hypertension and medical informatics - NCBI

HYPERTENSION AND MEDICAL INFORMATICS William McBride, MD; Carlos Ferrario, MD; and Paulefte A. Lyle Whitehouse Station, New Jersey and Winston-Salem, North Carolina

Background: Only about 27% of Americans with hypertension have their disease under control. Hypertension in the African-American population has a higher prevalence (32%) and is less likely to be treated or controlled compared with that in the caucasian population. Hypertension places a significant burden on patients and health care systems. Applications of medical informatics can facilitate the management of hypertension. Examples that illustrate the utility, status, and future potential of medical informatics applications in the treatment of hypertension are presented. Methods: Relevant studies and review articles were accessed through a PubMed search of the English-language literature and for current Internet-based information for the time period of 1993-2002. Search terms included, but were not limited to, hypertension, medical informatics, medical information science, electronic medical records, Internet, and managed care. Results: There is evidence that medical informatics has a favorable impact on health care issues as it relates to patients and physicians. Although the use of computers to assist in managing hypertension is in its infancy, there are examples where informatics applications have a demonstrated clinical value. Conclusions: Management of hypertension needs much improvement. The use of computers and the Internet in health care is expanding and expected to have a positive impact on the management of chronic diseases, such as hypertension. (J Natl Med Assoc. 2003;95:1048-1056.)

Key words: hypertension * medical informatics * medical information science * electronic medical records

BACKGROUND Phase II of the Third National Health and Nutrition Examination Survey (NHANES III) showed that only about 53% of people with hypertension are receiving treatment, and only 27% of patients with hypertension have their disease under control with medication. Outcomes are even worse in the African-American population, which has a higher prevalence of hypertension (36.7% of AfricanAmerican men and 36.6% of African-American C 2003. From Merck & Co. Inc., Whitehouse Station, NJ (McBride, Lyle); Hypertension and Vascular Disease Center, Wake Forest University, Winston-Salem, NC (Ferrario). Send correspondence and reprint requests for J Natl Med Assoc. 2003;95:1048-1056 to: William McBride, MD, Merck & Co. Inc., PO Box 4, WP 39-242, West Point, PA 19486-0004; phone: (215) 652-7585; fax: (215) 993-3359; e-mail: [email protected]

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

women versus 25.2% of caucasian men and 20.5% of caucasian women) that is less likely to be treated or controlled than in the caucasian population." 2 For all races, hypertension commonly occurs with other cardiovascular risk factors, such as obesity, diabetes, and hyperlipidemia. End-organ damage in the cardiac, cerebrovascular, and renal systems is more common and occurs earlier in African Americans.34 Lack of control of hypertension, coupled with poor management of other risk factors, is responsible for continued excess mortality and morbidity from heart failure, myocardial infarction, and cerebrovascular disease, particularly in the southeastern United States.56 About half of first myocardial infarctions and two-thirds of first strokes occur in patients with blood pressure >160/95 mmlHg7. Hypertension is associated with two to three times higher risk for the development of heart failure.7 These poor clinical outcomes are burdensome to patients and very costly to the health care system. Medical informatics (the scientific field that VOL. 95, NO. 11, NOVEMBER 2003 1048

HYPERTENSION AND MEDICAL INFORMATICS

deals with the storage, retrieval, sharing, and optimal use of biomedical information, data, and knowledge for problem-solving and decision-making)8 can facilitate the planning of effective, population-based treatment approaches for patients with hypertension. Computer applications could be part of a multidisciplinary approach that could slow the growing epidemic of hypertension-related diseases. For example, patient characteristics, physician characteristics, treatment patterns within and among practices, and gaps in care can be identified using medical informatics applications. Feedback about practice patterns can be provided to health care professionals to help improve treatment of hypertension. Furthermore, data can be collected from clinical practice settings, in contrast with restrictive clinical trials in which treatment decisions are directed by a treatment protocol. Computerized collection of data from clinical practice can help determine how physicians are incorporating knowledge about the treatment of disease (for example, evidence-based hypertension treatment guidelines) in their practices.

METHODS This review: 1) discusses the problem of hypertension management and its impact on the health care market, 2) analyzes agencies and projects that use medical informatics to assess patterns of prescription drugs used to treat patients with hypertension, and 3) describes the expanding role of medical informatics for both providers and patients. Relevant studies and review articles were accessed through a PubMed search of the Englishlanguage literature and for current Internet-based information for the time period 1993-2002. Search terms included, but were not limited to, hypertension, medical informatics, medical information science, electronic medical records, Internet, and managed care. Examples from the literature and the Internet that illustrate the utility, status, and future potential of medical informatics applications in the treatment of hypertension are presented. Quantitative selection criteria were not applied.

RESULTS AND DISCUSSION Control of Hypertension in the Managed-Care Era Many cases of hypertension go unrecognized or, if diagnosed, are poorly treated.2 The reasons for this situation are complex. Because hypertension is 1049 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

a chronic disease that is asymptomatic during most of its course, many patients are complacent about their condition. Access to care is another problem for the increasing number of uninsured patients in the United States. Uncoordinated care is common for many chronic diseases that are managed on an outpatient basis, which often results in interruptions between diagnosis and treatment, and during treatment. However, even those patients with access to health care often have uncontrolled hypertension.2 Poor control rates of plasma cholesterol levels in patients with coronary heart disease and low prescriptions rates for angiotensin-converting enzyme (ACE) inhibitors in patients with heart failure are two well-documented examples of shortcomings in the outpatient management of chronic, cardiovascular conditions.9 There are other financial and administrative causes for diagnosis and treatment interruptions in medical practice in the United States. Reimbursement for medical services is often better for acute care and interventional procedures than for outpatient management. Furthermore, health care systems-especially in academic centers-frequently lack centralized patient information systems and, therefore, tracking the management of common chronic conditions can be challenging. Finally, because of the demands on primary care physicians to see more patients, outpatient visit times are short. This results in insufficient time for physician-patient interaction, including that which should be dedicated to educating patients about their disease. Over the past decade, managed-care programs have provided health insurance for millions of Americans. An expectation of managed care was improved health outcome with concurrent control of spiraling health care costs. The population-based principles of managed care theoretically could be the basis of improved management of hypertension. Jacobs'° linked the relationships between hypertension and managed-care principles: 1) hypertension is a common disease that provides the opportunity for population-based management; 2) the risk stratification for hypertension is well defined-therefore, resources can be allocated to patients according to risk strata; 3) guidelines for the treatment of hypertension exist that can serve as a foundation for disease management; and 4) hypertension is managed largely in the outpatient setting, which is emphasized by managed care. The efficacy of blood VOL. 95, NO. 11, NOVEMBER 2003


pressure control rates was introduced in 2000 as a part of the Health Plan Employer Data Information Set (HEDIS) measure. Successful HEDIS measures are required for health plans to gain accreditation by the National Committee on Quality Assurance. Although managed-care systems were introduced to enhance control of chronic diseases, data supporting the view that managed-care plans have improved hypertension management remain elusive. The reasons are multifactorial. Managed-care organizations disseminate guidelines, although the degree of adherence to such guidelines by participating physicians is variable. Managed-care organizations often have adversarial relationships with providers and patients. Finally, it is difficult for a managed-care plan to manage patients with a chronic disease, such as hypertension, because many patients migrate from one plan to another. Several issues challenging the optimal treatment of patients, including those with hypertension, are related to the cost-containment features of managedcare plans. Reimbursement schemes, such as capitation, may not have been successful because some physicians react to them by limiting comprehensive care to patients in an effort to control their internal costs. A physician may attempt to control costs by prescribing drugs that may be less expensive but not as well tolerated as alternative drugs. The side effects of the less-expensive drug may result in poor patient compliance. In the Medical Outcomes Study, Greenfield et al.'I asserted that capitation did not influence medical outcomes of patients with hypertension or diabetes compared with fee-for-service care. However, this was an observational study rather than a randomized clinical trial, and the patients were followed only for seven years-a relatively short period for this long-term condition. In addition, the sample size was small, and the results may not be generalizable because the study took place in only three metropolitan areas. Another issue related to reimbursement that has yet to be adequately addressed is the general lack of financial incentives for provider quality performance. Finally, an inadequate information management infrastructure in many managed-care organizations makes it difficult to identify and stratify patients, understand provider practice patterns, and implement early decision support. A recent report demonstrated that a decision support system improved guideline adherence and reduced costs'2outcomes attractive to managed-care organizations. JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

As insurance companies and providers invest in computers and medical informatics, there is the expectation that the management of hypertension and other chronic medical problems may improve. In summary, many patients with hypertension are receiving their medical care through managedcare systems. This vehicle of medical care provides positive opportunities for the effective management of hypertension, as well as challenges that are important to address, perhaps via medical informatics tools, both for the benefit ofpatients and the efficiency of health care administration.

MEDICAL INFORMATICS IN THE MANAGEMENT OF HYPERTENSION Population-Based Projects Population-based databases provide researchers with the opportunity to assess clinical practice patient and practice characteristics, which can be contrasted with the necessarily restrictive characteristics of clinical trial settings. Population-based approaches to hypertension utilize existing administrative databases to evaluate practice patterns or follow well-defined methodology and collect primary data from a sample of patients to create a database. The latter methodology, while more detailed and accurate than is an administrative database, is more time-consuming and costly. In a study of 15 Boston hospitals, it was reported that a review of discharge abstracts identified only one-half the number of comorbidities compared with the number identified from full medical record review."3 Despite the inherent weaknesses, which should decrease over time with systems and processes enhancements, administrative data have been used to investigate health service questions in large patient populations. Performing chart abstraction on a sample of the study population can validate administrative selection criteria. Three examples of population-based studies of hypertension follow that used medical informatics methodology to determine frequency distribution of medication use, discontinuation rates of medications, and economic factors associated with hypertension drug management.

Hyperlension Management in a Large Health-Maintenance Organization Administrative data were used to evaluate the effects of the 1992 guidelines of the Joint National Committee on Prevention, Detection, Evaluation, VOL. 95, NO. 1 1, NOVEMBER 2003 1050


and Treatment of High Blood Pressure'4 (INC V) on drug treatment patterns of hypertension in a nonstaff model health-maintenance organization (HMO).'" Patients with a new ICD-9 hypertension diagnosis (preceded by at least a 12-month interval without a diagnosis of hypertension) that was corroborated by hypertensive drug use were included in the survey. The selection process was validated by reviewing a random selection of 174 charts. Using strict criteria for the diagnosis of hypertension, 154 of the 174 audited patient charts confirmed the diagnosis of hypertension, for a positive predictive value of 90.2%. Although JNC V guidelines emphasized betablockers and diuretics as first-line agents in patients with uncomplicated hypertension or nonmenopausal women, ACE inhibitors and calcium channel blockers were frequently used for both new starts and add-on treatment in the study population. In addition, the study results indicated high rates of therapy discontinuation for diuretics, which were recommended as second-line treatment by JNC V (40.6%) compared with beta-blockers (28.2%), calcium channel blockers (27.2%), and ACE inhibitors (25.3%). Strict compliance with treatment guidelines may not be appropriate for certain patients; however, the results of this survey suggested unawareness of or lack of consideration for treatment guidelines for treating hypertension.

Manitoba Drug Program Information Network Study Patterns of prescription drug use during 1996 in Manitoba, Canada, were studied.'6 Data were compiled from 283 community pharmacies using the province health department's Drug Program Information Network (DPIN). This information system identifies all claims for prescription drugs dispensed throughout Manitoba. Approximately 6.7 million prescriptions were dispensed in 1996. The authors of the study tabulated the types of drugs dispensed, mean number of prescriptions per resident, drug use variations according to age and gender characteristics, and costs per patient, as well as overall cost to the region's various reimbursement programs. The three most commonly dispensed drugs were used to relieve nervous system disorders (such as anxiety or depression) and pain (1,274 prescriptions per 1,000 residents), to treat cardiovascular disorders (1,122 prescriptions per 1,000 residents), and to control infections (863 1051 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

prescriptions per 1,000 residents). The DPIN enabled the province's health care planners to assess whether physicians and health centers in Manitoba were making optimal use of cost-effective disease management approaches. The high quality of the pharmaceutical data permitted the Manitoba health authorities to identify communities with healthy and unhealthy subpopulations, because the latter group tended to use a higher volume of prescription drugs. For example, the Manitoba health researchers found that the characteristics of ACE inhibitor use paralleled that of other drugs evaluated; that is, in areas of the province where health is generally poorer, more ACE inhibitor prescriptions were dispensed and subjects were of lower socioeconomic status. Further research will be necessary to determine compliance rates for ACE inhibitors, and how patients receiving these drugs use other areas of the health system, such as hospitalizations. ACE inhibitors could be appropriate drugs for lower socioeconomic groups if the use of more expensive services, such as hospitals, was decreased due to better hypertension control with these agents.

Hypertension Management in the California Medicaid (Medi-Cal) Program McCombs et al.'7 studied Medi-Cal claims data to evaluate the economic effect of interrupting or terminating antihypertensive medication. Paid claims data from 400,000 Medi-Cal recipients were analyzed. Patients over age 40 with at least one antihypertensive prescription were identified, and all additional paid claims from each patient over a 12-month period were analyzed. Using a strict definition of interruption or termination of therapy, a final analytic data set of 6,419 hypertensive patients was identified for detailed clinical and

economic analysis. The results of the study showed that: 1) 86% of newly diagnosed patients with hypertension interrupted therapy over the 12-month period, 2) patients with interrupted therapy consumed an additional $873 per patient in health care resources, and 3) increased costs were primarily due to increased hospital expenditures. The highest rate of continuation was seen with the ACE inhibitors (33%); other rates of continuation were calcium channel blockers (28%), beta-blockers (17%), central-acting drugs (17%), and diuretics (5%). VOL. 95, NO. 1 1, NOVEMBER 2003


Summary The three examples of population-based studies of hypertension used computer-generated databases to determine frequency or distribution of medication use, discontinuation rates of medications, and economic factors associated with drug management of hypertension. Population-based data may assist health care professionals in formulating disease interventions that might improve clinical and economic outcomes associated with hypertension. As we continue to expand our use of computers and the Internet in medical care, it is likely that interventions will be based on information obtained from new ever-changing technology. Challenges to the availability and utility of population-based data are significant. The challenges include equipment costs, programming, training, obtaining complete data, and the time required for analysis and review.

THE EXPANDING ROLE OF COMPUTERS IN THE MANAGEMENT OF HYPERTENSION The use of computers in clinical medicine is expanding. Computers can assist in case finding, recall, risk stratification, guideline implementation, and obtaining aggregate data about practice patterns. Computers will play an increasingly important role in the detection and treatment of hypertension, and the follow-up of such patients. Appropriate security associated with patient privacy is very important for all computerized patient data. A review of the use of computers in hypertension indicated that: 1) there are few randomized clinical trials regarding the use of computers in medicine; 2) computers have the potential to improve the detection, treatment, and follow-up of hypertension in community practices; and 3) available randomized clinical trial data indicate an improvement in patient administration (e.g., recording blood pressure data, occurrence of patient follow-up visits). However, the review was inconclusive regarding physician performance or blood pressure control.'8 Further studies will be needed to evaluate the effectiveness of computers in quality performance programs related to hypertension. There is an increased demand for quality in the medical market place, and computers play an important role in the measurement and enhancement of quality. Hypertension recently became a HEDIS measure, which has stimulated an increase in the JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

number of hypertension disease management programs that involve practicing physicians with managed-care contracts, from 2-3% in 1996 to 1997, to 10-13% in 1998.9 Quality performance programs have become a requirement of resident training education'9, which should result in an increasing number of quality improvement projects in residency programs in the United States. As these quality improvement programs expand, there will be more of an opportunity to use computers and determine if health and economic outcomes improve. In this section, we provide examples of the use of electronic medical records (EMRs), the Internet, and other computer-based systems that support the treatment of hypertension.

Role of EMRs in the Management of Hypertension An EMR-the digital creation and storage of the traditional paper-based medical record-can be designed to collect comprehensive demographic, clinical, treatment, and economic information. There are numerous potential benefits of EMR use: individual or simultaneous access from multiple locations, a variety of views of data, capabilities of structured data entry and aggregate data analysis, decision support, electronic data exchange, and legibility.20 An extension of the use of EMR could be the inclusion of a decision support system. For example, the system known as HyperCritic2' was designed and tested by general practitioners with regard to the treatment of patients with hypertension: medication type, dose, concomitant therapy, blood pressure level, potential adverse reactions, and laboratory parameters. Overall, the system-generated treatment critiques that were provided to the physician users were considered to be as beneficial as those provided by physician reviewers; however, there was variability among the physician reviewers with regard to their acceptance of the critiques. Anonymous primary care EMR data in the Practice Partner Research Network (PPRNet) was analyzed to evaluate adherence to process and outcome measures for chronic illness, including hypertension.22 This study showed that treatment guidelines were not being followed, and that most patients did not reach desired clinical outcomes. In the case of hypertension, 78.5% of patients had their blood pressure measured at their first clinic visit, and only 45% of patients with systolic blood VOL. 95, NO. 1 1, NOVEMBER 2003 1052


pressure >140 mmHg or diastolic blood pressure >90 mmHg for more than three measurements were diagnosed as having hypertension. Increased compliance with guidelines was commonly observed in practices using EMRs compared with those that did not. Physician feedback regarding the use of EMRs has been mixed. Reported benefits of EMRs are improved patient management; communication between physicians, staff, and patients; and job performance.23 In the report by Wager et al.,23 there was a perception of reduced practice expenses but no documentation of actual expense reduction. As EMR technology progresses to increase physician workflow and save time, and as benefits to disease management are demonstrated, adoption of EMRs should accelerate.2425

The Internet and Health Care for Patients with Hypertension A recent survey of 10,000 physicians by the Healtheon Corporation reported that Internet e-mail was used by 85% ofphysicians, with 63% of the physicians using e-mail on a daily basis.26 Another survey done by PSL Consulting and the American Medical Association found that 75% of physicians in the United States, Canada, and the United Kingdom used the Internet. The survey found that 95% of the physicians access electronic medical journals; whereas 59-86% accessed professional association web sites.26 There are no reliable or consistent data on the number of physicians who use e-mail regularly for clinical encounters. Internet tools can be difficult to create, identify, or implement; the Healtheon Corporation has been trying to "web enable" major national insurance companies to create infrastructure for shared webbased systems since 1997 and is still far away from achieving its goal.27 Computer-assisted physician education began in the 1960s. There are few randomized clinical trials comparing computer-assisted instruction with conventional self-study methods. A recent randomized clinical trial comparing a web-based tutorial system with self-study printed materials for the clinical management of patients with acute myocardial infarction found that online tutorials produced greater learning efficiency (Internet groups needed less time for studying) and were more satisfying than conventional self-study of printed materials.28 The Hypertension, Dialysis, and Clinical 1053 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

Nephrology web site29 is an example of an Internet site devoted to the dissemination of information to medical professionals. The site features a questionand-answer section; summaries of articles, abstracts, and meeting presentations; and links to review articles. Links to other relevant organizations, medical provider networks, and databases are also provided. Although education about hypertension management is provided to physicians in several Internet sites, evaluation of web-based learning efficiency, retention, and satisfaction related to hypertension has not been reported. Patient use of the Internet has fueled a rising tide of consumerism related to health.30 Patients seek information from the Internet for their physical complaints, established diseases, treatment, and lifestyle modification issues. Herzlinger reported that there were 30 million people who used the Internet for health information in 1999 and estimated that the number will reach 70 million in 2000.31 Interestingly, Herzlinger estimated that 35% of Medline use is from nonphysicians. The Internet might lead to patients demanding access to their physicians by e-mail 24 hours per day and expecting efficient practice administration and timely electronic exchanges with their physician's office. In this application of computers for medicine, appropriate security for patient data will be particularly challenging. Patients acquire information and draw support from people around the world through online chat rooms and disease-focused interest groups. The number of consumer-oriented list groups is estimated to be 90,000 (patients receive e-mails regarding their disease or medical interest), and the number of medical-related news groups for consumers is reported to be 30,000.27

Other Computer-Based Tools Related to Hypertension Epstein and Sherwood33 have defined disease management as an explicit systemic populationbased approach to identify persons at risk, to intervene with specific programs of care, and to measure clinical and other outcomes. Electronic tools that are designed to improve hypertension control rates have some of the characteristics that fit this definition of disease management. In this section, we briefly discuss telemedicine, telepharmacy, "remote doctoring," and computer-based decision support systems. VOL. 95, NO. 11, NOVEMBER 2003


Telemedicine has been defined as the use of telecommunications to provide medical information and services.34 Currently, teleradiology is the most common application of telemanagement. Telemedicine is being initiated in correctional facilities and home health care. Telesurgery using remote-control robots is being investigated by the military and others.35 Friedman et al.36 found that the weekly use of an automated telephone monitoring system over a six-month period improved medication adherence rates, blood pressure control, and patient satisfaction in hypertension subjects. The telephone monitoring system used in that study cost about $65 per year. A major barrier to telemedicine is the lack of reimbursement for most of these types of services. With advanced technology and the Internet, telepharmacy (electronic prescribing) has been made possible.37 Telepharmacy offers physicians tools to prescribe appropriate medication more appropriately and more safely. Telepharmacy facilitates compliance with formulary requirements. Features of telepharmacy include alerts to promote safety (e.g., potential drug interactions), lists of frequently prescribed drugs that can be reflective of provider and payer preferences, lists of drugs associated with treatment guidelines, automatic exchange recommendations if a prescribed drug is nonformulary, and cost information. Implementation of a telepharmacy program begins with entering the prescription into the system using a personal digital assistant or other data-entry vehicle. The prescription is transmitted over the Internet (protected by encryption) to the appropriate pharmacy. Bates25 suggests that telepharmacy improves drug safety and concludes that the telepharmacy system of the future will include robots used to fill prescriptions (complex prescriptions will be filled by a pharmacist), bar coding, automated dispensing devices, and computerization of the medication management system. MedImpact, a Pharmacy Benefit Management company, successfully implemented electronic prescribing.37 Prescription process time decreased by 30%, physician acceptance reached 35% within six months, electronically transmitted prescriptions were 15% less costly, and there was a high degree of satisfaction from providers and staff. A specific study evaluating the use of a telepharmacy management system has not been reported for hypertension; however, we should assume that antihypertenJOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

sive drugs are included in studies regarding telepharmacy because it is a common disease and blood pressure medications are frequently prescribed. The Internet has enabled a variety of "remote doctoring" models to appear in the market place.27 One category of the model is "e-visits," where physicians formally respond to patients' e-mails. Another category is the disease management model that sends patients' clinical measurements from their home computers or medical devices to centralized nursing stations. Kassirer32 predicted that such visits would replace regular personal visits; however, physicians may not embrace these forms of technology because currently there is no reimbursement for these types of services. Other obstacles relate to malpractice and the intrusion into a physician's private life by work-related e-mail traffic. Computer-based decision support systems should be helpful in physician education and decision-making for individual patients. The ATHENA Decision Support System promotes guideline-concordant choice of antihypertensive drug therapy in relation to comorbid diseases.38 It can be integrated into several EMR systems. Similarly, the PRODIGY system is a guideline-based, decisionsupport system designed to assist general practitioners in England in choosing the appropriate therapeutic action for their patients with hypertension or asthma.39

CONCLUSIONS Disappointingly low numbers of patients with properly diagnosed and/or controlled hypertension demonstrate that management of hypertension is difficult and needs much improvement. Currently, there is no evidence that hypertension control rates have markedly improved in this era of managed care. The use of computers in clinical practice is expanding and expected to have a positive impact on the management of chronic diseases, such as hypertension. In the future, there may be increased use of electronic medical records with universal links to pharmacy, laboratory, and administrative databases. Pharmacy management systems and decision support could be available at the point of care. The role of the Internet is expected to be instrumental in the electronic health care revolution. As medical informatics tools become more widely available and used, the success rates associated with the management of hypertension are expected to improve. VOL. 95, NO. 11, NOVEMBER 2003 1054


REFERENCES 1. Nieto FJ, Alonso J, Chambless LE, et al. Population awareness and control of hypertension and hypercholesterolemia. The Atherosclerosis Risk in Communities Study. Arch Intern Med. 1995; 155:677-684. 2. Hyman DJ, Pavlik VN. Characteristics of patients with uncontrolled hypertension in the United States. NEJM. 2001;345: 479-486. 3. Francis CK. Hypertension, cardiac disease, and compliance in minority patients. Am J Med. 199 1;(Suppl I A):29-36. 4. Flack JM, Mensah GA, Ferrario CM. Using angiotensin converting enzyme inhibitors in African-American hypertensives: a new approach to treating hypertension and preventing targetorgan damage. Ciur, Med Res Opin. 2000; 16:66-79. 5. Jones D, Basile J, Cushman W, et al. Managing hypertension in the southeastern United States: applying the guidelines from the Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI). Am JMed Sci. 1999;318:357-364. 6. Ferrario C, Yunis C, Mendys P, et al. Control of hypertension and blood pressure in the southeast region of the United States: a call for action. Cardiovasc Rev Rep. 1999;20:379-386. 7. American Heart Association. Heart Disease and Stroke Statistics 2003 Update. Dallas, TX: American Heart Association; 2002. 8. Greenes RA, Shortliffe EH. Medical informatics. An emerging academic discipline and institutional priority. JAMA. 1990;263: 114-120. 9. Sueta CA, Chowdhury M, Boccuzzi SJ, et al. Analysis of the degree of undertreatment of hyperlipidemia and congestive heart failure secondary to coronary heart disease. Am J Cardiol. 1999;83: 1303-1307. 10. Jacobs RP. Hypertension and managed care. Am J Managed Care. 1998;4:S749-756. 11. Greenfield S, Rogers W, Mangotich M, et al. Outcomes of patients with hypertension and noninsulin-dependent diabetes mellitus treated by different systems and specialties. Results from the medical outcomes study. JAMA. 1995;274:1436-1444. 12. Persson M, Mjorndal B, Carlberg B, et al. Evaluation of a computer-based decision support system for treatment of hypertension with drugs: retrospective, nonintervention testing of cost and guideline adherence. JIntern Med. 2000;247:87-93. 13. Schwartz M, lezzoni LI, Moskowitz MA, et al. The importance of comorbidities in explaining differences in patient costs. Meclical Care. 1996;34:767-782. 14. The fifth report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC V). Arch Inter-n Med. 1993; 153:149-152. 15. Hanchak NA, Boccuzzi SJ, MacPherson S, et al. Recent trends in antihypertensive therapy. J Am Coll Cardiol. 1996;27

(Suppl A):356A. 16. Metge C, Black C, Peterson S, et al. The population's use of pharmaceuticals. Medical Care. 1 999;37:JS42-JS59. 17. McCombs JS, Nichol MB, Newman CM, et al. The costs of interrupting antihypertensive therapy in a Medicaid population. Meclical Care. 1994;32:214-226. 18. Montgomery AA, Fahey T. A systemic review of the use of

1055 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

computers in the management of hypertension. J Epidemiol Conimmnity Health. 1998;52:520-525. 19. Accreditation Council for Graduate Medical Education. Institutional organization and commitment: quality assurance. Accreditation Council for Graduate Medical Education. September (1998): 1-5. 20. Van Bemmel JH, Musen MA. In: Handbook of Medical Informatics. 1st ed. The Netherlands: Springer-Verlag. 2000;99-115. 21. van der Lei J, van der Does E, Man in 't Veld AJ, et al. Response of general practitioners to computer-generated critiques of hypertension therapy. Meth Informn Med. 1993;32: 146-153. 22. Ornstein SM, Jenkins RG. Quality of care for chronic illness in primary care: opportunity for improvement in process and outcome measures. Am J Managed Care. 1999;5:621-627. 23. Wager KA, Lee FW, White AW, et al. Impact of an electronic medical record system on community-based primary care practices. JAm Board Fanmili, Practice. 2000;13:333-348. 24. Drezner JL. Understanding adoption of new technologies by physicians [Medscape General Medicine web site]. Available at: http://www.medscape.com/. Accessed February 7, 2000. 25. Bates DW Using information technology to reduce rates of medication errors in hospital. Br Med J 2000;320:788-791. 26. Kalsman MW, Acosta DA. Use of the Internet as a medical resource by rural physicians. J An Board Fain Pract. 2000; 13: 349-352. 27. Kleinke JD. Failed promise of the health care Internet. Health Affairs. 2000; 19:57-71. 28. Bell DS, Fonarow GC, Hays RD, et al. Self-study from web-based and printed guideline materials: a randomized, controlled trial among resident physicians. Ann Intern Med. 2000; 132:938-946. 29. Hypertension, Dialysis, and Clinical Nephrology web site. Available at: http://www.hdcn.com. Accessed January 4, 2002. 30. Wiebe D. Doctors swept up by connected consumers [Medscape web site]. Available at http://wvw.medscape.com/Home/ MoneyMedicine/MedscapeArticles/ehealth.html. Accessed November 18, 2000. 3 1. Reece RL. The health care consumer gospel according to Harvard School: a talk with Regina Herzlinger, DBA. The Ph1ysician Executtive. 2000:36-40. 32. Kassirer JP Patients, physicians, and the Internet. Health Affairs. 2000;19:115-123. 33. Epstein RS, Sherwood LM. From outcomes research to disease management: a guide for the perplexed. Ann Intern Med. 1996; 124:832-837. 34. Perednia DA, Allen A. Telemedicine technology and clinical applications. JAMA. 1995;273:483-488. 35. Brown N. What is telemedicine? [Telemanagement Research Center University of Oregon web site]. Available at: http://distance.ohsu.edu/-BBuser/document/MINF5 10/ telemedicine.html. Accessed October 10, 2000. 36. Friedman RH, Kazis LE, Jette A, et al. A telecommunications system for monitoring and counseling patients with hypertension. Impact on medication adherence and blood pressure control. Am J Hvpertens. 1996;9:285-292. 37. DaSilva RV, Rivkin S. Increasing efficiency through point of prescribing technology. Drug Benefits Trends. 2000; 12:45-49.

VOL. 95, NO. 11, NOVEMBER 2003


38. Goldstein MK, Hoffman BB, Coleman RW, et al. Implementing clinical practice guidelines while taking account of changing evidence: ATHENA DSS, an easily modifiable decision support system for managing hypertension in primary care. Proc AMIA Symp. 2000;300-304. 39. Johnson PD, Tu S, Booth N, et al. Using scenarios in chronic disease management guidelines for primary care. Proc AMIA Svmp. 2000;389-393.

RECRUITMENT ANNOUNCEMENT

THE W. K. KELLOGG FOUNDATION SCHOLARS IN HEALTH DISPARITIES 2004 GRANT CYCLE The W.K. Kellogg Foundation and the Center for the Advancement of Health are pleased to announce recruitment for the 2004 cohort

of W.K. Kellogg Foundation Scholars in Health Disparities.

We Welcome Your Comments The Journal of the National Medical Association welcomes your Letters to the Editor about articles that appear in the JNMA or issues relevant to minority health care. Address correspondence to ktaylor@ nmanet.org.

Neuroradiologist The Department of Radiology at Beth Israel Deaconess Medical Center (BIDMC), Harvard Medical School, is seeking applicants for a faculty position in Neuroradiology. Responsibilities include

solutions for health disparities by race/ethnicity, gender and income/socioeconomic status.

clinical service; teaching of medical students, residents, and fellows; and research in Neuroradiology. The Department has five whole body GE and Siemens MRI systems (Two 3-T, Three 1.5T). One of the 3T systems is dedicated to research. There is a small bore

Research topics funded by this program include the ways by which social, economic, political, environmental and educational

high field imaging and spectroscopy research system as well. There are seven multislice CT instruments (a mixture of 16, 8, and 4 slice

This program challenges post doctoral investigators from a variety

of disciplines such as economics, sociology, political science, public health and law, to examine the causes and consider policy

inequalities and institutional racism play a role in the creation of

health disparities, and the development of policy initiatives that might reduce these disparities.

units). Angiography is performed in a bi-plane digital room. There are ample research opportunities with excellent support in tech-

Benefits: * Stipends of up to $52,500/year

nique development, MR spectroscopy, perfusion imaging, and image processing. Candidates for this position must have board cer-

* Fringe benefits

* Research/travel expenses Program Sites: Up to 15 Scholars will be funded to undertake

tification in diagnostic radiology, have or be eligible for CAQ in neuroradiology, and be eligible for licensure in the state of Massachusetts. Salary and academic rank will be commensurate

postdoctoral research at one of the following participating sites: ... Society. Health and * Harvard Center for Health fo and Societywith Harvard Center experience and qualifications. Interested applicants should * University of Michigan Institute for Social Research send curriculum vitae to David B. Hackney, MD, Chief of * Morgan State University Public Health/Urban Health Policy Center Neuroradiology, Professor of Radiology, Harvard Medical School, * University of California, San Francisco Center on Social Department of Radiology, Beth Israel Hospital, 330 Brookline Ave., Disparities in Health Shapiro 4th Floor, Boston, MA 02215 (617) 667-2552; fax (617) * Columbia University Mailman School of Public Health Department of Epidemiology 667-8212; e-mail [email protected]. * Morehouse School of Medicine National Center for Primary Care .

Application: For further information and application materials, visit www.cfah.org or contact Barbara Krimgold, Director of the Scholars Program, at bkrimgoldecfah.org after October 15, 2003.

Beth Israel Deaconess Medical Center is an EEO/AA employer. Visit our website at http://radiology.bidmc.harvard.edu

Deadline for receipt of applications is January 15, 2004.

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

VOL. 95, NO. 11, NOVEMBER 2003 1056