Cardiovascular Risk Factor Control in Communities—Update From the ASH Carolinas‐Georgia Chapter, the Hypertension Initiative, and the Community Physicians' Network

Abstract

The prevalence of hypertension dictates that blood pressure must be managed effectively in primary care. The American Society of Hypertension (ASH) regional chapters and clinical hypertension specialists represent a positive response by ASH to the growing problems of hypertension and metabolic syndrome—related risks and disease. To have a significant public health effect, the impact of clinical hypertension specialists must be leveraged. Key activities in the community include educating other providers locally, delivering care for complex referral patients, and fostering growth of a practice network with a central database in collaboration with academic partners. The database supports practice audit and feedback reports to enhance quality improvement, identify continuing medical education topics, and facilitate clinical trials to test new therapeutic and best‐practice approaches to risk factor management. The ASH regional chapters serve as a forum for community and academic hypertension specialists to collaborate with like‐minded individuals and organizations. The collaboration among the ASH Carolinas—Georgia chapter, the Hypertension Initiative, and the Community Physicians' Network provides a model for other ASH chapters and health delivery groups to partner in delivering continuing medical education programs focused on cardiovascular risk factor management, recruiting practices into the network, and developing and maintaining a centralized patient database. Evidence suggests that this collaboration is facilitating application of evidence‐based medicine and risk factor control.

Hypertension affects approximately 65 million people in the United States. ¹ The National Health and Nutrition Examination Survey (NHANES) 1999–2000 data indicate that approximately 15% of normotensive, 30% of prehypertensive, and 60% of hypertensive patients may have the metabolic syndrome. ² The growth in numbers of hypertensive patients and patients with the metabolic syndrome is likely to continue, since US citizens are becoming older, more obese, and more ethnically diverse. ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ Cardiovascular and renal disease will likely escalate in the absence of more effective prevention. Treatment and control of hypertension, hyperlipidemia, and diabetes reduce cardiovascular morbidity and mortality. ⁸ , ⁹ , ¹⁰ Translating clinical trial efficacy into effective community‐based practice, however, has proven challenging. ¹¹ , ¹²

A single approach is unlikely to close the gap between efficacious treatment in clinical trials and effective clinical practice, given local and regional differences in demographics, socioeconomic and educational status, population density, health attitudes and beliefs, medical care delivery systems, and payment plans. Facilitating positive change in preventing cardiovascular and renal disease requires an approach appropriate for the community and region. In the southeastern United States, the prevalence of obesity, hypertension, diabetes mellitus, sedentary lifestyles, and high‐fat diets is greater, and the proportion of African Americans is larger, than in other regions of the United States. ¹³ The Southeast has less managed care and tends to have more uninsured and federally insured patients than other regions. ¹⁴ These factors may contribute to higher rates of stroke, total cardiovascular disease, and end‐stage renal disease in the Southeast. ¹³

The number of clinical hypertension specialists (CHSs) is unlikely to be sufficient to directly manage or provide referral care for all uncontrolled hypertensive patients. It is important, therefore, to leverage the impact of each CHS through activities in addition to patient care. There are 3 vital activities: (1) raising awareness of the importance and feasibility of controlling hypertension, hyperlipidemia, and diabetes; (2) educating physicians and other health care providers locally on best‐practice approaches to cardiovascular risk reduction (ie, fostering an opinion leader); and (3) helping to identify key topics for continuing medical education (CME) programs and guide the evaluation of quality improvement initiatives to prevent and treat cardiovascular risk factors and outcomes.

The American Society of Hypertension (ASH) regional chapters, initiated and supported by ASH, provide a constructive forum for individuals and groups with complementary resources and expertise to collaborate on a common mission. The ASH Carolinas—Georgia chapter has partnered with the Hypertension Initiative and the Community Physicians' Network (CPN) to form a clinical network to foster improvements in cardiovascular risk factor management and reduce ethnic disparities in health care in the Southeast. This paper describes the partnership and reports certain results from the 2000–2005 period.

ASH CAROLINAS—GEORGIA CHAPTER

The ASH Carolinas—Georgia chapter, incorporated in June 2001, serves as a forum for CHSs from academic centers and the community to improve the awareness, prevention, treatment, and control of cardiovascular risk factors. Educating and assisting health care providers in applying evidence‐based medicine in daily clinical practice is central to this mission. The chapter provides multiple CME programs for physicians and other health care providers in the tri‐state region. ¹⁵

The ASH Carolinas—Georgia chapter facilitates growth in the number and regional geographic diversity of ASH CHSs. One goal is to have one CHS for every 20 primary care physicians with a distribution approximating that of the local population. ¹⁶ The chapter endeavors to raise scholarship funds for qualified physicians to participate in the intensive hypertension CME course that has been offered biennially by ASH before the CHS certification examination.

HYPERTENSION INITIATIVE

The Hypertension Initiative, based at the Medical University of South Carolina, began in April 1999 with the goal of facilitating the state's transition from a leader in cardiovascular disease to a model of cardiovascular health. ¹⁷ The Hypertension Initiative initially focused on primary care providers. The provider focus began with the Experts in Hypertension seminar series, a CME‐accredited program, funded primarily by unrestricted grants from the pharmaceutical industry.

CME activities shared the following objectives: (1) raise awareness of the epidemiology of hypertension and the feasibility of improving control of hypertension, hyperlipidemia, and diabetes; (2) educate providers about evidence‐based guidelines and clinical trials for each of these 3 medically modifiable major risk factors that can positively impact daily practice; (3) facilitate participation in the practice network and database; and (4) encourage physicians to become CHSs. The tri‐state now has 97 CHSs. The impact of these CHSs on the control of BP among approximately 5 million hypertensive patients in the Carolinas—Georgia region through direct patient care is necessarily modest, however, requiring strategies for leveraging the expertise of the CHSs throughout the clinical practice network.

The Hypertension Initiative Clinical Practice Network and Database

An essential mission of the Hypertension Initiative is the participation of the primary care providers throughout the clinical practice network in the network database. More than 80% of patient data are obtained from a variety of electronic medical record systems (EMRS) in practices with this technology. The database is audited quarterly. Data originate from a variety of practice settings, including community clinics, Veterans Affairs (VA) primary care clinics, residents' (physicians in training) clinics, and solo and group practices. Only one blood pressure (BP) measurement per visit is recorded in the database, with the last BP measurement in the seated position receiving the highest priority. The analytic database contains patient demographics, vital signs, diagnoses, medications, and laboratory values.

Quarterly Provider Reports

Providers in the practice network receive quarterly feedback reports describing the demographic characteristics of their patient panel. Reports include mean values for systolic and diastolic BP in all hypertensive patients and for patients with diabetes and renal disease, the percentage of patients achieving various BP goals, and the frequency of use of the various classes of antihypertensive medications. Comparable values are provided with respect to patients with hyperlipidemia and diabetes. The reports compare each provider's patient panel to all patients in the network. The reports are not graded and relative performance is not shared with other providers or clinics or third‐party payors. Current reports include a continuity of care index (percent of patients seen in the past 6 months), and future reports will include a therapeutic inertia index. The Hypertension Initiative has created structured queries that can be applied to data obtained from practices with an EMRS to assess quality performance indicators (Health Plan Employer Data and Information Set [HEDIS] and Doctor's Office Quality—Information Technology [DOQ‐IT]) for cardiovascular and other health conditions. Some indicators align with pay‐for‐performance criteria, but the Hypertension Initiative has no plans to share this information with third‐party payors without expressed written consent from providers and clinics.

THE COMMUNITY PHYSICIANS' NETWORK

The CPN, the third member of the collaboration, is a consortium of approximately 140 practices in Georgia. ¹⁸ , ¹⁹ , ²⁰ , ²¹ The CPN, which was established in 2002 with funds from the National Center for Minority Health and subsequent support from the Medtronic Foundation, promotes participation of African American and other minority physicians and patients in quality improvement initiatives and clinical trial protocols.

The CPN includes mainly urban clinics serving predominantly low‐income African American patients; the Hypertension Initiative comprises clinics in rural, suburban, and smaller urban environments and includes clinics that collectively serve patients with more diverse demographic and economic characteristics. The Hypertension Initiative provides expertise in data acquisition, management, and reporting, ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ while the CPN provides professional and community consultants to enhance the effectiveness of the academic—community partnership. ¹⁸ , ¹⁹ , ²⁰ , ²¹ At present, data from the CPN practice sites generally include only one time point. The CPN and Hypertension Initiative are working to implement serial data monitoring at all sites to support the data audit and feedback program and facilitate the design and evaluation of quality improvement initiatives.

PROGRESS REPORT, 2000–2005

The number of practices and patients in the network and database has grown over the past 5–7 years. The combined database now includes information on 521,068 patients with visits during 2004–2005 and includes 313,034 hypertensive, 290,933 hyperlipidemic, 118,979 diabetic, and 22,760 nephropathic patients (Table). Roughly half of the patients are from VA clinics, where approximately 97% of patients are men, which accounts for the fact that 69% of all patients in the database are men. Men also represent a higher percentage of patients with risk factors.

Table.

Selected Descriptive Characteristics for 521,068 Patients With Visits During 2004–2005 and the Subsets With Hypertension (HTN), Hyperlipidemia (HLP), Diabetes Mellitus, and Nephropathy

Characteristic	All	HTN	HLP	Diabetes	Nephropathy
No. of patients	521,068	313,034	290,933	118,979	22,760
Age, mean ± SD, y	56±19	64±14	63±14	64±14	69±13
Men/women, %	69/28	79/20	80/19	80/19	89/11
White—black ratio	23:12	23:14	24:12	23:16	25:21
HTN present, %	60	100	81	93	98
Mean BP, mm Hg
All patients	129/75	134/76	131/75	132/74	134/73
HTN patients only	134/76	134/76	133/75	133/74	134/73
BP control levels, %*
<150/95 mmHg	86/81	81/81	85/83	83/82	79/79
<140/90 mm Hg	72/63	63/63	69/65	66/64	63/62
<130/80 mmHg	40/31	31/31	36/32	35/33	36/35
HLP present, %	56	75	100	82	85
LDL, mean ± SD, mg/dL
All patients	110±33	107±33	114±33	102±33	99±34
HLP patients only	114±33	111±33	114±33	104±33	101±34
LDL control levels, %*
<130 mg/dL	74/71	77/74	71/71	82/80	83/81
<100 mg/dL	42/33	45/38	33/33	53/49	56/52
Diabetes present, %	23	35	34	100	62
HbA1c <7%, %†	53/40	53/41	53/43	53/40	50/44
BP indicates blood pressure; LDL, low‐density lipoprotein cholesterol; and HbA1c, glycosylated hemoglobin. *First percentage denotes control for all patients in column, while the second percentage denotes control only for the patients with a diagnosis of that risk factor (HTN or HLP). †Control to HbA1c <7% is shown only for diabetic patients. The first percentage denotes percentage of diabetics with HbA1c <7% for those with an HbA1c value, while the second percentage is for all diabetic patients, assuming those without a value have values >7%.

Quality Improvement

Among a cohort of 208,517 hypertensive patients with at least 5 visits in different 6‐month periods from 2000–2005, BP control to <140/90 mm Hg increased steadily (Figure 1). Among a cohort of 82,442 diabetic hypertensive patients meeting the same visit criterion, control of BP to <130/80 mm Hg, low‐density lipoprotein cholesterol to <100 mg/dL, and glycosylated hemoglobin to <7% showed progressive improvement (Figure 2, top). Among this same group of hypertensive diabetic patients, the percentage receiving an angiotensin‐converting enzyme (ACE) inhibitor and/or angiotensin receptor blocker (ARB) has risen steadily over time (Figure 2, bottom).

Figure 1.

Serial data for blood pressure (BP) control to <140/90 mm Hg in a cohort of 208,517 hypertensive patients, including diabetic hypertensives, who had BP values during at least 5 visits in different 6‐month periods from January 2000 through January 2005.

Figure 2.

(Top) Serial data for blood pressure (BP) control to <130/80 mm Hg, low‐density lipoprotein cholesterol (LDL) to <100 mg/dL, and glycosylated hemoglobin (HbA_1c) to <7% for a cohort of 82,442 diabetic patients with at least 5 visits in different 6‐month periods from January 2000 through January 2005. LDL data are provided only for the 61,075 patients with a diagnosis of hyperlipidemia and LDL values on a minimum of 5 visits in different 6‐month periods from 2000 to 2005. (Bottom) The percentage of diabetic patients receiving an angiotensin‐converting enzyme inhibitor (ACE) and/or an angiotensin receptor blocker (ARB) from 2000 to 2005.

Many factors are likely responsible for the positive trends in risk factor control and application of evidence‐based medicine. The factors may include CME programs provided by the ASH Carolinas—Georgia chapter and the CPN, the practice data audit and feedback reports from the Hypertension Initiative, and efforts to leverage the CHSs. The VA clinics and family health centers have also implemented efforts to improve cardiovascular risk factor control. We have not attempted to parse the contribution of various factors to the improvement in risk factor control and evidence‐based prescribing patterns. These data, nevertheless, document the magnitude of progress and serve to identify and focus CME program content and other interventions to drive quality improvement. One example of an observation from the database that provides the basis for intervention is the magnitude and impact of therapeutic inertia on hypertension control rates in our clinical practice network. ²⁷

The network database is used to identify CME program topics that are more or less useful and interventions to reduce disparities in treatment. For example, many CME programs emphasize an underuse of ACE inhibitors and ARBs in African Americans, who are more likely than whites to experience the target organ complications ameliorated by these agents. Our database, on the other hand, has consistently shown that ACEIs and ARBs are prescribed equally to African American and white hypertensive patients. ²⁴ , ²⁵ , ²⁶

In network practices, disparities in hypertension control do not appear to reflect differences in routine diagnostic testing and prescribing patterns for antihypertensive medications. ²² , ²³ , ²⁴ , ²⁵ Disparities in control of hyperlipidemia, in contrast, represent in part the underprescribing of statins and other lipid‐lowering medications to African Americans. ²² , ²³ While control rates for both hypertension and hyperlipidemia are lower for African Americans than whites in the practice network, the underlying factors are not the same for both risk factors. These data provide the basis for more focused, and hopefully more effective, interventions to reduce disparities.

Ethnic disparities in hypertension control for black and white men were approximately 40% less at VA clinics than at non‐VA clinics in the network. ²⁶ This finding was explained mainly by higher rates of BP control among African American men at VA than non‐VA clinics. These data suggested that more open access to care and medications at VA than non‐VA clinics was particularly beneficial to BP control in African American men. The observation raises the possibility that improving access to care and medications would lead to a substantial narrowing of ethnic disparities in hypertension control.

The CPN hypertension registry has provided important information on the status of treatment of African American and other minority patients ¹⁸ , ¹⁹ : 90% of patients in the registry are African Americans, 68% are women (mean age, 47±13 years), 32% have preferred provider organization insurance, 30% have health maintenance organization insurance, 30% have Medicare or Medicaid, and 5.3% pay individually or are uninsured. Eleven percent of patients included in the CPN hypertension registry have <8 years of formal education. They have multiple comorbid conditions: diabetes (44%), premature coronary heart disease (13%), renal disease (8%), left ventricular hypertrophy on electrocardiogram (9%), chronic heart failure (5%), and previous myocardial infarction (5%). Overall, 36.1% are at goal BP; among diabetics, however, only 17.3% are at goal BP.

These findings have informed CME programs. The CPN has conducted 6 major regional CME educational initiatives, reaching more than 800 primary care physicians, over the past 24 months. These CME programs include topics on hypertension and the metabolic syndrome. ¹⁸ , ¹⁹ , ²⁰ , ²¹

Future Directions

Although the partnership has enjoyed initial success in growth of the network and, more importantly, in control of risk factors, there is still room for improvement in the control of hypertension, hyperlipidemia, and diabetes. ²² , ²³ , ²⁴ , ²⁵ , ²⁶ To have a sustained impact on improving control rates for hypertension, hyperlipidemia, and diabetes, and reducing ethnic disparities, it appears that our current fragmented system of care delivery will have to undergo significant transformation to an integrated patient‐centered system. Such systems will include an EMRS and personal health records, which should regularly provide data to a patient data repository. ²¹

To facilitate further improvements in the management of cardiovascular risk factors and other chronic health conditions, queries have been written to include pay‐for‐performance indicators to providers in participating practices using an EMRS. The Hypertension Initiative has no plans to provide these data to payors unless specifically requested in writing by the provider.

Clinics serving rural areas, ethnic minorities, and uninsured or underinsured patients are well represented in the network and database; however, many of these clinics do not have EMRS and they report that cost is a barrier to acquiring this technology. The partnership is vigorously pursuing various options to address this gap.

The patient database could serve to better inform our understanding of individual variability in response to pharmacologic and lifestyle measures. One intriguing possibility is establishing a DNA bank that could be used to identify genetic markers predictive of and/or associated with biologic variability in responses to and outcomes of treatment.

Another promising area identified through the database is therapeutic inertia. Analyses of patients seen 4 or more times in 2003 with at least one uncontrolled BP confirmed previous reports that therapy is intensified in only a very small proportion of patient visits (approximately 1 in 8). ²⁷ The analyses further suggested that reducing therapeutic inertia by half could get BP control rates to levels reported in clinical trials (at least 65%), which are essential to achieving the Healthy People 2010 goal of BP readings <140/90 mm Hg in half of all hypertensive patients in the community. ²⁸ BP control rates in our database for the end of 2004 and beginning of 2005 indicate control rates approximating 65%. It will be of interest to determine whether a decline in therapeutic inertia is contributing to the steady improvement in risk factor control.

Geocoding and Rural Urban Classification Assignment (RUCA) data can be used to refine assessment of health disparities. For example, significant geographic variation in BP control has been observed. Geographic variation in ethnic disparities of BP control was substantial, with some zip codes showing few (<5%) and other zip codes showing much larger (>15%) black—white differences. ²³ Geocoding and RUCA data can help inform interventions and target implementation to improve resource use and outcomes.

SUMMARY

The ASH Carolinas—Georgia Chapter, CPN, and Hypertension Initiative represent a partnership with a common mission of facilitating the transition of the tri‐state region from a leader in cardiovascular disease ²⁹ , ³⁰ to a model of cardiovascular health, with elimination of health disparities. Other such efforts have typically appeared to be resource intensive and often include a small number of practices using a common EMRS. In contrast, the present cost of maintaining our infrastructure is approximately $1 per patient annually, with a comparatively larger number of medical practices using a variety of EMRSs, as well as a smaller number of practices without any EMRS. The partnership intends to leverage CHSs by engaging them in the tripartite mission of educating and influencing the practice of their colleagues in the community, providing excellent care for referred patients, and participating in health services research activities driven by a central database. The database serves to identify CME topics and to guide the design, implementation, and evaluation of interventions to improve outcomes, including controlled clinical trials. The Hypertension Initiative will continue to collaborate, seek additional partners, and innovate in pursuit of the common mission. In the process, we hope to build on a template for ameliorating the projected health and economic burden of hypertension and metabolic syndrome—related risk and disease.

Disclosure: This work was supported in part by grants from the Duke Endowment; the American Society of Hypertension; P01HS1087 (EXCEED) from the Agency for Healthcare Research and Quality; 5R01HL072377 and HL04290 from the National Heart, Lung, and Blood Institute; P60‐MD00267 (EXPORT) from the National Center for Minority Health Disparities; and the Stroke Belt Elimination Initiative from the US Department of Health and Human Services. The Morehouse Community Physicians' Network is supported by grants 5P20RR11104 from the National Center for Research Resources' Research Centers in Minority Institutions program to the Morehouse Clinical Research Center, 5P20MD00272 to the Healthcare Disparities Center from the National Center for Minority Health and Health Disparities, and the Medtronic Foundation.