Guidelines for Assessing Outcomes ofAntihypertensive Treatment

Michael Weber, MD

Because of the changing nature of medical practice andthe need to more accurately monitor and quantify theclinical benefits and cost-effectiveness of treatment, theSixth Report of the Joint National Committee (JNC VI)recommends that the outcomes of antihypertensive ther-apy be classified as short, intermediate, and long term.Short-term outcomes are most relevant to the practitio-ner and include such measures as blood pressure con-trol, laboratory changes, and quality of life. In contrast,long-term outcomes, typically measured in randomizedclinical trials, are of particular interest to policy makersand guidelines writers, and focus on whether treatmentsaffect survival and the incidence of major cardiovascular

events. Intermediate-term outcomes, usually measurablewithin months of starting treatment, deal with such clin-ical surrogates as treatment-induced changes in left ven-tricular structure, arterial compliance, and renal func-tion. No longer are the traditional short-term outcomesadequate to describe a new drug; hypertension special-ists, formulary committees, healthcare economists, andeven regulatory agencies now expect sponsors to planstudies that define a drug’s full range of outcomes. Thisarticle briefly discusses some of the criteria for theseoutcome measures. Q1999 by Excerpta Medica, Inc.

Am J Cardiol 1999;84:2K–4K

A lthough awareness of hypertension is greater thanit was 2 decades ago, national surveys, such as

the National Health and Nutrition Examination Survey(NHANES III), indicate that hypertension is still notadequately controlled.1 It is estimated that less thanone quarter of hypertensive patients have blood-pres-sure controlled at the target level of 140/90 mm Hg.Although intensive management of hypertension re-duces the incidence of stroke and heart failure, coro-nary events are still the most common result of hy-pertension.

Hypertension-related morbidity and mortality willnot decrease until changes are made in treatment pro-tocols, and these changes must be based on quantifi-able outcome measures. The Sixth Report of the JointNational Committee (JNC-VI) identifies short-, inter-mediate-, and long-term outcomes for the evaluationof antihypertensive therapy.2 Short-term outcomes,such as blood pressure levels, are commonly mea-sured at the initiation of antihypertensive therapy.Both short- and intermediate-term outcomes are usedto assess how successfully hypertension therapy con-trols blood pressure and prevents organ damage.Long-term outcomes assess the success of hyperten-sion treatment in the prevention of coronary disease,heart failure, stroke, renal failure, and mortality overmany years.

Because the ultimate goal of treating hypertensionis prevention of the aforementioned, long-term effects,one can make a strong argument for including long-term outcomes in hypertension management proto-cols. However, long-term outcomes generally takeseveral years to document, so the cost–effectiveness

of gathering these data versus their potential healthbenefits becomes a separate, measurable outcome.

SHORT-TERM OUTCOMESShort-term outcomes (e.g., blood pressure levels,

treatment compliance, clinical chemistries, and sideeffects) assess the initial effects of therapy. Table Iprovides a complete list of short-term outcomes andhow they are measured.

Short-term outcomes are generally given numericalvalues. Recent data from studies such as the Hyper-tension Optimum Treatment (HOT) trial have shownthat clinical events occur less frequently in patientswhose diastolic blood pressure (DBP) levels are in thelow 80s, compared with patients whose DBP levelsare in the upper 80s.3 In addition, reducing bloodpressure levels to,125/75 mm Hg appears to mosteffectively preserve renal function.4 The JNC-VI alsoadvocates a lower target blood pressure level of,130/85 mm Hg for patients with concomitant dis-ease or evidence of organ damage.2 Addressing theseshort-term outcomes may reduce the incidence oflong-term morbidity.

Many antihypertensive therapies have side effectsthat diminish patients’ quality of life and thus lead topoor compliance. Quality-of-life validation instru-ments were introduced to allow more quantifiablemeasures of how hypertension and hypertension treat-ment affect patients’ life styles. These instrumentsallow physicians to select drugs, or adjust drug doses,on the basis of improving a patient’s quality of life(i.e., diminishing the side effects that discourage pa-tients from continuing with treatment).

Other quantifiable, short-term measures includeroutine laboratory tests, the number of drugs used, thefrequency of physician visits, and the cost of treat-ment.

Some short-term outcomes are more difficult tomeasure because they rely on self-reporting by pa-tients. These include lifestyle modifications and pa-

From the Department of Medicine, The Brookdale University Hospitaland Medical Center, Brooklyn, New York, USA.

Address for reprints: Michael Weber, MD, Department of Medi-cine, The Brookdale University Hospital and Medical Center, StateUniversity of New York, Health Science Center at Brooklyn, OneBrookdale Plaza, Brooklyn, New York, 11212-3198.

2K ©1999 by Excerpta Medica, Inc. 0002-9149/99/$20.00All rights reserved. PII S0002-9149(99)00402-6

tient global evaluations. Despite the subjective natureof self-evaluation, a patient’s perception of the suc-cess of therapy is an important variable of long-termcompliance.

INTERMEDIATE-TERM OUTCOMESTable II presents intermediate-term outcomes,

which are measured 6 months after the start of anti-hypertensive therapy. At this point, presumably, doseadjustments have been made and the patient’s treat-ment regimen and short-term measurements are fairlystable. The purpose of intermediate-term outcomes isto monitor signs indicating organ damage, such asimpaired renal function, electrocardiographic changes,or altered left ventricular function. There is no clearlyestablished causal link between intermediate-term out-comes and long-term clinical events. However, the rela-tion between chronic hypertension and clinical eventssuggests that monitoring of intermediate-term outcomesmay aid in the early identification of organ damage.

Intermediate-term outcomes include all short- andintermediate-term clinical and economic outcomes,such as lost work days and health resources utilization(Table II). Lost workdays during initiation of treat-ment may be attributed to visits to the clinic for bloodpressure monitoring and general assessment, whereasdays missed in the intermediate phase might resultfrom hypertension-related complications. Physiciansmust consult with patients during this phase to deter-mine whether missed workdays or healthcare utiliza-tion are clearly related to hypertension symptoms ortreatment. The economic impact of these outcomes isquantifiable.

LONG-TERM OUTCOMESThe events of cardiac, cerebrovascular, renal, and

vascular diseases are associated with long-term hyper-tension. Prevention of morbidity and mortality due toclinical events is the ultimate goal of antihypertensivetherapy. However, it generally takes years, even dec-

TABLE I Short-Term Outcome Measures

Outcome How Measured

Blood pressure Numerical values (direct• Office measure)• Indices of 24-hr efficacy

Symptomatic side effects Present or absentMild/moderate/severe on

analog scaleQuality of life Numerical values

• Patients• Perception of spouses or others

Compliance with treatment Yes/no• Drug utilization, self-reported• Timely prescription refills Percentage achievement of• Adherence to clinical

appointmentsoptimal performance

Lifestyle modifications Numerical values (direct• Weight control measure or self-reported)• Weekly aerobic exercise Yes/no• Alcohol reduction• Smoking cessation

Routine clinical chemistries Numerical value• Electrolytes (measured variables)• Renal function• Glucose

No. of drugs taken Numerical value(measured variables)

Frequency of physician visits Numerical value• Complexity of treatment (measured variables)

Frequency of clinical tests Numerical value(measured variables)

Total direct medical costs of treatment* Numerical value(calculated)

Indirect costs of treatment Numerical value• Travel expenses (calculated cost and• Loss of productivity direct measure)• Increased insurance costs

Global evaluation: patient Analog scale• Subjective perception of

satisfactionGlobal evaluation: physician Analog scale

• Subjective perception ofsatisfaction

*Costs 5 drug cost 1 monitoring cost 1 side effects cost 2 savings inmanagement of underlying disease.

TABLE II Intermediate-Term Outcome Measures

Outcome How Measured

All short-term clinical andeconomic outcomes

See Table I

Concomitant metabolic risk factors Numerical values (direct• Lipid profile measure)• Glucose tolerance, HbA1c

Renal outcome measures Numerical value (direct• Renal function measure)• Proteinuria/microalbuminuria

Cardiac outcomes: ischemia Present/absent: yes/no• ECG changes New findings: yes/no

Cardiac outcomes: left ventricularstructure and function

Numerical values (directmeasure): yes/no

• ECHO left ventricular mass• Doppler diastolic function• ECG left ventricular

hypertrophy

Cardiac outcomes: arrhythmia Present/absent• Symptoms Numerical values• ECG/ambulatory monitoring

Arterial compliance Numerical value (direct• Noninvasive estimates of

arterial stiffnessmeasure)

Clinical evidence of Present/absent: yes/noatherosclerosis (or measure of

• New-onset angina pectoris severity of vascular• Findings of carotid stenosis changes)• Renovascular findings• Changes in optic fundi

Patient days lost from work Numerical value(measured variable)

Health resources utilization Numerical value• Related hospitalizations• Clinical visits• Advanced imaging or other

tests

ECG 5 electrocardiogram; ECHO 5 echocardiogram; HbA1c 5 hemoglo-bin A1c.

A SYMPOSIUM: ANGIOTENSIN II RECEPTOR ANTAGONISTS 3K

ades, for the long-term outcomes of hypertension andantihypertensive therapy to become apparent.

Long-term outcomes include all short- and inter-mediate-term outcomes, as well as the incidence ofhypertension-related clinical events and mortality (Ta-ble III). Mortality due to cardiovascular disease can beattributed, at least in part, to hypertension, whereasmortality due to causes other than cardiovascular dis-ease may indicate that antihypertensive therapy waseffective during the patient’s lifetime.

Long-term outcomes may aid assessments of thecosts of antihypertensive treatment and the value ofantihypertensive therapy in the context of reducing

morbidity and mortality. By using large populationstudies, one can evaluate the cost per year of each lifesaved, as well as quality-adjusted life years. In addi-tion, long-term evaluation of antihypertensive agentsmay identify the drugs of optimal benefit to patients(e.g., those with antiproliferative properties that caninhibit vascular changes).

CONCLUSIONMost clinical studies use only short-term outcome

measures to determine the efficacy of a particular drugbecause they can be assessed immediately and eco-nomically. The causal connection between short- andintermediate-term outcomes and long-term morbidityand mortality is still under investigation. However, all3 ranges of outcomes should be considered in assess-ments of how effectively antihypertensive therapiesprolong life and reduce clinical events.

Short-term measures are easily attainable. How-ever, intermediate- and long-term measures are ob-tained over many years of patient monitoring, and theprocess is costly. If clinical trials are expanded toinclude short-, intermediate-, and long-term outcomes,they may lead to hypertension management regimensthat reduce morbidity and mortality, improve qualityof life, and provide economic justification for pre-scribing newer, promising drugs.

1. Burt VL, Culter JA, Higgins M, Horan MJ, Labarthe D, Whelton P, Brown C,Roccella EJ. Trends in the prevalence, awareness, treatment, and control ofhypertension in the adult US population: data from the health examinationsurveys, 1960 to 1991.Hypertension1995;26:60–69.2. The sixth report of the Joint National Committee on prevention, detection,evaluation, and treatment of high blood pressure.Arch Intern Med1997;157:2413–2466.3. Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S,Menard J, Rahn KH, Wedel H, Westerling S. Effects of intensive blood-pressurelowering and low-dose aspirin in patients with hypertension: principal results ofthe Hypertension Optimal Treatment (HOT) randomised trial: HOT Study Group.Lancet1998;351:1755–1762.4. Lazarus JM, Bourgoignie JJ, Buckalew VM, Greene T, Levey AS, Milas NC,Paranandi L, Peterson JC, Porush JG, Rauch S, Soucie JM, Stollar C. Achieve-ment and safety of a low blood pressure goal in chronic renal disease: theModification of Diet in Renal Disease Study Group.Hypertension1997;29:641–650.

TABLE III Long-Term Outcome Measures

Outcome How Measured

All short-term and intermediate-termclinical and economic outcomes

See Tables I and II

Mortality Yes/no• Cardiovascular• Noncardiovascular

Cardiac events Yes/no• Sudden death• Myocaridal infarction• Angina pectoris• Coronary artery bypass surgery• Percutaneous transluminal

coronary angioplasty• Congestive heart failure• Clinically significant arrhythmia

Cerebrovascular events Yes/no• Strokes, hemorrhagic or

thrombotic• Transient ischemic attacks

Renal events Yes/no• Renal insufficiency• Renovascular disease Numerical value

Aortic and peripheral vasculardisease

Yes/no

Cost–effectiveness Numerical values (calculated)• Cost per year of life saved• Cost of health resources• Patient and indirect costs• Cost of lost productivity

4K THE AMERICAN JOURNAL OF CARDIOLOGYT VOL. 84 (2A) JULY 22, 1999