[Reprinted from the Journal of the Association for Academic Minority Physicians, Vol. 9, No. 3, July 1998, pp 48 - 52.]

ABSTRACT: Hypertension and diabetes are overrepresented in the African-American population and can be particularly devastating in this population. These diseases share genetic predisposition, medical risk factors, and environmental influences as etiologic factors, and they may be interrelated, at least in part, by obesity and accompanying hyperinsulinemia. Noncompliance with treatment plans is a significant barrier to health improvement in both diseases, but increased attention to patient involvement in care is a potential solution to this long-standing problem.

The Baltimore Alliance for the Prevention and Control of Hypertension and Diabetes was established in January 1998 to promote care to the underserved community of West Baltimore, Maryland, and to improve outcomes of hypertension and diabetes. Based at the University of Maryland School of Medicine, the Baltimore Alliance comprises a community health worker program, a church-based education and screening effort, managed care and pharmaceutical company (Hoechst Marion Roussel) partners, a health policy and services research group, and inpatient/ outpatient clinical care sites in the health system.

Key words: academic/industry partnership, Baltimore Alliance, community outreach, diabetes, disease management, hypertension.

In the United States, hypertension affects about 23% of the general population and 32% of African Americans¹. Diabetes (types I and 11 combined) affects about 6% of the general population but more than 11 % of African Americans². The prevalence of hypertension is two- to threefold higher in diabetic compared with nondiabetic populations³. These prevalence data suggest that 1.5 to 2.2 million African Americans in the United States have both diabetes and hypertension. Hypertension and diabetes are diseases in which compliance with treatment plans tends to be poor and out-comes are suboptimal.

Like many conditions, hypertension and diabetes increase in prevalence in people who have (1) a genetic predisposition to the disease, (2) medical risk factors, such as obesity, for the disease, and (3) suboptimal life-style factors, such as poor diet, little exercise, or substance abuse. The prevalence of hypertension is about twofold higher in a diabetic compared with a nondiabetic population. The reason for the association is unclear. In the United States, about 90% of diabetics have type II diabetes. Classically, the presence of hypertension in type I diabetes mellitus is attributed to diabetic nephropathy. Coexisting hypertension and diabetes complicate care and are probably related to a combination of genetic factors, insulin resistance (usually secondary to obesity), increased sodium sensitivity in the vasculature, and increased adrenergic responsiveness².

Insulin resistance, secondary to obesity with increased circulating insulin levels, is believed to be a modifiable risk factor for the prevention of hypertension in the diabetic patient. Hyperinsulinemia has a number of negative effects, including weight gain, enhanced atherogenesis, vascular remodeling (thickening), and increased platelet aggregation⁴. Behaviors that improve insulin action, such as increased exercise and weight loss, return elevated blood pressures to the normal range, even without sodium restriction.^{5 6 7}

One of the main problems in achieving acceptable control of hypertension and diabetes is patient compliance. Nationally, among hypertensive patients, only I of 4 or 5 patients has adequately controlled blood pressure (<140/90 mm Hg). Noncompliance is related to the fact that hypertension may progress silently. Patients tend to seek and comply with care when they have pain, are inconvenienced, or sufficiently fear a negative outcome. Asymptomatic hypertension is characterized by none of these drivers. New cases are usually discovered during annual physical examinations, as an incidental finding when a patient visits a health care professional for another reason, or during community outreach screening programs. If care is instituted and "the cure is worse than the disease" (fairly likely in the case of asymptomatic hypertension), compliance decreases.

Factors that may lead to noncompliance with therapy include onerous life-style changes, drug side effects, and failure to perceive benefit in the short term or believe that benefit will accrue in the long term. Other barriers to compliance may include cost (money and time) and priorities perceived by the patient to be higher than compliance with the plan. Obviously, these barriers must be removed before compliance can be achieved and outcomes can improve.

In diabetic patients, poor compliance also has its roots in the need to change habits established and consolidated over a lifetime. The expectation that a patient will change these habits in weeks or months, the first time around, is a laudable but unrealistic goal.

A critical success factor for goal attainment is that the goals be SMART: specific, measurable, attainable, realistic, and time locked. The clinician's goal should not be "to lower blood pressure" (or blood glucose), but rather, for example, "to lower blood pressure to 140/90 mm Hg, with minimal erosion in quality of life, within I month." The patient's goal for life-style changes should not be "to lose weight, quit smoking and drinking, and exercise more," but rather "to lose X pounds per month to a body weight of Y by instituting this diet plan and that exercise plan, then wean from cigarettes using Z approach, and finish within so many months, then..." and so on.

One other item is vitally important to help ensure patient compliance. Patients must feel empowered, motivated, involved, and, to the appropriate degree, in control of their care. Education is a good first step but does not, in and of itself, improve outcomes. That is, it is necessary but not sufficient. Patient education should be viewed as an empowering first step that provides a tool for the patient to use in controlling his or her disease.

In addition to learning about their disease, patients should be trained and motivated to keep records of symptoms, examination and laboratory data (such as blood pressure or blood glucose measurements), and an ongoing list of questions for their care provider. This involvement stimulates continual compliance and makes the encounter with the care provider a more efficient, rewarding experience.

With regard to medications, some health care providers are now advocating, and a bold few are implementing, patient-controlled dose titration (mostly but not entirely in consultation with a health care provider) in response to blood pressure or blood glucose readings performed by the patient. While this clearly is not the right Strategy for all patient personality or capability types, it does approach the ultimate in patient involvement and empowerment. When patients take ownership of their disease, learn about it, chart their progress with their exercise and diet plan, record their blood pressure or blood glucose data-when they feel in control-they comply better and longer. If a treatment plan is done to them as opposed to done with them, it is no surprise that compliance is reduced and outcomes are poor.

The Baltimore Alliance for the Prevention and Control of Hypertension and Diabetes (the Baltimore Alliance) was established in January 1998 with the stated goals of (1) increasing enrollment of patients in diagnosis and treatment programs and (2) improving retention of patients in treatment programs and, thereby, improving patients' health outcomes -- clinical, humanistic, and economic.

To achieve these goals, we coordinate, facilitate, and assist ongoing programs at the University of Maryland School of Medicine in Baltimore and initiate innovative approaches in the prevention and management of hypertension and diabetes. We recognize the importance of serving all members of the community but place special emphasis on underserved patient populations. Our working philosophy of mobilization, cultural relevance, and partnership will, we hope, make us a valued and sought-after ally of health care providers and the community in general. We work from evidence-based disease management protocols, measure the effects of our interventions, and strive to continuously improve our methods and outcomes.

The Alliance, based at the University of Maryland School of Medicine, is largely a service organization that draws on the expertise of its components to provide resources for improved patient care, satisfaction, and outcomes. The components of the Baltimore Alliance include:

ENABLE. An Americorp project, Enhancing Neighborhood Action by Local Empowerment, is a community health worker initiative modeled on the "Barefoot Doctor" program in China. It serves high-risk patients in the West Baltimore community who have diabetes and/or hypertension (among other disease states) to help drive compliance with both treatment regimens and appropriate contact with health professionals. Trained volunteers visit with patients in their community to help, remind, and inform them.

CHAMP. The Community Health and Monitoring Program is a community-based health education and prevention program that provides screening, referrals, monitoring, and culturally relevant health education to the African-American population through church and other volunteer efforts.

Managed care organizations (MCOs). MCOs that share the values and goals of the Alliance are welcomed as partners in the effort to improve overall out-comes. As we will discuss later, we believe that all outcomes are candidates for continual improvement. Clinical outcomes are most important to us, because we believe that improved clinical outcomes, achieved properly, can help improve humanistic (quality of life and satisfaction with care encounters) and economic outcomes, in both the short and the long term.

Clinical experts. The Alliance is fortunate to be able to work with, draw on the expertise of, and provide synergies to the Division of Hypertension and the Joslin Center for Diabetes at the University of Maryland School of Medicine. In addition to providing excellent care and having their own well-established programs, these centers help lead the disease management design teams and are valuable resources for provider and patient education.

Clinical care sites. The University of Maryland School of Medicine is composed of a large tertiary care hospital, special care sites, and five comrnunity-based care sites. The community care sites (UniversityCare) are distributed throughout West Baltimore and are both outreach and primary care sites. One of these sites will act as the nerve center and central data clearinghouse for the Alliance's efforts and projects related to the clinical care of outpatients in the system and the disease management efforts. A special site, University of Maryland Family Medicine, is a freestanding site on the campus that provides care, expertise, and logistical support for the hypertension and diabetes efforts of the Alliance.

Center for Health PolicylHealth Services Research. This component of the Alliance provides experts in applying research methods to assess and predict the utilization, quality, and outcome of services offered throughout the health care system. In addition to epidemiologic and biostatistical and computer-assisted telephone interviewing services, the Center has experts in existing and developing health policy matters.

Other components. Partnership affiliations also exist with the following organizations:

• Hoechst Marion Roussel Pharmaceuticals provides significant funding and novel personnel support. Hoechst's in-kind contribution of the Alliance's executive director both preserves the budget for the work of the Alliance and provides needed expertise in project management, disease management, and other experience and skills. Local Hoechst personnel, in addition to their classic job role, help drive the goals of the Alliance, increasing knowledge and visibility of the Alliance and its goals. They provide additional support, such as provider and patient education, and budget support from local funds. During the disease management implementation phase, these personnel help sell the concept and support its adoption.

• Legacy 2000, a Baltimore Times newspaper effort, is a men's health initiative that goes beyond hypertension and diabetes to address issues impacting the health of African-American men in the Baltimore area.

• Millenium Healthcare is a Baltimore health education and services organization with goals similar to those of the Baltimore Alliance. Leadership in both organizations saw the sense of combining and cooperating in the effort to achieve synergies despite limited resources.

As might be expected, the Baltimore Alliance is organized with a bit of a twist on how we view things. Using the "inverted pyramid" concept, we see the small apical base as our Advisory Board of Directors. This board is composed of health system and community leaders as well as community members and patients. It performs the usual board functions. On the next level, a Steering Team composed of representatives of the components meets about every other week to "do the work" of the Alliance. This team pushes programs and ideas out to community leaders on the next level of the pyramid and cooperates with them to serve the huge "base" of the pyramid, the community members of West Baltimore.

In addition to the risk-factor screening and clinical services mentioned previously, the Alliance offers provider education programs that range from physician continuing medical education and nursing and pharmacy continuing education programs to disease management training and quality updates. Quality updates will be accomplished through the "provider report card" approach described elsewhere and currently implemented in many health systems. Patient education and training will be provided in standard and some not-yet-standard ways. We previously described our belief in the importance of effective patient education as a cornerstone of compliance. In addition to traditional one-on-one and classroom-style training, we will have a newsletter service to guide, inform, and motivate readers. We have already established a web site (http://som1.umaryland.edu/BaltAlliance) to facilitate the updating and development of information during the growth phase of the project. The site informs the reader about the Baltimore Alliance, hypertension, and diabetes and lists information about University-Care sites and accessing care in the system. It also provides easy links to related sites of potential interest.

Disease management (DM) will be used as a method of standardizing care and measuring the beneficial effects of the Alliance's interventions. The executive director of the Alliance trained and worked in disease management (and continues to work in that project as a minor time commitment) at the University of Pennsylvania Health System. We subscribe to two definitions of disease management. Each definition approaches the topic from a slightly different, yet equally important perspective, and both have certain core elements in common:

Subscribing to both definitions ensures that we place equal emphasis on the process of designing and implementing evidence-based guidelines as well as the outcomes they are intended to achieve. We believe the DM approach will allow us to continue to put patients first, meet the needs of all partners in the health care effort, measure the effects of our interventions, and continuously improve our outcomes. We do not view DM as just drug therapy management or out-comes assessment, but rather as composed of all of the following components:

• Prevention programs
• Identification of at-risk patients (those as yet undiagnosed)
• Establishment of current baseline practices and outcomes
• Diagnostic guideline construction
• Severity stratification of patient population
• Comprehensive therapeutic guideline construction
  • Life-style interventions
  • 0ther nondrug therapy
  • Drug therapy
• Referral guideline construction
• Follow-up guideline construction
• Health care provider prompts, reminders, and education
• Patient education efforts and materials
• Comprehensive outcomes assessment
  • Clinical outcomes
  • Overall humanistic outcomes:
    • General quality of life
    • Disease-specific quality of life
    • Patient satisfaction with episode-of-care encounter or wellness intervention
• Overall informatic needs
  • Covered lives demographics
  • Intranet provision of provider prompts and guidelines
  • Intranet provision of patient education and motivation material
  • Outcomes data collection
• Continuous improvement efforts and methods
• Patient-centered goals
• Program-centered goals

Attention to all these components by the design teams and during the implementation stage of the project will help ensure a complete and well-rounded approach to the problem. Extensive provider involvement in the design stage will help ensure buy-in and create a core of champions for the system as it is rolled out to the clinical care sites. Further, such early involvement will guarantee that the system meets the real-world needs of care providers by remaining easy to use and improving care. In this way, the disease management system is valued and appreciated by the end users (the care providers) rather than viewed as "more annoying paperwork" that gets between the patient and the care provider.

After identifying our current baseline practices and their outcomes, we will institute the hypertension and diabetes systems uniformly across the health system. Outcomes will be assessed at 6 and 12 months and then annually. Providers will be encouraged and given incentives to comply with the disease management system with minimal variation. When the care provider believes there is a need to deviate from the DM plan for an individual patient, the reasons for those deviations will be captured and used to improve the program. The DM program will therefore continually evolve and improve. New advances in hypertension and diabetes care will easily and efficiently be slotted into the program, as will new approaches to patient education and outcomes assessment.

While these are perhaps the main strengths of a disease management approach, others include the following: (1) improvement in care quality and patient outcomes (2) management of or reduction in overall health costs (3) improvement of both health and satisfaction (in the case of the latter, for patients and health care providers), and (4) in the increasingly consumer-oriented environment in which health care systems find themselves, disease management positions the system to attract risk contracts and maintain and/or expand its "business."

It has been said that "the devil" of disease management programs is not in the details of design, but rather in the details of implementation. To be embraced by health care providers, DM programs must be simple, easy to use, and improve care. Perhaps most important, they must not impose an undue increase in work load (especially paperwork) on already busy health care providers. Keeping these factors in mind and instituting DM programs in an organized, timely progression with sufficient training and support will help ensure their success.

The Baltimore Alliance looks forward to providing those resources, removing obstacles, and rewarding success. It also looks forward to reporting outcomes improvements and goal attainment in the future.

The authors gratefully acknowledge the assistance of Lewis H. Roht, MD, NTH, and the Department of Epidemiology, Division of Medical Affairs, at Hoechst Marion Roussel Pharmaceuticals, Inc.

'Hoechst Marion Roussel Pharmaceuticals, Inc, Kansas City, Missouri, and the Baltimore Alliance for the Prevention and Control of Hypertension and Diabetes, Baltimore, Maryland.

'Department of Family Medicine, University of Maryland School of Medicine, and the Baltimore Alliance for the Prevention and Control of Hypertension and Diabetes, Baltimore, Maryland.

Address correspondence and reprint requests to Joseph C. Gerber, RPh, PhD. Baltimore Alliance, University of Maryland, 520 W Lombard St (rear), Gray Lab-201-D, Baltimore, MD 21201.

Supported as a Patients First Partnership by the University of Maryland School of Medicine and Hoechst Marion Roussel Pharmaceuticals, Inc (HMR). HMR provided funding as well as personnel and other in-kind support.

REFERENCES

¹ Burt VL, Wheiton P, Rocelia EJ, Brown C, et al. Prevalence of hypertension in the US adult population: Results from the Third National Health and Nutrition Survey, 1988-199 1. Hypertension 1995;25:305-313.

² US Dept of Health and Human Services. CDC National Fact Sheet for Diabetes. Centers for Disease Control and Prevention, Nov 1, 1997.

³ Arauz-Pacheco C, Raskin P. Diabetes m,llitus perspectives on therapy: Management of hypertension in diabetes. Endocrinol Metab Clin North Am 1992;21:371-379.

⁴ Bakris GL. Pathogenesis of hypertension in diabetes. Diabetes Reviews 1995;3:460- 476.

⁵ Tuck MI., Sowers J. Domfeld L, Kledzig G, et al. The effect of weight reduction on blood pressure, plasma renin activity, and plasma aidosterone levels in obese patients. N Engl J Med 1981:30,4:930-933.

⁶ Sonne-Holme S, Sorenson TIA, Jensen G, Schnohr P. Independent cffects of weight change and attained weight on the Prevalence of arterial hypertension in obese and non-obese men. BMJ 1980;299:767-770.

⁷ Reisin E, Able R, Modan M, Silverberg DS, et al. Effect of weight loss without salt restriction on the reduction ofblood pressure in overweight hypertensive patients. N Engi J Med 1978;298:1.6.