Author + information
- Received May 20, 2014
- Revision received August 25, 2014
- Accepted September 8, 2014
- Published online December 2, 2014.
- William B. Borden, MD∗∗ (, )
- Thomas M. Maddox, MD, MSc†,‡,
- Fengming Tang, MS§,
- John S. Rumsfeld, MD, PhD†,
- William J. Oetgen, MD, MBA‖,
- J. Brendan Mullen, BSFS¶,
- Sarah A. Spinler, PharmD (AQ Cardiology)#,
- Eric D. Peterson, MD, MPH∗∗ and
- Frederick A. Masoudi, MD, MSPH††
- ∗George Washington University, Washington, DC
- †VA Eastern Colorado Health Care System, Denver, Colorado
- ‡University of Colorado School of Medicine, Denver, Colorado
- §Mid America Heart Institute, Kansas City, Missouri
- ‖American College of Cardiology, Washington, DC
- ¶National Quality Forum, Washington, DC
- #Philadelphia College of Pharmacy, University of the Sciences, Philadelphia, Pennsylvania
- ∗∗Duke University, Durham, North Carolina
- ††Division of Cardiology, Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado
- ↵∗Reprint requests and correspondence:
Dr. William B. Borden, Department of Medicine, George Washington University, 2150 Pennsylvania Avenue, NW, Washington, DC 20037.
Background Since 2003, the Seventh Report of the Joint National Committee (JNC-7) has been the predominant guideline for blood pressure management. A 2014 expert panel recommended increasing the blood pressure targets for patients age 60 years and older, as well as those with diabetes or chronic kidney disease.
Objectives The purpose of this study was to examine the effect of the 2014 expert panel blood pressure management recommendations on patients managed in U.S. ambulatory cardiovascular practices.
Methods Using the National Cardiovascular Data Registry PINNACLE Registry, we assessed the proportion of patients who met the 2003 and 2014 panel recommendations, highlighting the populations of patients for whom the blood pressure goals changed.
Results Of 1,185,253 patients in the study cohort, 706,859 (59.6%) achieved the 2003 JNC-7 goals. Using the 2014 recommendations, 880,378 (74.3%) patients were at goal. Among the 173,519 (14.6%) for whom goal achievement changed, 40,323 (23.2%) had a prior stroke or transient ischemic attack, and 112,174 (64.6%) had coronary artery disease. In addition, the average Framingham risk score in this group was 8.5 ± 3.2%, and the 10-year ASCVD risk score was 28.0 ± 19.5%.
Conclusions Among U.S. ambulatory cardiology patients with hypertension, nearly 1 in 7 who did not meet JNC-7 recommendations would now meet the 2014 treatment goals. If the new recommendations are implemented in clinical practice, blood pressure target achievement and cardiovascular events will need careful monitoring, because many patients for whom the target blood pressure is now more permissive are at high cardiovascular risk.
The treatment of hypertension is a cornerstone for the prevention of stroke, acute coronary events, chronic heart failure, and kidney disease. Clinicians rely on national clinical practice guidelines for recommendations on whom to treat, when to treat, to what goals, and with what medications. This guidance is particularly important, because of the need to balance the benefits of cardiovascular risk reduction with the costs and potential complications of therapy.
In 2003, the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7) was widely implemented by the medical community (1). This report established blood pressure goals of <140/90 mm Hg for the general population and <130/80 mm Hg for patients with diabetes and/or chronic kidney disease. A follow-up Eighth Joint National Committee (JNC-8) guideline was initially drafted with funding from the National Heart, Lung, and Blood Institute. The National Heart, Lung, and Blood Institute ultimately decided not to proceed with the updated guideline, instead focusing on research and providing support for professional societies to write their own advisories (2). However, the expert panel members who were initially appointed to JNC-8 chose to independently publish their recommendations. The primary difference between JNC-7 guidelines and the 2014 panel recommendations was the raising of some of the blood pressure treatment targets (3), including revised targets of <150/90 mm Hg for patients age 60 years and older and <140/90 mm Hg for those with diabetes and/or chronic kidney disease.
The 2014 expert panel report represents a major shift in the treatment of hypertension, and concern has been expressed about the public health impact of these less aggressive recommendations on efforts to prevent cardiovascular disease. Despite the substantial improvements in cardiovascular disease morbidity and mortality in the United States over the past few decades, only 50% of patients with hypertension currently receive treatment; thus, critics—including some members of the JNC-8 panel writing group itself—argue that the 2014 panel recommendations could slow, halt, or even reverse these gains (4), especially among at-risk patient groups, such as elderly women and African Americans (5). One recent analysis of a U.S. population survey estimated that 5.8 million fewer patients would require blood pressure treatment according to the new report (6). Whereas those data extrapolated the potential effect of the 2014 panel recommendations in the general population, we sought to understand the potential effect on national U.S. cardiovascular practice, using data from the National Cardiovascular Data Registry (NCDR) Practice Innovation and Clinical Excellence (PINNACLE) clinical registry program. This study assessed the proportion of patients managed in cardiovascular outpatient practices meeting the 2003 JNC-7 guideline and the 2014 panel recommendations and characterized the patients for whom the blood pressure targets changed under the new recommendations.
The NCDR PINNACLE Registry served as the study data source. Cardiology practices voluntarily participate in and submit data to the PINNACLE Registry as part of a national office-based cardiovascular quality improvement program (7,8). The data are collected at the point of care through either a paper chart abstraction form or a validated electronic health record mapping algorithm designed to comprehensively capture required data elements. These data elements include demographics, insurance status, and detailed clinical information, including symptoms, medical conditions, vital signs, medications, and laboratory values. Registry data quality assurance is maintained through rigorous data definitions, standard data collection and transmission, and periodic data quality checks (9,10).
We identified all patients age 18 years or older with a diagnosis of hypertension with clinical encounters in the PINNACLE Registry version 1.2 between January 1, 2008, and December 31, 2012. A diagnosis of hypertension was determined through a chart notation of the diagnosis, a measured systolic blood pressure >140 mm Hg, or a diastolic blood pressure >90 mm Hg. In addition, we performed a sensitivity analysis by limiting the population to those patients who met blood pressure criteria on 2 or more occasions. We excluded patients with missing blood pressure measurements and those with implausible diabetes mellitus history data, presumably reflecting data-entry errors.
Clinical outcomes and characteristics
The primary study outcome measure was whether patients with hypertension would meet the 2003 JNC-7 guidelines and/or the 2014 panel treatment goal recommendations. Patients met the 2003 JNC-7 guidelines if they had a blood pressure of <140/90 mm Hg or <130/80 mm Hg for patients with diabetes mellitus. For comparison purposes to the 2014 panel recommendations, we also specifically assessed patients age 60 years and older who met a blood pressure goal of <140/90 mm Hg. Patients met the 2014 panel recommendations if they had a blood pressure of <140/90 mm Hg if they were age 18 to 59 years or had diabetes mellitus at any age, and <150/90 mm Hg if they were age 60 years or older. We were unable to assess for a history of chronic kidney disease, because such data were not recorded in the PINNACLE Registry version 1.2. If a patient had multiple clinical encounters during the study period, then the most recent blood pressure measurement was used. Sensitivity analyses were performed using the first blood pressure measurement and an average of blood pressure measurements.
Patients were classified as having diabetes mellitus if they had a clinical diagnosis of diabetes mellitus, a hemoglobin A1c ≥6.5%, a fasting blood glucose ≥126 mg/dl, or any blood glucose of ≥200 mg/dl. Patients were designated as having coronary artery disease if they had a clinical diagnosis of coronary artery disease or if they had a prior history of coronary artery bypass surgery or percutaneous coronary intervention. We calculated 10-year Framingham risk scores (11) and 10-year atherosclerotic cardiovascular disease (ASCVD) risk scores using the Pooled Cohorts Equation in the 2013 American College of Cardiology and American Heart Association Guideline on the Assessment of Cardiovascular Risk (12,13). We also assessed the antihypertensive medication classes that patients were prescribed, including diuretics, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, beta-blockers, and calcium-channel blockers.
The analyses performed were descriptive of patients with various characteristics meeting the 2003 JNC-7 and the 2014 panel recommendations. Patients with diabetes mellitus were categorized accordingly and were not included in the age categories. We first described the demographic and clinical characteristics of patients achieving recommended goals, including their prescribed medications. We then assessed the patients who met the 2014 panel recommendations but did not meet the 2003 JNC-7 guidelines. These patients had diabetes mellitus and blood pressure ≥130/80 mm Hg and <140/90 mm Hg, or, if age 60 years and older, a blood pressure ≥140/90 mm Hg and <150/90 mm Hg. All analyses were performed with SAS version 9.3 (SAS Institute, Cary, North Carolina).
The PINNACLE Registry included data on 1,711,326 patients from 2008 to 2012. Among the 1,344,526 patients with hypertension, 1,699 were excluded because they were <18 years of age, 39,345 were excluded because they lacked blood pressure measurements, and 118,229 were excluded because of practices reporting implausible diabetes rates (0.1%, 2.9%, and 8.8% of the total population with reported hypertension, respectively). The final study population consisted of 1,185,253 patients.
In this cohort, 706,859 (59.6%) achieved the 2003 JNC-7 guideline goals and 880,378 (74.3%) achieved the 2014 panel recommendations, with a population of 173,519 (14.6%) who met the 2014 recommendations but not the 2003 guidelines. Because the goal was the same for patients <60 years of age without diabetes, 195,200 (68.0%) met both the 2003 guidelines and the 2014 recommendations, whereas among patients 60 years of age and older without diabetes, 385,779 (66.9%) met the 2003 goals and 466,002 (80.8%) met the 2014 panel recommendation goals. Among patients with diabetes mellitus, 125,880 (39.2%) met the 2003 guideline goals and 219,176 (68.2%) met the 2014 recommendation goals (Central Illustration). The characteristics of patients meeting the 2003 JNC-7 guidelines and the 2014 panel recommendations are described in Table 1 and Online Table 1. The relative frequency of antihypertensive drug classes prescribed was similar in these 2 populations; beta-blockers were prescribed in >50% of patients within each group followed by diuretics, angiotensin-converting enzyme inhibitors, calcium-channel blockers, and angiotensin receptor blockers (Table 1).
The characteristics of the 173,519 patients for whom the recommendations differ are listed in Table 1. Overall, 112,174 (64.6%) had a coronary artery disease diagnosis, 93,296 (53.8%) had diabetes mellitus, 46,910 (27.0%) had a history of heart failure, 43,984 (25.3%) had a prior myocardial infarction, and 40,323 (23.2%) had a prior stroke or transient ischemic attack (TIA). In calculating the risk scores, 646,895 patients meeting 2003 JNC-7 guideline goals, 806,314 patients meeting the 2014 panel recommendations, and 159,419 patients for whom the recommendations changed were missing data for the 10-year Framingham risk score. There were missing data for the ASCVD risk score for 551,228 patients meeting 2003 JNC-7 goals, 683,694 patients meeting 2014 panel recommendations, and 132,466 patients for whom the recommendations changed. Patients with missing risk score data were generally younger and had fewer comorbidities. Among those for whom the recommendations changed with calculable risk scores, the average Framingham risk score was 8.5 ± 3.2%, and the 10-year ASCVD risk score was 28.0 ± 19.5%. By comparison, the patients meeting the 2003 JNC-7 guideline goals and the 2014 panel recommendation goals had average Framingham risk scores of 4.8 ± 4.0% and 5.5 ± 4.2% and 10-year ASCVD risk scores of 18.6 ± 16.8% and 20.6 ± 17.8%, respectively (Central Illustration). The sensitivity analyses yielded similar results.
This analysis conducted in a real-world cardiology practice clinical registry demonstrates that about three-quarters of patients with hypertension met the 2014 panel recommended blood pressure goals, whereas less than two-thirds met the 2003 JNC-7 guideline goals. Among those eligible for treatment who did not meet JNC-7 goals, 14.6% would meet the 2014 panel goals, which was similar to a recent national population analysis that estimated 15.9% (6). With the more permissive 2014 blood pressure targets, patients meeting the new goals had significantly higher cardiovascular risk than those meeting the 2003 guidelines. Specifically, the patients for whom the recommendations changed had an average 10-year risk of myocardial infarction or death of 8.5%, and, when stroke risk was included, it rose to 28%. The 2014 panel recommendations are based on randomized-controlled trial data. However, if these recommendations are adopted by practitioners treating real-world populations of high-risk patients such as those in this study, close monitoring, particularly with clinical registries, will be necessary to ensure that improvements in cardiovascular event rates, particularly stroke, do not reverse course.
An estimated 78 million Americans have hypertension, representing 1 in every 3 adults overall and 2 in every 3 adults age 60 years and older (14). Beginning in the 1970s, substantial public health and clinical efforts have taken hold, with improvements in blood pressure awareness, treatment, and control. Hypertension control improved from 27.3% in 1988 to 1994 to 50.1% in 2007 to 2008, although substantial opportunities clearly still exist to further improve that risk factor control (15). Coincident with this improvement in blood pressure control and control of other risk factors has been a strong steady decline in cardiovascular morbidity and mortality. This association is especially clear with stroke, where an almost linear epidemiologic relationship exists between blood pressure and stroke risk (16). The United States has seen a nearly 4-fold reduction in stroke deaths with age-adjusted stroke mortality rate, decreasing from 88 per 100,000 in 1950 to 23 per 100,000 in 2010 (16).
The 2014 panel recommendations concluded that the existing randomized-controlled trial evidence was insufficient to support a systolic blood pressure goal of 140 mm Hg in people age 60 years and older, unless they were already tolerating such treatment, and thus raised the goal to 150 mm Hg. The epidemiologic association between improved population blood pressure control and decreased cardiovascular morbidity and mortality is strong, and the clinical trials demonstrating the benefit of treating high blood pressure are convincing. In a meta-analysis of 72 randomized controlled trials, Law et al. (17) showed that lowering systolic blood pressure by 10 mm Hg and diastolic blood pressure by 5 mm Hg reduced coronary heart disease events by 22% and strokes by 41%, including a 15% reduction in coronary heart disease events and a 23% reduction in strokes in trials where the pre-treatment systolic blood pressure was 140 to 159 mm Hg. Evidence exists for treating to a systolic blood pressure target of about 140 mm Hg in both younger and older individuals (4,18–20). However, when specifically using randomized-controlled trial evidence to address the question of a blood pressure target for people older than age 60 years, the 2014 panel members felt the data did not demonstrate additional benefit to a systolic blood pressure goal of <140 mm Hg (3,21,22).
That recommendation was considered controversial, and a faction of the writing group felt strongly enough to publish their perspective (4). A minority of the panel dissented with that recommendation, because they thought there was insufficient evidence of harm with the existing goal to make the change, especially in light of the significant gains experienced in cardiovascular morbidity and mortality. These dissenting panel members felt that the burden of evidence was on demonstrating a clear benefit to increasing the blood pressure goal and that such evidence was inadequate (4). Our study lends support to these concerns by demonstrating the substantial cardiovascular risk in the patients for whom the recommendations change and highlights the uncertainty of how the new recommendations will ultimately affect real-world patients.
In light of the concerns about loosening the blood pressure goal for older patients and those with diabetes mellitus, our study provides a detailed perspective on patients receiving care in ambulatory cardiology practices. Our findings demonstrate that these patients, particularly those for whom the blood pressure goals changed, are at baseline substantial risk of coronary heart events and stroke. Several studies have specifically examined blood pressure lowering in older adults. The SHEP (Systolic Hypertension in the Elderly Program) trial, in which the mean age was 72 years, showed that a reduction in mean systolic blood pressure from 155 to 143 mm Hg was associated with a 32% reduction in cardiovascular events at 5 years (18). The HYVET (Hypertension in the Very Elderly Trial) showed a similar result in patients with a mean age of 83.6 years, in whom a reduction in mean systolic blood pressure from 158.5 to 143.5 mm Hg resulted in a 34% reduction in cardiovascular events at 2 years (19). Although these SHEP and HYVET findings show achieved blood pressures, rather than the goal blood pressures in the 2014 panel recommendations, the results inform how measured blood pressure changes are associated with reductions in cardiovascular events.
Using these randomized controlled trial risk reduction results and the 10-year ASCVD risk of 27.9% estimated for the older population in the PINNACLE Registry for whom the adequacy of blood pressure control using recommended targets changed, treating patients from a systolic blood pressure of about 150 to about 140 mm Hg is estimated to result in a 9- to 10-point 10-year ASCVD absolute risk reduction to 19.0% using the SHEP data and 18.4% using HYVET data. This is equivalent to a number-needed-to-treat over 10 years of 10 to 11 patients to prevent 1 cardiovascular event. This estimate is consistent with national projections in a similar at-risk population that a 12 mm Hg reduction over 10 years from a baseline systolic blood pressure of 140 to 159 mm Hg would result in a number-needed-to-treat of 11 patients to prevent a cardiovascular event (23). Using these data about patients age 60 years and older, treatment of the more than 80,000 older patients in our study population affected by the change in targets to a systolic blood pressure goal of 140 mm Hg could potentially avert approximately 8,000 cardiovascular events over 10 years.
These estimates of averted cardiovascular events do not account for potential adverse treatment effects or the costs of therapy. However, they highlight the substantial absolute risk of patients affected by the 2014 panel report recommendations and the potential for lost treatment benefit. On a national scale, 1 analysis estimates that 13.5 million patients with hypertension who had previously not met JNC-7 guideline goals would now meet treatment goals in 2014 recommendations (6). Our study population from cardiology practices has higher risk than the general population, yet represents an important cohort with profound implications for the broader U.S. population.
First, our study population is restricted to U.S. cardiology practices that participate in the PINNACLE Registry. The population in our study has greater cardiovascular comorbidity and better blood pressure control than a general population sample (14). Although this can be construed as a weakness, we believe that our study provides important insights more broadly to similarly high-risk real-world individuals with hypertension. Second, there may be variability in the measurement of blood pressure. Some PINNACLE Registry patients only have a single blood pressure measurement, which may not be reflective of their overall blood pressure pattern. Our sensitivity analysis of using more than 1 blood pressure measurement yielded similar results. Third, the PINNACLE Registry obtains data directly from clinical care processes, mostly from electronic health records rather than specialized research abstracters, and thus, data collection may be more incomplete. Last, missing data may have led to an overestimation of the 10-year Framingham and ASCVD risk scores. For example, the average age was 62.4 and 64.0 years, the rate of diabetes mellitus was 19.1% and 23.8%, and prior stroke or TIA occurred in 11.8% and 15.9% of patients missing and not missing Framingham risk scores, respectively. Similarly, the average age was 65.4 and 68.0 years, the rate of diabetes mellitus was 25.8% and 31.8%, and prior stroke or TIA occurred in 21.9% and 22.8% of patients missing and not missing ASCVD risk scores, respectively. Even though the risk scores may be overestimated, the fundamental finding of elevated cardiovascular risk in this population is unchanged.
This study found that 14.6% of patients in ambulatory cardiology practices who previously failed to meet the 2003 JNC-7 guidelines recommendations now meet the 2014 panel recommendations. These patients have high rates of prior cardiovascular disease and high 10-year risk of cardiovascular events. With the several other blood pressure management guidelines available, it remains to be seen how practice patterns change based on the 2014 panel recommendations. Some of that uptake may be influenced by whether the 2014 panel recommendations are incorporated into performance measures, public reporting, and value-based purchasing programs. Given the size and underlying cardiovascular risk of the population affected by the changes in the 2014 panel recommendations, close monitoring will be required to assess changes in practice patterns, follow blood pressure control, and, importantly, determine any changes in cardiovascular morbidity and mortality.
COMPETENCY IN SYSTEMS-BASED PRACTICE: The 2014 expert panel recommends higher blood pressure targets than the 2003 JNC-7 report, so fewer patients meet criteria for treatment of high blood pressure.
TRANSLATIONAL OUTLOOK: Clinical outcomes studies are needed to gauge the effect of application of the 2014 expert panel recommendations on cardiac events in well-defined patient populations.
This research was supported by the American College of Cardiology Foundation’s National Cardiovascular Data Registry (NCDR). The views expressed in this paper represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at http://www.ncdr.com. Dr. Borden was supported as a Nanette Laitman Clinical Scholar while he was at Weill Cornell Medical College. Dr. Maddox is supported with a Veterans’ Affairs career development award. Dr. Rumsfeld is the Chief Science Officer for the National Cardiovascular Data Registry. Dr. Peterson has received research financial support from Eli Lilly, Janssen, AstraZeneca, and Boehringer Ingelheim; and is the principal investigator of one of the NCDR's Data Analytic Centers. Dr. Masoudi is the Senior Medical Officer for the National Cardiovascular Data Registry; and has a contract with the American College of Cardiology. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atherosclerotic cardiovascular disease
- Joint National Committee
- National Cardiovascular Data Registry
- Practice Innovation and Clinical Excellence
- transient ischemic attack
- Received May 20, 2014.
- Revision received August 25, 2014.
- Accepted September 8, 2014.
- American College of Cardiology Foundation
- Gibbons G.H.,
- Shurin S.B.,
- Mensah G.A.,
- et al.
- James P.A.,
- Oparil S.,
- Carter B.L.,
- et al.
- Krakoff L.R.,
- Gillespie R.L.,
- Ferdinand K.C.,
- et al.
- Chan P.S.,
- Oetgen W.J.,
- Buchanan D.,
- et al.
- Messenger J.C.,
- Ho K.K.,
- Young C.H.,
- et al.
- Masoudi F.A.,
- Ponirakis A.,
- Yeh R.W.,
- et al.
- Wilson P.W.,
- D’Agostino R.B.,
- Levy D.,
- et al.
- Goff D.C.,
- Lloyd-Jones D.M.,
- Bennett G.,
- et al.
- American Heart Association. 2013 prevention guidelines tools: CV Risk Calculator. 2013. Available at: https://my.americanheart.org/professional/StatementsGuidelines/PreventionGuidelines/Prevention-Guidelines_UCM_457698_SubHomePage.jsp. Accessed April 1, 2014.
- Go A.S.,
- Mozaffarian D.,
- Roger V.L.,
- et al.
- Lackland D.T.,
- Roccella E.J.,
- Deutsch A.F.,
- et al.
- Law M.R.,
- Morris J.K.,
- Wald N.J.