely to report 0 to 1 healthcare visits annually, whereas controlled patients were more likely to report ≥4 annual visits. The proportion of uninsured patients did not change in controlled hypertensives but increased in uncontrolled patients over time. Body mass index rose over time in both groups and was higher in controlled patients. Low-density lipoprotein cholesterol declined over time in both groups and was lower in controlled patients in 1999 to 2004 and 2005 to 2008. The prevalence of diabetes mellitus increased over time and was more common in controlled than uncontrolled patients. The percentage of current cigarette smokers did not differ between controlled and uncontrolled hypertensives but was lower among uncontrolled patients in 1999 to 2004 than the other 2 time periods.
The percentage of hypertensive patients with CKD did not change significantly with time and was not different in uncontrolled and controlled patients. Prevalent clinical cardiovascular disease was more common in controlled than uncontrolled patients in 1988 to 1994 and 1999 to 2004 but did not change with time. Framingham 10-year CHD risk was higher in controlled than uncontrolled patients in 1988 to 1994 and 1999 to 2004.
Number of Antihypertensive Medications
The percentage of all hypertensive individuals who were untreated declined, whereas the percentages on 2 and ≥3 BP medications increased over time (Figure 1A). Among all uncontrolled patients, the percentage of untreated subjects declined with time, but the mean remained >50% in 2005 to 2008 (52.2%; 95% confidence interval, 48.1 to 56.3;Figure 1B). The percentage of treated uncontrolled patients on 1 medication fell whereas the percentage on ≥3 medications rose with time (Figure 1C). Among treated uncontrolled (Figure 1C) and treated controlled (Figure 1D) hypertensive patients, the number of BP medications reported and changes over time were similar. The proportion on 1 BP medication fell and the proportion on ≥3 BP medications rose. All controlled hypertensives reported taking BP medications. However, ≈4% to 5% of them did not bring any antihypertensive medications to the examination, nor did they identify any BP medications during the interview, including “unspecified” (Figure 1D).
Uncontrolled Hypertensive Patients Grouped by Number of Antihypertensive Medications
The proportion of untreated patients fell between 1988 to 1994 and 2005 to 2008, whereas the percentage on 1 to 2 medications did not change and the percentage on ≥3 medications rose (Table 2). The proportion of uncontrolled hypertensive patients who were uninsured increased with time. Systolic BP was lower and diastolic BP was higher in untreated than treated patients. Untreated hypertensive patients were younger and more likely to be men, infrequent healthcare users, and uninsured and to have 10-year CHD risk <10% than treated patients. Conversely, untreated patients were less likely to have ≥4 healthcare visits annually, stage 2 hypertension, diabetes mellitus, CKD, and 10-year CHD risk >20% than treated patients, especially those on ≥3 BP medications, ie, aTRH.
Controlled Hypertensive Patients Grouped by Number of Antihypertensive Medications
In general, there were more differences by number of medications within uncontrolled than within controlled hypertensive patients (Table 3). Whereas all controlled hypertensive patients reported taking BP medication, 4% to 5% across all survey periods did not verbally identify or physically bring any of them during their assessment. Untreated hypertensive patients were younger than those on ≥3 medications. Framingham 10-year CHD risk generally increased with the number of medications, with significant differences in 1999 to 2004 and 2005 to 2008. The percentage of controlled hypertensive patients on ≥4 medications was 2.3% (95% confidence interval, 1.3 to 3.3) in 1988 to 1994, 5.2% (95% confidence interval, 3.6 to 6.8) in 1999 to 2004, and 7.3% (95% confidence interval, 5.1 to 9.4) in 2005 to 2008.
Therapeutic Inertia in Uncontrolled Hypertensive Patients With ≥2 Healthcare Visits Annually
Using the definition provided in Methods, we find that the percentage of patients with no therapeutic inertia rose with time over the 3 NHANES periods, whereas the percentage of patients with high-level therapeutic inertia declined (Table 4). Nevertheless, more than half of uncontrolled hypertensive patients in all time periods had a moderate or high level of therapeutic inertia.
Classes of Antihypertensive Medications
The class of antihypertensive medications reported by controlled and uncontrolled hypertensive patients taking 1, 2, and ≥3 such medications is provided for each of the 3 NHANES periods in Table 5. The NHANES databases do not indicate the order that the antihypertensive agents were added for patients who reported taking 2 and ≥3 BP medications. In all time periods, angiotensin-converting enzyme inhibitors, β-blockers, and calcium channel blockers were the agents most often reported by controlled and uncontrolled hypertensive patients reportedly taking a single antihypertensive medication. Diuretics, especially thiazide-type diuretics, showed the largest increase in percentage use as BP medication number rose from 1 to ≥3. There was a limited number of differences significant at P<0.01 between uncontrolled and controlled hypertensive patients who reported taking each class of medication in the 3 NHANES time periods. Nondihydropyridine calcium antagonists were the class of antihypertensive medication with the largest number of significant differences between controlled and uncontrolled patients and were taken more often by those who were uncontrolled.
In multivariable logistic regression, the clinical variables consistently and independently associated with untreated hypertension across the 3 NHANES periods included male sex, patients with 0 to 1 healthcare visits a year, body mass index <25 kg/m2, less than stage 3 CKD, and 10-year Framingham 10-year coronary risk 10% to 19% and <10% compared with the reference groups (Figure 2).
Uncontrolled Hypertension on 1 to 2 Blood Pressure Medications
All of the independent variables shown inFigures 3 and 4 were examined simultaneously in multivariable logistic regression analysis. The clinical variables independently associated with uncontrolled hypertension on 1 to 2 BP medications included male sex (1999 to 2004, 2005 to 2008), black race (1988 to 1994, 1999 to 2004), age, and 10-year CHD risk >20% (all time periods for both). Patients on thiazide-type diuretics (1988 to 1994, 1999 to 2004) and angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers (1999 to 2004, 2005 to 2008) were less likely to have hypertension uncontrolled on 1 to 2 BP medications. In contrast, patients reportedly taking nondihydropyridine calcium channel blockers were more likely to have treated uncontrolled hypertension (1988 to 1994, 2005 to 2008).
Clinical Factors Independently Associated With Apparent Treatment-Resistant Hypertension Among All Uncontrolled Patients
In all time periods, patients reporting ≥4 healthcare visits annually were more likely to have aTRH, ie, uncontrolled on ≥3 BP medications (Figure 4). Other clinical factors consistently and independently associated with aTRH in multivariable logistic regression included obesity, CKD, and 10-year CHD risk >20%. Increasing age and black race were independently linked with aTRH in 2005 to 2008, and patients with 10-year CHD risk 10% to 20% were more likely to have treatment-resistant hypertension in 1988 to 1994 and 1999 to 2004.
No significant differences were seen in the percentages of controlled and uncontrolled hypertensive patients taking various classes of antihypertensive medications in 2005 to 2008, with 1 exception (Figure 5). Nondihyrdropyridine calcium channel blockers were taken more often by uncontrolled than controlled hypertensive patients.
The proportion of hypertensive patients with BP ≥140 mm Hg systolic and/or ≥90 mm Hg diastolic declined from 73.2% in 1988 to 1994 to 52.5% in 2005 to 2008 (P<0.001; Table 1 and Figure 1).1,2 Despite progress, >30 million hypertensive patients remain uncontrolled in the United States. The primary objective of this study was to identify consistent characteristics and trends that distinguished uncontrolled from controlled hypertensive patients. The study focused on pursuing this objective in 3 groups of uncontrolled patients: those who are untreated and those reportedly taking 1 to 2 and ≥3 BP medications.
Hypertension treatment increased steadily from 1988 to 2008,1,2,10 yet more than half of uncontrolled patients were untreated in the 3 NHANES periods. Although hypertension awareness improved over the same time period, roughly two thirds of untreated patients in all surveys were unaware (Table 2). Patient characteristics independently associated with untreated hypertension, including both unaware and aware patients, across the 3 NHANES periods included male sex, infrequent health care, lean body mass index, absence of CKD, and 10-year Framingham CHD risk <10% and 10% to 20% (Figure 2), ie, better cardiovascular and renal health. Infrequent health care emerged as a major issue; a mean of ≥40% of untreated hypertensive patients in all survey periods had 0 to 1 healthcare visits annually. Lack of insurance is a factor limiting healthcare use, regardless of income.24 Uninsured status rose with time among uncontrolled patients (Table 1) and was independently linked with untreated hypertension in 2 of 3 NHANES periods (Figure 2), an association noted previously.25
The findings among untreated hypertensive patients suggest that increasing healthcare use in traditional or nontraditional settings, eg, worksites,26 is critical in reducing the burden of untreated hypertension. Raising the perceived value of regular preventive healthcare services among those without clinically overt disease emerges as an important complementary educational strategy. Screening efforts to raise hypertension awareness, especially among medically underserved groups and individuals at lower risk, should include inquiries on regular use of primary care and plans to engage infrequent users and nonusers in ongoing primary care. Initiatives to reduce the growing number of uninsured adults also emerge as important in addressing the substantial burden of untreated uncontrolled hypertension.
Focus on Uncontrolled Hypertension Treated With 1 to 2 Blood Pressure Medications
Approximately one third of all uncontrolled patients reported taking 1 to 2 antihypertensive medications in all 3 NHANES periods. However, among treated uncontrolled patients, the proportion on 1 to 2 medications fell (Table 3). One logical option for improving BP control in this comparatively large group of uncontrolled patients is an additional antihypertensive medication. Clinical variables independently linked to uncontrolled hypertension on 1 to 2 BP medications included increasing (older) age and 10-year Framingham coronary risk >20% in all 3 time periods and male sex, Hispanic ethnicity, and infrequent healthcare in 2 NHANES periods (Figure 3).
The comparatively higher risk among uncontrolled than controlled hypertensive patients on 1 to 2 medications provides further justification for adding a second or third antihypertensive medication to improve BP control. Their mean systolic BP was >150 mm Hg across surveys, with many having stage 2 hypertension, which should also prompt treatment intensification. Whereas infrequent health care was a distinguishing characteristic of treated uncontrolled hypertension on 1 to 2 BP medications, roughly 85% to 90% of this group had ≥2 healthcare visits annually. Thus, the majority appear to be seen sufficiently frequently to allow treatment intensification.
Our previous work identified therapeutic inertia as a significant contributor to uncontrolled hypertension, accounting for ≈19% of the total variance in BP control.5 Although the NHANES database does not include information required to calculate therapeutic inertia,4,5 it was arbitrarily defined on the basis of BP, number of antihypertensive medications, and 10-year CHD risk. Whereas high-level therapeutic inertia declined with time (Table 1), therapeutic inertia was high or moderate in more than half of uncontrolled hypertensives who reported ≥2 healthcare visits annually in all 3 NHANES periods (Table 4). Therapeutic inertia is not a provider-only trait but appears to reflect a patient-provider interaction.5 In addition to provider interventions to reduce therapeutic inertia, educating patients, especially those at higher risk, on the importance of BP control may facilitate efforts to overcome therapeutic inertia.
The data also suggest that medication selection affects control. Patients uncontrolled compared with controlled on 1 to 2 BP medications were less likely to report taking a diuretic (1988 to 1994, 2005 to 2008) and less likely to report taking an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker (1999 to 2004, 2005 to 2008). The findings align with the efficacy of diuretics and renin-angiotensin system blockers, especially in combination, for BP control.27 Although significant only in 2005 to 2008, nondihydropyridine calcium channel blockers may also improve BP control, consistent with the antihypertensive efficacy of nondihydropyridine calcium channel blocker as monotherapy and in combination with angiotensin-converting enzyme inhibitor combinations.14,28
Focus on Apparent Treatment-Resistant Hypertension
The proportion of treated uncontrolled hypertensive patients reportedly taking ≥3 BP medications increased from 15.9% in 1988 to 1994 to 28.0% in 2005 to 2008 (Table 3 and Figure 1C). These patients appear to have treatment-resistant hypertension, although 30% to 50% may have pseudohypertension, which includes nonadherence, inaccurate measurements, and artifacts, eg, office or white-coat effect.7,29 Although our report focuses on uncontrolled aTRH, the American Heart Association position statement also identified patients with controlled hypertension taking ≥4 medications as treatment resistant.7 With this definition and data presented in the tables and the Results section, 11.8% of all hypertensive patients in 2005 to 2008 have aTRH, an increase from 5.5% in 1988 to 1994 and 8.5% in 1999 to 2004.
In multivariable analysis, clinical factors independently linked with uncontrolled aTRH across surveys included frequent healthcare use (≥4 healthcare visits annually), which indicates that the medical community has numerous opportunities to improve BP control. The patients with aTRH are further characterized by obesity, CKD, and 10-year Framingham coronary risk >20% (Figure 4). In 2005 to 2008, older age and black race and in 1999 to 2004 female sex emerged as independent clinical correlates of aTRH. The clinical factors independently associated with aTRH identified are consistent with characteristics previously cited.7 Patients with aTRH are generally at significant risk for cardiovascular disease and likely to benefit from better BP control.
Limited effectiveness of current antihypertensive medications and strategies for combining them constitute a barrier to better hypertension control. Of particular note was the infrequent use of aldosterone antagonists among patients with aTRH (Table 5 and Figure 5). Several trials indicate that the addition of an aldosterone antagonist lowers BP 20 to 25/10 to 15 mm Hg in patients with treatment-refractory and -resistant hypertension.30,31 More effective antihypertensive medications or strategies for combining them are urgently needed because the burden of treatment resistant hypertension is likely to grow given an aging population with more obesity and related complications, including diabetes mellitus and CKD. Personalized medicine, including hemodynamic and renin-guided therapeutics,32,33 and genetic testing34 offer alternative or complementary approaches to constructing more effective medication regimens for treatment-resistant patients.
This study has limitations. Although NHANES represents the noninstitutionalized civilian US population, the sample size is relatively small. Thus, not all clinically and epidemiologically important differences may be detected. Second, medications were determined by self-report, which may be inaccurate, although >85% of hypertensive patients brought their medications to the 1999 to 2004 and 2005 to 2008 examinations. Adherence was not assessed and medication dose and frequency were not recorded. Third, hypertension was defined by BP on a single visit. Thus, hypertension status may have been incorrectly assigned to some patients, especially the unaware, untreated subset, which had lower BPs than uncontrolled treated patients. Fourth, although lifestyle change is also important to BP control, our analysis focused on treatment. Fifth, sample size, although substantial, is comparatively small for subgroup analyses, which increases the potential for falsely accepting the null hypothesis, ie, no differences between groups. Although appropriate for the large number of statistical comparisons performed in the analysis, the concern of limited power is exacerbated by recognizing statistical significance only at P<0.01.
Despite the limitations, our analyses of a nationally representative hypertensive patient sample from 3 NHANES periods suggest that national efforts to increase healthcare insurance and use of primary care could further improve hypertension awareness and treatment, leading to higher control rates. These efforts are important for untreated uncontrolled hypertensive patients who are more often men, nonobese, and without overt cardiovascular and renal diseases. Uncontrolled hypertensive patients on 1 to 2 BP medications could benefit from an additional antihypertensive medication. These patients are generally seen frequently enough to have therapy intensified. They are older and have greater risk for CHD than their controlled counterparts taking 1 to 2 BP medications; ie, there appear to be compelling reasons to intensify therapy. This group is also more likely to include men, blacks, and Hispanics and to have stage 2 hypertension, although these characteristics were not significant in all time periods. Interventions to reduce therapeutic inertia, which should include efforts directed at both providers and patients, could be useful for improving their BP control and reducing risk.
A growing proportion of hypertensive patients appear to be resistant to treatment. Strategies to enhance therapeutic efficacy and to promote therapeutic adherence and lifestyle change are especially important for this group. Trends toward an older and more obese population with a higher prevalence of diabetes mellitus and CKD could lead to a growing burden of treatment-resistant hypertension.7 To counter these potentially adverse trends and to improve overall hypertension control, integrated healthcare research, policy, and delivery initiatives are needed to reverse growth in unhealthy lifestyles and obesity,35,36 to enhance regular use of a medical home, to reduce therapeutic inertia, to develop more efficacious antihypertensive agents, and to improve methods for selecting effective antihypertensive combinations for individual patients.
Sources of Funding
This work was supported in part by the state of South Carolina (Hypertension Initiative), National Institutes of Health (NIH) R01 National Heart, Lung and Blood Institute HL07080684, NIH R01 National Institute of Diabetes and Digestive and Kidney Diseases DK067615, NIH R01 National Institute of Neurological Disorders and Stroke NS058728, and US Army W81XWH-10–2-0057. The funding agencies were not involved in the design or conduct of the study; collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript.
Dr Egan has received grant support from Daiichi-Sankyo, Novartis, and Takeda; has served as a lecturer with honoraria on CME-accredited programs for the American Society of Hypertension Carolinas-Georgia-Florida Chapter and International Society of Hypertension in Blacks; and has been a consultant to NicOx. Dr Ferdinand has served on the speakers' bureau at AstraZeneca, Novartis, Forest, and Daiichi-Sankyo; has received honoraria from AstraZeneca, Novartis, and Forest; and has served as a consultant/advisory board member for AstraZeneca, Novartis, and Forest. The other authors report no conflicts.
Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz.
- Received March 7, 2011.
- Accepted June 29, 2011.
- © 2011 American Heart Association, Inc.
US Census Bureau. Statistical abstract of the United States: 2011. http://www.census.gov/compendia/statab/2011/tables/11s0007.pdf. Accessed February 17, 2011.
- Rose AJ,
- Berlowitz DR,
- Manze M,
- Orner MB,
- Kressin NR
- Okonofua EC,
- Simpson K,
- Jesri A,
- Rehman S,
- Durkalski V,
- Egan BM
- Wang JT,
- Vasan RS
- Calhoun DA,
- Jones D,
- Textor S,
- Goff DC,
- Murphy TP,
- Toto RD,
- White A,
- Cushman WC,
- White W,
- Sica D,
- Ferdinand K,
- Giles TD,
- Falkner B,
- Carey RM
- Cutler JA,
- Sorlie PD,
- Wolz M,
- Thom T,
- Fields LE,
- Roccella EJ
Centers for Disease Control. National Center for Health Statistics Research Ethics Review Board. http://www.cdc.gov/nchs/nhanes/irba98.htm. Accessed February 17, 2011.
- Burt VL,
- Cutler JA,
- Higgins M,
- Horan MJ,
- Labarthe D,
- Whelton P,
- Brown C,
- Roccella EJ
- Chobanian AV,
- Bakris GL,
- Black HR,
- Cushman WC,
- Green LA,
- Izzo JL,
- Jones DW,
- Materson BJ,
- Oparil S,
- Wright JT Jr.,
- Roccella EJ
- Appel LJ,
- Wright JT Jr.,
- Greene T,
- Agodoa LY,
- Astor BC,
- Bakris G,
- Cleveland WH,
- Charleston J,
- Contreras G,
- Faulkner ML,
- Gabbai FB,
- Gassman JJ,
- Hebert LA,
- Jamerson KA,
- Kopple JD,
- Kusek JW,
- Lash JP,
- Lea JP,
- Lewis JB,
- Lipkowitz MS,
- Massry SG,
- Miller ER,
- Norris K,
- Phillips RA,
- Pogue VA,
- Ranall OS,
- Rostand SG,
- Smorgorzewski MJ,
- Toto RD,
- Wang X
- Stevens LA,
- Coresh J,
- Feldman HI,
- Greene T,
- Lash JP,
- Nelson RG,
- Rahman M,
- Deysher AE,
- Zhang Y,
- Schmid CH,
- Levey AS
- Selvin E,
- Manzi J,
- Stevens LA,
- Van Lente F,
- Lacher DA,
- Levey AS,
- Coresh J
Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 1997;20:1183–1197.
- Harris MI,
- Goldstein DE,
- Flegal KM,
- Little RR,
- Cowie CC,
- Widemeyer H-M,
- Eberhardt MS,
- Byrd-Holt DD
National Center for Health Statistics; Centers for Disease Control and Prevention. The National Health and Nutrition Examination Surveys: analytic and reporting guidelines. Last update: December 2005. Last correction, September 2006. http://www.cdc.gov/NCHS/data/nhanes/nhanes_03_04/nhanes_analytic_guidelines_dec_2005.pdf. Accessed February 17, 2010.
- Ostchega Y,
- Hughes JP,
- Wright JD,
- McDowell MA,
- Louis T
- de la Sierra A,
- Segura J,
- Banegas JR,
- Gorostidi M,
- de la Cruz JJ,
- Armario P,
- Oliveras A,
- Ruilope LM
- Chapman N,
- Dobson J,
- Wilson S,
- Dahlöf B,
- Sever PS,
- Wedel H,
- Poulter NR
- Taler SJ,
- Textor SC,
- Augustine JE
- Ford ES,
- Zhao G,
- Li C,
- Pearson WS,
- Mokdad AH
Defining the characteristics of uncontrolled hypertensive patients may facilitate efforts to improve blood pressure control. Data were analyzed for 13 375 hypertensive adults from the National Health and Nutrition Examination Surveys for 1988 to 1994, 1999 to 2004, and 2005 to 2008. Multivariable logistic regression was used to identify clinical characteristics associated with untreated hypertension, hypertension uncontrolled on 1 to 2 blood pressure medications, and apparent treatment-resistant hypertension on ≥3 blood pressure medications. More than half of uncontrolled hypertensives were untreated, including 52.5% in 2005 to 2008, with about two thirds of them unaware of their hypertension. Untreated hypertensive patients were more often men, infrequent users of primary health care, lean, and without clinical cardiovascular or renal disease. Most treated uncontrolled patients reported taking 1 to 2 blood pressure medications, a proxy for therapeutic inertia. This group, which was older and had higher 10-year coronary heart disease risk than patients controlled on 1 to 2 medications, comprised 34.4% of all uncontrolled and 72.0% of treated uncontrolled patients in 2005 to 2008. Apparent treatment-resistant hypertension increased from 15.9% (1998 to 2004) to 28.0% (2005 to 2008) of treated patients (P<0.001). Clinical characteristics associated with apparent treatment-resistant hypertension included ≥4 visits a year, obesity, chronic kidney disease, and 10-year coronary heart disease risk >20%. Adherence to health lifestyles and medications is important for all patients. Raising hypertension awareness among infrequent users of primary health care and linking them to a medical home could reduce untreated hypertension. Uncontrolled patients on 1 to 2 blood pressure medications could benefit from an additional antihypertensive medication. More effective selection of antihypertensive combination therapy is important for the growing proportion of patients with apparent treatment-resistant hypertension.