Improving Treatment Effectiveness in Hypertension I T IS a disturbing fact that only 27% of patients with hypertension have their blood pressure (BP) con- trolled at normotensive lev- els.' This has occurred despite the availability of effective antihyper- tensive drug treatment. It is appar- ent that, unless we can improve our present methods of delivering treat- ment and improving compliance, we will continue to leave more than two thirds of the hypertensive popula- tion with their BP uncontrolled. This article will suggest possible rea- sons and remedies for this poor per- formance. It evaluates diet treat- ment, drug therapy, and interventions for improving compliance and sug- gests changes for improvement. They are not meant to be final recommen- dations. Rather, the article is in- tended to focus concern about the subject, to suggest some possible so- lutions, and to open a constructive dialogue. DIET TREATMENT Most hypertensive patients are asymptomatic. Because they feel well, they are not motivated to make any drastic changes in their lifestyle. For example, weight-reducing diets in obese patients are successful only over the short term. After 3 to 5 years, almost all have returned to their previous weight.2,3 Most patients do not adhere to restrictive diets for long periods. The presently recommended approach to beginning treatment, es- pecially in mild hypertension, is diet therapy. The most frequently pre- scribed dietary interventions are From the Department of Veterans Affairs, Veterans Affairs Medical Center, and Georgetown University School of Medicine, Washington, DC. weight reduction and sodium re- striction A recent trial of diet plus exercise was evaluated for the treat- ment of hypertension4 The diet was low in energy intake and sodium. In- tensive indoctrination and motiva- tion sessions were used to obtain compliance. The diet-exercise inter- vention was maintained for 1 year. At the end of that time, body weight was reduced by an average of 4.5 kg and BP had fallen by an average of 10.6 mm Hg systolic and 8.1 mm Hg diastolic. Although this antihyper- tensive response was significant, it would not be great enough to nor- malize the BP of most patients with pretreatment systolic BP greater than 150 mm Hg or diastolic levels greater than 97 mm Hg. Also, addition of various antihypertensive drugs re- sulted in a significantly greater fall in BP. Whether their results with diet alone could be maintained over the long-term is questionable (see the discussion of the Trial of Non- pharmacologic Interventions in the Elderly below). It is also question- able whether the intensive educa- tional effort for maintaining the diet could be duplicated in a general practice setting. Recently, a diet rich in fruits and vegetables and low in total and saturated fats was found to lower the BP by 11.4 mm Hg systolic and 5.5 mm Hg diastolic in hypertensive pa- tients (Effects of Dietary Patterns on Blood Pressure trial).* The dura- tion of the dietary intervention was only 8 weeks. All meals were pre- pared at the various centers to en- sure compliance. The subjects ate most of their meals at the centers. Further investigation will be neces- sary to determine whether this diet could be effective in the usual clini- cal setting and over the long-term. A longer-term trial of dietary in- tervention was carried out by the Trial of Nonpharmacologic Inter- ventions in the Elderly Collabora- tive Research Group.6 They com- pared the effects of withdrawal of antihypertensive drug treatment in 4 groups of patients treated as fol- lows: (1) low-sodium diet, (2) weight-reducing diet, (3) both di- ets combined, and (4) usual care (no dietary interventions). Special edu- cational efforts were used to help gain compliance. Antihypertensive drug treatment was withdrawn 3 months after randomization, and the patients were followed up for a me- dian of 29 months. During the ini- tial period after withdrawal of anti- hypertensive drugs, normal BP was maintained in approximately 93% of patients in the 3 diet intervention groups and in 87% of the usual- care patients. However, during the succeeding months, despite contin- ued dietary restrictions, more and more patients experienced a return of their hypertension requiring the reinstitution of drug treatment. At 29 months of follow-up, the per- centage of patients receiving the low- sodium diets who remained normo- tensive still without drug treatment had fallen from 93% to 38%, with the trend suggesting that essentially all patients would have required drug treatment within 5 years (Figure 4 of their article). These results indi- cate the same problem that exists with weight-loss diets. There is loss of effectiveness over the long-term despite intensive efforts at maintain- ing compliance. A meta-analysis of 28 con- trolled trials of low-sodium diets in hypertensive patients indicated an average decrease of only 3.7 mm Hg in systolic BP for a lOO-mmol/d re- duction in sodium excretion.' The average diastolic reduction of 0.9 mm Hg was not significant. By con- trast, the combination drug treat- ARCH INTERN MED/VOL 159, NOV 22, 1999 WWW.ARCHINTERNMED.COM 2517 Fixed-Dose Antihypertensive Drug Combinations* Bisoprotot fumarate-hydrochiorothiazide Bisoprotot fumarate-hydrochiorothiazide Metoprolot tat-hate-hydrochtorothiazide Metoprolot tartrate-hydrochtorothiazide Nadotot-bendroftumethiazide Nadotot-bendroftumethiazide Propranoloi hydrochloride (extended release)-hydrochlorothiazide Propranoloi hydrochloride (extended release)-hydrochlorothiazide Timoiot mateate-hydrochlorothiazide Timoiot mateate-hydrochlorothiazide ACE inhibitors and diuretics ACE inhibitors and diuretics Benazepril hydrochloride-hydrochtorothiazide Benazepril hydrochloride-hydrochtorothiazide Captoprit-hydrochlorothiazide Captoprit-hydrochlorothiazide Enataprit mateate-hydrochiorothiazide Enataprit mateate-hydrochiorothiazide Lisinoprtl-hydrochlorothiazide Lisinoprtl-hydrochlorothiazide Angiotensin Ii receptor antagonists and diuretics Angiotensin Ii receptor antagonists and diuretics Losartan po~ssium~ydrochloro~~~~de Losartan po~ssium~ydrochloro~~~~de Calcium antagonists and ACE inhibitors Calcium antagonists and ACE inhibitors Amiodipine besyta~e-benazeprit hydrochloride Amiodipine besyta~e-benazeprit hydrochloride Diitiazem hydrochloride-enalaprit maleate Diitiazem hydrochloride-enalaprit maleate Verapamil hydrochloride (extended release)-trandolapril Verapamil hydrochloride (extended release)-trandolapril Fetodipine-enataprtt mateafe Fetodipine-enataprtl mateafe Ziac Ziac Lopressor HCT Lopressor HCT Corzide Corzide tnderide LA tnderide LA Timotide Timotide Lotensin HCT Lotensin HCT Capozide Capozide Vaseretic Vaseretic Prinzide, Zestoretic Prinzide, Zestoretic Hyzaar Hyzaar Lotrel Lotrel Teczem Teczem Tarka Tarka LeXXet LeXXet Drug Combination Drug Combination p-Adrenergic blockers and diuretics p-Adrenergic blockers and diuretics Atenolol-chlorthatidone Atenolol-chlorthatidone Brand Brand Tenoretic Tenoretic *This partial listing is representative of currently popular combinations. ACE indicates angiotensin-convetiing eniyme. ment described below lowered av- erage systolic and diastolic BP by 22 and 12 mm Hg, respectively. It is more difficult to motivate patients to change their diet than it is to have them take medication. Diets impose a major and unpleasant change in life- style. When diets fail to control the BP, patients may become discour- aged and discontinue diet therapy completely. For these and other rea- sons, drug therapy should be given preference over diet treatment. few visits so as to rule out the pres- ence of "white coat" hypertension. However, once treatment is begun, drug titration should proceed cau- tiously but expeditiously to achieve BP control. COMBINATION DRUG THERAPY DRUG TREATMENT To encourage compliance, treat- ment should be simple and should interfere as little as possible with the patient's usual lifestyle. Treatment should reduce the BP promptly and effectively with few or no side ef- fects and continue to do so over the long term. The antihypertensive drugs in use today are usually well tolerated, are mostly free of serious adverse effects, and seldom cause minor side effects. Depending on how it is prescribed, treatment with drugs can be simple or complicated, effective or ineffec- tive. Optimal drug therapy should meet the following criteria: effective BP control, simplicity (1 tablet once per day), safety, cost-effectiveness, and little or no interference with normal lifestyle. Combination therapy (Table) is more effective than mono- therapy.8,9 It will control the hyper- tension in patients who have failed to respond to several monothera- pies.8 Also, doses of each compo- nent are usually lower than those re- quired for monotherapy, which tends to minimize side effects. The time required from begin- A fixed-dose combination tab- ning treatment to control of the hy- let (Ziac) composed of the PI- pertension can be important. Pa- blocking drug bisoprolol fumarate tients may become discouraged by in doses of 2.5, 5, and 10 mg plus a series of ineffectual treatments. It hydrochlorothiazide, 6.25 mg, was is therefore important to control the compared with titrated doses of am- hypertension as expeditiously as lodipine besylate or enalapril male- possible. Of course, no treatment ate, both given as monotherapy.' should be instituted during the first Goal diastolic BP (590 mm Hg or a reduction 210 mm Hg) was achieved in 71% of those taking the combination, 69% receiving a&lo- dipine alone, and 45% randomized to enalapril. Adverse effects were un- common with all drugs and were least frequent with the combina- tion therapy). Retail cost of treat- ment with Ziac is approximately $0.60 per day. Thiazides plus P-blockers have until recently been the only combi- nation drug treatment tested in long- term morbidity-mortality trials. One example is the Systolic Hyperten- sion in the Elderly Program,`O in which treatment consisted of chlorthalidone, to which atenolol could be added if needed to achieve goal BP (5 160 mm Hg systolic). At the 5-year visit, 69% of the treated patients achieved goal BP with ei- ther diuretic alone or the combina- tion with P-blockers. Stroke and car- diovascular complications were significantly reduced in the drug treatment group. Several combinations of a cal- cium channel blocker and an angio- tensin-converting enzyme (ACE) inhibitor are commercially avail- able. Fixed-dose combinations of diltiazem hydrochloride and enala- pril maleate" were compared with monotherapy in a trial involving 891 hypertensive patients. Two dose lev- els of diltiazem hydrochloride were tested, 120 and 180 mg, each com- bined with 5 mg of enalapril male- ate (Tagem). After 12 weeks of treat- ment, the reduction of diastolic BP averaged 7.6 and 8.3 mm Hg below pretreatment BP with the low- and high-dose diltiazem combinations, respectively, which was a signifi- cantly greater reduction than with either drug given alone. Adverse re- actions were no more frequent with the combinations than with the single drugs. Similar results were ob- served in another trial that used a combination of benazepril hydro- chloride and amlodipine.12 Recently, 2 large trials used a calcium channel blocker to which an ACE inhibitor or other drug was usually added. Treatment was effec- tive in both trials in preventing car- diovascular events. The Syst-Eur Trial13 used nitrendipine in pa- tients with isolated systolic hyper- tension. Their results were similar ARCH INTERN MED/VOL 159, NOV 22,1999 WWW.ARCHINTERNMED.COM 2518 to those of the Systolic Hyperten- ,. sion_in the Elderly Program. The other trial (Hypertension Optimal Treatment trial) studied the effects of different levels of diastolic BP re- duction on cardiovascular morbid- ity and mortality.14 They found that reductions below 90 mm Hg were optimal. No additional benefit oc- curred below 85 mm Hg. The cal- cium channel blocker used was felo- dipine. Both of these trials used titration of doses not only of the pri- mary drug but also of the added drugs. This makes the therapeutic program more complicated than the simple fixed-dose combination. Such complexities may make treatment less successful in the real world. A diuretic was included in the Ziac combination described previ- ously because it is the most effec- tive of all drugs in enhancing the an- tihypertensive activity of other agents. This was demonstrated in a Veterans Affairs randomized trial in patients whose BP remained greater than 140/90 mm Hg after 2 succes- sive monotherapies.' They were then assigned in a randomized double- blind manner to various 2-drug com- binations. All possible combina- tions of 6 major classes of drugs were tested. The combinations that in- cluded a diuretic achieved a sys- tolic BP less than 140 mm Hg in 77% of patients and a diastolic BP less than 90 mm Hg in 69%. Combina- tions containing drugs other than a diuretic were less effective, even in- cluding the combination of an ACE inhibitor and a calcium channel blocker. The response rates to the thiazide combinations were impres- sive considering that these were pa- tients with treatment-resistant hy- pertension whose BP had failed to normalize with 2 consecutive mono- therapies. Despite their reputation, di- uretics have not demonstrated seri- ous toxic effects in the various clini- cal trials on prevention of morbidity and mortality. 15.1h Furthermore, thia- zides when used in combination are effective in small doses, which fur- ther minimizes the occurrence of side effects. Other fixed-dose combina- tions include enalapril maleate, 5 or 10 mg, with hydrochlorothiazide, 12.5 or 25 mg, and a combination of the angiotensin II receptor an- tagonist losartan potassium, 50 mg, with hydrochlorothiazide, 12.5 mg. In the Veterans Affairs trial de- scribed previously,' the ACE inhibi- tor-diuretic combination lowered the BP from a pretreatment mean of 156/ 100 mm Hg to 134/88 mm Hg. In an- other Veterans Affairs cooperative study, hydrochlorothiazide, 25 mg twice daily, plus captopril, 12.5 mg 3 times daily, reduced the average di- astolic BP by 16.6 mm Hg, a reduc- tion to normal levels in nearly all patients.l' Combinations with a diuretic plus an ACE inhibitor should be use- ful in the presence of a variety of co- morbid conditions involving the heart and kidneys. The ACE inhibi- tors reduce mortality both during myocardial infarction'* and in the post-myocardial infarction pe- riod." They also reduce left ven- tricular remodeling, which often oc- curs after myocardial infarcti0n.l" Both ACE inhibitors and diuretics markedly benefit patients with con- gestive heart failure,2',22 and they are both effective in reducing left ven- tricular hypertrophy associated with hypertension.23 In the kidneys, ACE inhibitors significantly reduce the microalbu- minuria often associated with hyper- tension.24 The ACE inhibitors lower intraglomerular BP, thereby slow- ing the progression of glomerular fibrosis2j including that occurring in diabetic nephropathy.26 Many physicians are reluctant to increase doses of antihyperten- sive drugs to effective levels. A re- cent survey found that 82% of phy- sicians failed to increase doses when indicated.27 Fortunately, the di- uretic combinations with a B-blocker or ACE inhibitor produce a high re- sponse rate with the initial doses, and only a single step-up is provided for patients who require additional medication. The mode of adminis- tration is simple: 1 tablet once daily of the weaker strength, and if the BP does not fall to normal, switch to 1 tablet per day of the second strength. The ACE inhibitors do not inter- fere with sexual activity or with cog- nitive functions. They do not in- duce weakness or fatigue or cause orthostatic hypotension. Small doses of diuretics probably also do not. The combination, therefore, is not only highly effective, it is also very well tolerated. COMPLIANCE Noncompliance is probably the ma- jor cause of failure to control hyper- tension, yet it is usually the most ne- glected. It is difficult to motivate asymptomatic patients in appar- ently vigorous good health to take medication every day for indefinite periods. Most of them are not con- cerned about the risk of complica- tions that may possibly occur far in the future. To help overcome their lack of interest, risk factors must be clearly stated and repeatedly emphasized. While the physician should lead this educational effort, most of it can usu- ally be delegated to knowledgeable and motivated nurses or social work- ers. Educational materials should also be provided, including pam- phlets and video programs such as those available from the American Heart Association or the National High Blood Pressure Education Pro- gram. Attention also must be paid to side effects that can cause patients to drop out of treatment. When- ever a new treatment is initiated, the major side effects should be dis- cussed and the patient assured that, if any should occur, another treat- ment will be substituted. There has been little research on compliance in the treatment of hypertension. The few published controlled studies suggest that com- pliance is improved by suitable in- terventions. In one study, 400 pa- tients were randomly assigned either to various compliance-promoting in- terventions or to no intervention.28 Three compliance-promoting in- terventions were used: (1) a lo- minute interview and counseling session by the physician at the time of initiating treatment, (2) enlist- ing a member of the patient's fam- ily to monitor pill taking and other compliance matters in the home, and (3) several group sessions led by a social worker. The patients were fol- lowed up for 5 years. All-cause mor- tality was 57.3% less in the experi- mental group than in the control patients. The BP was controlled in ARCH INTERN MED/VOL 159, NOV 22, 1999 UWW.ARCHINTERNMED COM 2519 65% of the experimental group compared with only 22% of the controls. In another trial, a hyperten- sion educational program was car- ried out in 2 rural counties in Ken- tucky.29 A neighboring county that had no educational program served as the control. A hypertension reg- istry was developed. Patients in the intervention counties received pe- riodic mailings concerning the vari- ous risks associated with hyperten- sion and the importance of treatment in preventing complications. This in- formation was reinforced by local newspaper articles and radio pro- grams. Lectures were given to local nurses on the importance of teach- ing compliance to their patients. At the end of 5 years of follow-up, the percentage of patients in the inter- vention group whose BP was con- trolled below 140/90 mm Hg in- creased from 24.8% to 39.7% (P<.OOl). There was no significant change in the control group. Car- diovascular risk factors were re- duced in the 2 intervention commu- nities and were increased in the control county. However, these and other studies30 have excited little interest in the problem of compliance. Home recording of BP is an- other procedure that is used not only to promote compliance but also as a guide in adjusting dosage. The BP is taken once or twice daily by the spouse or the patient and a record is kept of the date, time, and level of BP. Increased motivation is pro- vided by making the patient an ac- tive partner in the process of con- trolling the BP. The patient also can see that persistent treatment is nec- essary for maintaining the BP at nor- motensive levels. However, some patients become upset with any upward fluctuations in BP. If they cannot accept that such fluctua- tions are normal, home recordings should be discontinued. One controlled trial that tested the effectiveness of home BP record- ings found that, after self-BP record- ings were instituted, compliance rose from 65% to 81%,31 a significant in- crease compared with the control group. In another trial, randomly se- lected patients recorded their BP at home.32 These patients showed a sig- nificantly greater degree of compli- ance than the control group. In a third tria1,33 patients who were known to be poor compliers either were randomized to usual clinic care or were asked to record their BP at home and to adjust dosage accord- ing to the readings. Compliance was improved in the home BP group as judged by pill counts and clinic attendance. Results of these studies are encouraging. However, larger- scale controlled trials are needed to more adequately evaluate this method. Care of patients by specially trained nurses has been advocated in the treatment of hyperten- sion34.35 and diabetes.36 Increased compliance has been claimed to re- sult from this procedure. More time can be spent with each patient than the physician can usually spare. The nurse can emphasize the need for faithful compliance to the regimen and can reiterate the importance of BP control in the prevention of com- plications Of course, the effective- ness of such a program will depend on the dedication and competence of the nurse. Haynes et a137 reviewed a num- ber of control trials that tested vari- ous interventions for improving compliance. They included 5 trials in hypertensive patients. Four of the 5 trials found statistically signifi- cant (although not great) improve- ment in compliance in the interven- tion groups. The most effective interventions included single daily dosage, patient education regard- ing risk reduction, use of nurses in education, reinforcement and moni- toring compliance by the spouse, self-measurement of BP, and sup- port groups. Despite significant improvement with such programs, there were still many failures. The Canadian Coalition for Blood Pressure Prevention and Con- trol established a national advisory committee on compliance in the management of high BP.38 They re- ported that estimates of patients' non- compliance rates vary widely in dif- ferent surveys but average about SO%, including dropouts.39 They present 4 recommendations for improve- ment: (1) education of patient and family in the importance of risk re- duction, with self-measurement of BP in nonadherent patients; (2) once- daily dosage and simplified tfeat- ment regimens; (3) having nurses re- view with patients their compliance status at each visit; and (4) using less expensive medications when pos- sible. Noncompliance is the most im- portant remaining problem in the treatment of hypertension. It needs more attention and research fund- ing than it currently receives. CONCLUSIONS Despite the availability of a variety of effective antihypertensive drugs, BP is controlled in fewer than one third of hypertensive patients. Since present methods of treatment ap- pear inadequate, they need to be changed. The currently popular monotherapy aimed at treating co- morbid conditions as well as reduc- ing BP has not achieved the goal of normalizing the BP in an accept- able percentage of patients. Fixed-dose combination tab- lets, such as diuretic+-blockers or diuretic-ACE inhibitors, confer much higher response rates. They provide a simple l-pill once-daily regimen that facilitates compliance while the low dose of each component mini- mizes side effects. The diuretic- ACE inhibitor combination also pro- vides treatment against many of the comorbid conditions frequently as- sociated with hypertension. In addition, moderate exer- cise should be encouraged, not only because it may produce further mod- est fall in BP but also because it reduces the risk of coronary heart disease.40 Other compliance-promoting procedures that have proved useful include (1) repeated emphasis on prevention of risk factors by clinic personnel in addition to pam- phlets, video presentations, etc; (2) giving nurses more responsibility for follow-up; (3) enlisting the help of the patient's spouse or other family member in monitoring daily pill tak- ing; and (4) recording the extent of noncompliance by having the phar- macist report the time intervals be- tween refills. Success in obtaining increased compliance has been claimed for the use of BP recording in the home. Favorable results have also been ARCH INTERN MED/VOL 159, NOV 22,1999 WWWARCHINTERNMED COM 2520 reported for the use of specially trained nurses to act as primary caregivers for patients with hyper- tension. The problem of compli- ance has also been recognized and discussed in the Sixth Report of the Joint National Committee on De- tection, Evaluation, and Treatment of High Blood Pressure.41 The management of noncom- pliance is difficult, time-consum- ing, and sometimes frustrating, but it appears to be most important for improving the effectiveness of treat- ment. We should at least attempt to improve on the present unaccept- ably high percentage of patients whose BPS are not controlled. To ac- complish this, we must be willing to try various approaches that may prove more effective than the pro- cedures used at present. Edward D. Freis, MD ~y~ert~sion Research Clinic 151E Veterans Ajfairs Medical Center 50 Irving St NW Washington, DC 20422 I thank Patricia Mohanfor her invalu- able assistance in the prepaF-ation of the manuscript. 1. Burt VL. Wheiton P, Rocceila EJ, et al. Preva- lence of hypertension in the US adult population: results from the Third National Health and Nutri- tion Evaluation Survey, 1988-l 991. Hype&n- sion. 1995;25:305-313. 2. Wadden TA, Sternberg JA, Letizia KA, et al. Treat- ment of obesity by a very low calorie diet, behav- iour and their contribution: a five-year perspec- tive. /ntJ Obes. 1989:13:39-46. 3. Garner DM, Waseiey SC. Confronting the failure of behavioral and dietary treatment for obesity, C/in PsycholRev. 1991;11:?29-780. 4. Liebson PR, Grandits GA, Diangumba S, et al. Comparison of five antihypertensive monothera- pies and placebo for change in left ventdcular mass in patients receiving nutritional hygienic therapy in the Treatment of Mild Hypertension Study (TOMHS). Ci~cu/a~io~. 1995;91:698-706. 5. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pres- sure: DASH Collabo~tive Research Group. bingo JMed. 1997;336:1117-1124. 6. Whelton PK, Appel LJ, Espeland MA, et al, far the TONE Collaborative Research Group. Sodium re- duction and weight loss in the treatment of hy- pertension in older persons: a randomized con- trolled Trial of Nonpharmacologic ~nte~entions in the Elderly (TONE) [published correction ap- pears in JAMA. 1998;279:1954]. JAMA. 1998; 279:83Q-846. 7 8 9 10 11. 12. 13. 14. 15. 16. 17. 18. Midgiey JP, Matthew AG, Greenwood CM, Logan AG. Effect of reduced dietary sodium on blood pressure: a meta-analysis of randomized con- trolled trials. JAMA. 1996:275:1590-1597. Materson BJ, Reda DJ, Cushman WC, et al. Re- sults of combination antihype~ensive therapy after failure of each of the components. J Hum HP pertens. 1995;9:791-796. Prisant LM, Mathew RN, Papadem~riou V, et al. Low-dose drug combination therapy: an aiterna- tive first-line approach to hypertension treat- ment. Am He&J. 1995;130:359-366. SHEPCooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA. 1991;265:3255- 3264. Cushman WC, Cohen JD, Jones RP, et al. Com- parison of the fixed combination of enalapriii diltiazem ER and their monnthe~pies in stage 1 to3essential hypertension. Am JHyperfens. 1998; 1 t :23-30. Frishman WH, Ram VS, ~cMahon G, et al. Com- parison of amlodipine and benazeprii in patients with systemic hypertension: a mndomized double- blind, placebo-controlled parallel-group study. J C/in Pharmacol. 1995;35:1060-1068. Staessen JA, Fagard R, L&garde 5, et al. Ran- domized double-blind comparison of placebo and active treatment for older people with isolated sys- tolic hype~ension (Syst-Eur Trial). Lancet 1997; 350:757-784. Hansson L, Zanchetti A, Carruthers SG, et al. Ef- fects of intensive blood-p~ssure io~ring and Iow- dose aspirin in patients with hypertension: prin- cipal results of the Hypertension Optimal Treatment (HOT) RandomizedTrial. Lancel. 1998;351:1755- 1762. Freis ED. Diuretic-induced hypokalemia: the de- bate over its relationship to cardiac ar~~hmias. PosfgradMed. 1987:81:123-129. FreisiD. Adverse effects of diuretics. OrugSafe& 1992;7:364-373. Freis ED, Materson BL. Captopril in mild to mod- erate hypertension: effect of low dose and of added diuretic. Prog fhsrmacQ~ 19~:5:87-92. Ryan TJ, Anderson JL, Antman EM, et al. ACC/ AHA guidelines for the management of patients with myocardial in~rction: executive summary: a report of the American College of Cardiology/ American Heart Association Task Force on Prac- tice Guidelines (Committee on Management of Acute Myocardial Infarction). Circulation. 1996; 94:2341-2350. 19. Pfeffer MA, Braunwald E, Maye LA, et al. Effect of captopril on mortality and morbidity after myo- cardiai infarction: results of the Survival and Ven- tricular Enlargement Study. N EnglJ Med. 1992; 327:669-672. 20. AIRE Study investigators. Effects of ramipril on mo~ali~ and morbidi~ of survivors of acute myo- cardial infarction with clinical evidence of heartfail- ure. Lancet. 1993;342:821-828. 21. Kuher J, Tarp-Pedenen C, Carlson JE, et al. Aciini- cai trial of the angiotensin-converting-enzyme in- hibitor trandolapril in patients with left ventricu- lar dysfunction after myocardial infarction: Trandoiapril Cardiac Evaluation (TRACE) Study Group. rVEr?~l JMed. 1995;333:1670-1676. 22. Oe~wania PC. Angiotensin-conveying enzyme in- hibitors in congestive heart failure. Arch intern Med. 1990;15O:t798-1805. 23. Schmieder RE, Martus P. Klingbeil A. Reversal of 24. 25. 26. 27. 28. left ventricular hypertrophy in essential hyperten- sion: a meta-analysis of randomized double- blind studies. JAMA. 1996;275:1507-1513. Bianchi S, Bigazzi R. Balderi G. et al. Microaibu- minuria in patients with essential hype~ension: effects of an angiotensin enzyme inhibitor and of a calcium channel blocker. Am JHyperfens. 1991; 4:291-296. Bjareck S, Nyberg G, Muier H. Beneficial effects of angiotensin converting enzyme inhibitor on re- nal function in patients with diabetic nephropa- thy. BMJ. 1986;293:976-974. Lewis EL, Hunskker LG, Pain RP. The effect of con- verting enzyme inhibition on diabetic nephropa- lhy. N Eng/ J Med. 1993;329:1456-1462. Sever PS. Blood pressure control for the hyper- tensive patient: what can we do better? AmdHy- pertens. 1997;1O(suppi 1):128S-130s. Morisky DE, Levine DM, Green LW, et al. Five- year blood pressure control and mo~ali~follow- ing health education for hypertensive patients. Am JPublic Health. 1983;73:153-162. 29. Kotchen JM, McKean HE, Jackson-Thayer S, Moore RW, Straus R, Kotchen TA. impact of a rural high blood pressure control program on hype~ension control and ~rdiovascular disease mortality. JAMA. 1986;255:2177-2182. 30. Farquhar JW. Wood PD, Bieitrose H. et al. Com- munity edu~tion for ~rdiovascuiar health. Lan- cet. 1997;1:1192-1195. 31. Edmonds 0, Forrester E, Grath H, etal. Does self- measurement of blood pressure improve patient compliance in hypertension? Hypertension. 1985; 3(suppi):S31-s34. 32. Haynes RB, Sackott DL. Gibson ES, et al. Im- provement in medication compliance in uncon- trolled hypertension. Lancef. 1976;1:1265- 1268. 33. Nessman DG, Carnahan JE, Nugent CA. Increasing compliance: patient-operated hyper- tension groups. Arch fnfefn Med. 1980;140: 1427-1430. 34. Logan RG, Miine BJ, Achber C, et al. Work-site treatment of hypertension by specially-trained nurses: a controlled trial. Lancef. 1979;2:1175- 1178. 35. Schultz JF, Sheps SC. Management of patients with hypertension: a hypertension clinic model. Mayo Clinic Proc. 1994;69:997-999. 36. Peters AL, Davidson MB, Ossorio RC. Manage- ment of patients with diabetes by nurses with support of subspecialization. HMO Pracf. 1995; 9:8-12. 37. Haynes RB, McKibbon KA, Kanani R. Systematic review of randomized trials of inte~ention to as- sist patients to follow prescriptions for medica- tions. Lancef. 1996;348:363-386. 38. Chockalingam A, Bacher M, Campbell N, et al. Ad- herence to management of high blood pressure: recommendations of the Canadian Coalition for High Blood Pressure Prevention and Control. Can J&b/k Healfh. 1998;89(suppl 1):5-7. 39. Feldman R, Bather M, Campbell N, et al. Adher- ence to pharmacoiogi~ management of hyper- tension. Can j Public Health. 1998;89(suppl 1): 16-18. 40. Papademetriou V, Kokkinos P. The role of exer- cise in the control of hypertension and cardiovas- cular risk. Curr Opin Nephrol Hyperfens. 1995;5: 459-462. 41. The Sixth Report of the Joint National Commit- tee on Prevention, Detection, Evaluation, and Treat- ment of High Blood Pressure. Arch fnfern Med. 1997;157:2413-2446. ARCH INTERN MED/VOL 159, NOV 22,1999 ~~.~R~HlNTERNMED.~OM 2521