The Cardiovascular Health Study: Design and Rationale Linda P. Fried, MD, MPH, Nemat 0. Borhani, MD, MPH, Paul Enright, MD, Curt D. Furberg, MD, Julius M. Gardin, MD, Richard A. Kronmal, PhD, Lewis H. Kuller, MD, DrPH, Teri A. Manolio, MD, MHS, Maurice B. Mittelmark, PhD, Anne Newman, MD, Daniel H. O’Leary, MD, Bruce Psaty, MD, PhD, Pentti Rautaharju, MD, PhD, Russell P. Tracy, PhD, and Philip G. Weiler, MD, MPH for the Cardiovascular Health Study Research Group (CHS). The Cardiovascular Health Study (CHS) is a population-based, longitudinal study of coronary heart disease and stroke in adults aged 65 years and older. The main objective of the study is to identify factors related to the onset and course of coronary heart disease and stroke. CHS is designed to determine the importance of conventional cardiovascular disease (CVD) risk factors in ok&r adults, and to identify new risk factors in this age group, especially those that may be protective and modifiable. The study design called f OTenrollment of 1250 men and women in each of four communities: Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Pittsburgh, Pennsylvania. Eligible participants were sampled from Medicare eligibility lists in each area. Extensive physical and laboratory evaluations were performed at baseline to identify the presence and severity of CVD risk factors such as hypertension, hypercholesterolemia and glucose intolerance; subclinical disease such as carotid artery atherosckrosis, left ventricular enlargement, and transient is&m& and clinically overt CVD. These examinations in CHS permit evaluation of CVD risk factors in older adults, particularly in groups previously under-represented in epidemiologic studies, such as women and the very old. The first of two examination cycles began inJune 1989. A second comprehensive examination will be repeated three years later. Periodic interim contacts are scheduled to ascertain and verify the incidence of CVD events, the frequency of recurrent events, and the sequel& of CVD. Ann Epidemiol 1991; J :263-276. KEY WORDS:

cardiovascular disease, aged, epidemiology, risk factors.

INTRODUCTION Risk factors for coronary heart disease (CHD) and stroke, and for other cardiovascular diseases (CVD), have been determined in middle-aged population groups, but the strength and importance of these factors in older individuals have not been clearly defined. Some studies have reported that the relative risks of CVD associated with From the Department of Medicine, The Johns Hopkins University, Baltimore, MD (L.P.F.), Department of Community Health, University of California, Davis, CA (N.O.B., P.G.W.), Mayo Pulmonary Services, Mayo Clinic and Foundation, Rochester, MN (P.E.), Department of Public Health Sciences, Bowman Gray School of Medicine, Winston-Salem, NC (C.D.F., M.B.M.), Department of Medicine, University of California, Irvine, Orange, CA (J.M.G.), Department of Biostatistics, University of Washington, Seattle, WA (R.A.K.), Department of Medicine, University of Pittsburgh, Pittsburgh, PA (L.H.K., A.N.), Division of Epidemiology and Clinical Applications, National Heart, Lung, and Blood Institute, Bethesda, MD (T.A.M.), Department of Radiology, Brigham and Women’s Hospital, Boston, MA (D.H.O.), Departments of Medicine and Epidemiology, University of Washington, Seattle, WA (B.P.), Heart Disease Research Centre, Dalhousie University, Halifax, Nova Scotia (P.R.), Department of Pathology, University of Vermont, Burlington, VT (R.P.T.). Address reprint requests to: Richard A. Kronmal, PhD, CHS Coordinating Center, Room 530, 1107 N.E. 45th Street, Seattle, WA 98105. 0 Published 1991 by Elsevier Science Publishing Co., Inc.

1047/2797/91/$00.00

264

Fried et al. THE CARDIOVASCULAR

HEALTH STUDY: DESIGN AND RATIONALE

AEP Vol. I, No. 3 February 1991: 263-276

conventional risk factors such as hypertension, total serum cholesterol, and cigarette smoking appear to weaken with advancing age (l-3). Other studies have even suggested a protective effect in the elderly of risk factors identified in younger populations (4-S). These conflicting observations raise the possibility that selective survival may significantly alter the relative importance of CVD risk factors in predicting the outcome of the disease, or even its incidence. Studies of risk factors for coronary heart disease and stroke in older adults (defined here as those aged 65 years and older) are needed for at least four reasons: 1. A major portion of morbidity and mortality from cardiovascular disease occurs in this age group (6). 2. The absolute number of persons in this age group is growing rapidly [7]. 3. There is a paucity of information regarding risk factors for CHD and stroke in this age group (8). 4. The clinical characteristics and course of cardiovascular diseases may vary with advancing age (9). Further, the reported high prevalence of asymptomatic and often extensive atherosclerosis at older ages (10) may make determination of overt disease as a risk factor for development of clinical disease more important in this age group than the identification of risk factors for atherosclerosis per se. Epidemiologic studies conducted in younger population groups have often focused on “conventional” risk factors, such as high blood pressure, abnormal lipids, and cigarette smoking. These risk factors are then related to the development and progression of atherosclerotic lesions. When atherosclerotic lesions are already present, however, other factors, such as those leading directly to the interruption of blood flow to vital organs, may be more important in the onset of clinical disease than these “conventional” risk factors. Such “hypothesized” risk factors, or “precipitants” of disease, include increased thrombotic tendencies (or decreased ability to lyse clots), increased vasomotor tone, inability to dilate the arterial wall contiguous to a plaque, decreased formation of collateral vessels, or impaired functional reserve due to either preceding or coexistent disease. In addition, factors that affect our perception of health and illness may significantly modify the presentation of illness in individuals with minimal or atypical symptoms. Major life changes associated with aging, such as serious illness or death of a spouse, changes in daily activity or social interaction associated with noncardiovascular illness, or loss of ability to function independently, may affect the clinical manifestations and the course of cardiovascular diseases as well. The Cardiovascular Health Study (CHS) is an observational, population-based longitudinal study of risk factors for coronary heart disease and stroke in adults aged 65 years and older. The main objective of the study is to assess these factors and identify those related to the onset and course of coronary heart disease and stroke in this age group. CHS is designed also to assess the associations of risk factors with pathologic changes known as “subclinical disease,” changes which have not produced recognizable clinical signs of disease. These subclinical changes may be the precursors of clinical disease, or the substrate upon which precipitants of clinical disease act. These changes can be detected through noninvasive measures. In CHS, sonography is utilized to detect carotid atherosclerosis; echocardiography is used to detect increased left ventricular mass and impaired cardiac systolic and diastolic function; electrocardiographic tracings are used to detect cardiac arrhythmias and silent myocardial ischemia; and spirometry will detect mild airway obstruction and pulmonary congestion secondary to congestive heart failure with respiratory muscle weakness measured by maximal inspiratory pressure. In contrast to subclinical disease, “clinical cardiovascular disease” is defined in CHS to denote symptomatic or otherwise overt illness, including angina pectoris, myocardial infarction, congestive heart failure, stroke, transient ischemic attack, claudication, and disability.

AEP Vol. 1, No. 3 Februq 1991: 263-276

Fried et al. THE CARDIOVASCULAR HEALTH STUDY: DESIGN AND RATIONALE

265

In this communication we describe the objectives and the design of the Cardiovascular Health Study. Details on specific methods are available in the CHS manuals of operation (11).

OBJECTIVES

OF CHS The objectives

of CHS are:

1. To quantify associations of conventional and hypothesized risk factors with CHD and stroke. 2. To assess the association of indicators of subclinical disease, identified by noninvasive measures such as carotid ultrasonography and echocardiography, with the incidence of CHD and stroke. 3. To quantify the association of conventional and hypothesized risk factors with subclinical disease. 4. To characterize the natural history of CHD and stroke, and identify factors associated with clinical course. 5. To describe the prevalence and distributions of risk factors, subclinical disease, and clinically diagnosed CHD and stroke.

DESIGN

AND

METHODS

Study Communities

and Centers

The CHS cohort was recruited from four U.S. communities: Forsyth County, North Carolina; Sacramento County, California, Washington County, Maryland; and Pittsburgh (Allegheny County), Pennsylvania. The Pittsburgh Field Center population was entirely urban; and the other three Field Centers recruited a mixture of urban and rural populations. The CHS design required each Field Center to recruit and examine not less than 1250 men and women aged 65 years and older during the first year beginning June 12, 1989, for a total of at least 5000 study-wide. This cohort will be followed for a period of three years with semiannual contacts, alternating between brief telephone calls (at 6, 18, and 30 months after entry) and interim clinic visits (at 1 and 2 years after entry). Three years after entry, the extensive physical and laboratory examinations performed at baseline will be repeated.

Sample Size and Power The recruitment goal of approximately 5000 participants is based on the following considerations: 1. Sufficient sample size to provide enough new CVD events in three years to permit assessment of risk factors associated with the incidence of CVD. 2. Adequate numbers of women and men to permit assessment of risk factors in both sexes. 3. Adequate numbers in each of four age groups (65-69, 70-74, 75-79, 80+) to permit assessment of risk factors in different age groups. This cohort was expected to have a CVD prevalence rate of 25%, yielding approximately 1250 participants with CVD at entry. The CVD incidence rate was expected to be 56 per 1000 per year, generating approximately 840 new CVD events in a three-year period. Power was estimated to be greater than 90% at three years for assessing associations of CVD events with sex, systolic blood pressure, and total cholesterol, with (Y = 0.05 (twotailed), assuming an incidence rate of 10%. Power was estimated to be greater than

266

Fried et al. THE CARDIOVASCULAR

HEALTH STUDY: DESIGN AND RATIONALE

TABLE 1 Size and age distribution frames for CHS

Age group (yrs)

Pittsburgh’ County n (%)

65-69 70-74 75-79

1,432 1,295 993

(29) (26) (20)

80+ TOTAL 65 +

1,240 (25) 4,960 (100)

AEP Vol. I, No. 3 February 1991: 263-276

of the Health Care Financing Administration Forsyth County n (%) 1,486 1,437 953

(30) (29) (19)

1,116 (22) 4,992 (100)

Sacramento County n (%) 1,330 1,407 969

(27) (29) (20)

1,145 (24) 4,851 (100)

sampling

Washington County n 1,070 1,291 1,113

(%) (20) (25) (21)

1,770 (34) 5,244 (100)

’ Voter Wards 4, 7, 14. 15 only.

80% at three years for assessing associations of CVD with left ventricular hypertrophy evaluated by electrocardiogram, and with fasting glucose, weight, and cigarette smoking, assuming a CVD incidence rate of 20%.

Sampling Each community sample was obtained by a two-step process from the Medicare eligibility lists of the Health Care Financing Administration (HCFA). HCFA estimates its eligibility file to be 98-99% complete for U.S. citizens aged 65 and older. These lists, ordered by Medicare claim number (Social Security number), were sampled systematically from a random start to produce sampling frames of approximately 5000 potential participants in each community. Use of the HCFA lists allows us to compare particie pants and non-participants on key demographic factors, and linkage with Medicare and vital statistics files for comparisons on hospitalizations and mortality. The size and age distributions of these four sampling frames were similar, except for a lower proportion of candidates aged 65-69 years, and a higher proportion of those aged 80-t in the Washington County sampling frame (Table 1). Random samples were selected from these sampling frames by a customized program designed to produce a cohort with a 60 : 40 female/male ratio in each of four age groups, with the following distribution within age strata: 35% aged 65-69, 25% aged 70-74, 20% aged 75-79, and 20% aged 80 years and older. These proportions were selected to produce approximately equal numbers of events in each age and sex stratum. Those eligible to participate included all persons living in the household of each individual sampled from the HCFA sampling frame, who were 65 years or older at the time of examination, were noninstitutionalized, were expected to remain in the area for the next three years, and were able to give informed consent and did not require a proxy respondent at baseline. Potentially eligible persons who were wheelchair-bound in the home at baseline or were receiving hospice treatment, radiation therapy or chemotherapy for cancer were excluded.

Recruitment

and Initial Examination

The baseline examination consisted of a home interview and clinic examination (Table 2). Limited data on demographics, illnesses, and physical function were collected in the home on all persons aged 65 years and older, to permit comparison of eligible and ineligible potential participants. Eligible participants giving informed consent answered standard questionnaires

Fried et al. THE CARDIOVASCULAR HEALTH STUDY: DESIGN AND RATIONALE

AEP Vol. 1, No. 3 February 1991: 263-276

TABLE 2 Components during follow-up

of the baseline examination

267

and planned repeat administrations Planned repeat administration

Component PSYCHOSOCIAL FACTORS of life (13)

Perceived health/quality Social networks

annually annually

Social support (14)

annually

(15)

Life events

(16)

semi-annually

Depression

(17)

annually

MEDICAL

CARE

Medications

annually

Medical history (21, 22)

annually

Hospitalizations

semi-annually

BLOOD

PRESSURE

Resting

annually

blood pressure

Orthostatic

blood pressure

three-year

examination

Ankle-arm

index

three-year

examination

three-year

examination

PHYSICAL

EXAMINATION

ANTHROPOMETRY PHYSICAL

annually

(24)

ACTIVITY

(18,

FUNCTIONAL

STATUS

NEUROLOGIC

FUNCTION

Cognitive

function

Transient

neurologic

DIETARY

three-year

19)

(20)

annually

(25, 26)

INTAKE

symptoms

(26)

examination

annually

(27)

three-year

examination

three-year

examination

ELECTROCARDIOGRAPHY Resting

ECG

Ambulatory SPIROMETRY

annually

(29) ECG

(30)

(31-33)

three-year

examination

three-year

examination

LABORATORY three-year

examination

three-year

examination

Plasma lipids (35)

three-year

examination

Hemostatic

three-year

examination

three-year

examination

three-year

examination

Serum chemistries Oral glucose tolerance

test (34)

factors (36)

ULTRASONOGRAPHY ECHOCARDIOGRAPHY

(37) (38)

assessing quality of life (12), social support (13), social networks (14), stressful life events (15), and personal habits. Information on prescription medication used in the preceding two weeks was collected directly from prescription bottles, and use of nonprescription drugs such as aspirin, sleeping pills and antihistamines was ascertained by questionnaire. Depression was assessed using the Center for Epidemiologic Studies Depression Scale (16), and the extent of physical activity was determined with modified Minnesota Leisure Time Activities (17) and Paffenbarger (18) questionnaires. Physical functioning was assessed using a modified version of the Health

268

Fried et al. THE CARDIOVASCULAR HEALTH STUDY: DESIGN AND RATIONALE

AEP Vol. I, No. 3 February 1991: 263-276

Interview Survey Supplement on Aging questionnaire (19), which assessed activities of daily living (ADL’s) and instrumental ADL’s. Medical history information included recent hospitalizations, the Rose questionnaires for angina pectoris and claudication (20), and the American Thoracic Society questionnaire on pulmonary symptoms (21). The duration of the home interview was approximately 90 minutes. Participants were asked to fast for 12 hours before coming to their clinic appointment; all examinations were scheduled in the morning. Sitting blood pressure, resting electrocardiography, and venipuncture were performed early in the examination. Sitting blood pressure was measured in the right arm of seated participants after a fiveminute rest using an appropriately-sized cuff and a Hawksley random zero sphygmomanometer, model 7076 (Hawksley and Sons Limited, Sussex, England). The average of two measurements of the first (systolic) and fifth (diastolic) Korotkoff sounds was used for analysis. Blood pressure and heart rate were also measured in the suprine position after a 30-minute rest (immediately following the echocardiographic examination) with a standard mercury sphygmomanometer (W.A. Baum Co., Inc., Copiague, NY), and repeated after three minutes of standing, to assess postural changes. Duplicate measurements of supine blood pressure in the right arm and both ankles were performed with a standard mercury sphygmomanometer and an 8 MHz Doppler probe attached to a double-headed stethoscope. These are used to assess arterial occlusive disease. A brief physical examination included assessment of pitting ankle edema, auscultation of the heart and lungs, and auscultation of the carotid arteries for bruits. Measures of physical performance included triplicate measurements of grip strength in each hand using a hand-held, Jamar dynamometer (Asimow Engineering Co., Los Angeles, Ca); a timed, 15foot walk at normal pace; and repeated chair stands from a sitting position with arms folded (22). Anthropometric measurements included weight, sitting and standing height, waist and hip circumferences, and heel-knee length. Heel-knee length, which can be used to estimate standing height (23), was included because not all participants were able to assume a completely upright posture. Bioelectric impedance was measured with a TVI-10 Body Composition Analyzer (Danninger Medical, Columbus, OH) as an estimate of body fat. Cognitive function was measured by the mini-mental state examination (24) and the digit-symbol substitution test (25). T ransient neurologic symptoms were assessed by questionnaire (26). Usual dietary intake was assessed using a modified NC1 (Block) food frequency questionnaire (27), administered via picture cards of foods sorted by frequency of use, and providing a qualitative assessment of intake. Supplementary information was collected about alcohol consumption, food-preparation practices, and use of prescribed diets, low-salt and low-fat foods. Twelve-lead resting electrocardiograms (ECGs) were obtained on all participants using the MAC PC-DT ECG recorder (Marquette Electronics Inc., Milwaukee, WI). Paper copies were produced for review at the Field Center. Electrocardiographic data were stored electronically and transmitted daily to the ECG Reading Center for analysis using the Novacoder ECG measurement and classification system (28). Twenty-four hour ambulatory ECGs were performed on a random subset of 1600 participants, using a Dynacord Model 420 Cassette Holter Recorder (Del Mar Avionics, Irvine, CA) to monitor leads V2 and V5. Recordings were analyzed for arrhythmic events and ischemic episodes by an automated analysis program (29). The forced vital capacity (FVC) and forced expiratory volume in one second

AEP Vol. 1, No. 3 Febnmy 1991: 263-276

Fried et al. THE CARDIOVASCULAR HEALTH STUDY: DESIGN AND RATIONALE

269

(FEVl) were measured with a water-sealed, Collins Survey II spirometer (WE Collins, Braintree, MA). An automated analysis program (PulmoScreen II, S&M Software Co., Doylestown, PA) was used to assess quality and reproducibility of effort after each maneuver, according to the standards of the American Thoracic Society (30), with predicted values from Crapo and colleagues (31). Methods and predicted values of Black and Hyatt (32) were used for maximal inspiratory pressure. Venipuncture was performed early in the clinic visit, after 12-hour fast. Citrated samples were sent for complete blood count (not including differential white cell count) at laboratories located close to the Field Centers. Multiple aliquots of plasma or serum were prepared and frozen at - 70°C and were shipped weekly on dry ice to the Central Blood Analysis Laboratory. Fasting serum chemistry analyses were performed on the Kodak Ektachem 700 Analyzer (Eastman Kodak Corp., Rochester, NY) and included measurement of potassium, creatinine, uric acid, albumin, and glucose. Serum insulin was measured by solid-phase radioimmunoassay, using serum-based standards (Diagnostics Products Corp., Los Angeles, CA). All participants except diabetics treated with insulin or oral hypoglycemic agents drank a 75 gram oral glucose load, and repeat venipuncture was performed two hours later for measurement of postchallenge serum glucose and insulin levels (33). Fasting plasma lipid analyses were performed on an Olympus Demand system (Olympus Corp., Lake Success, NY) and included measurement of total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides, standardized according to the Centers for Disease Control. Low density lipoprotein (LDL) cholesterol was calculated according to the equation by Friedewald and associates (34). Plasma fibrinogen levels were measured with a clot-based endpoint, using a BBL fibrometer (Becton-Dickinson, Cockeysville, MD) using a modification of the von Clauss method (35). Assays for factors VII and VIII were performed using factor-deficient plasma on a General Diagnostics Coag-A-Mate X2 (Organon Teknika Corp., Durham, NC), and standardized against World Health Organization reference materials. Also, a serum and plasma repository has been established to allow future studies. Duplex ultrasonography of the carotid arteries was performed with a Toshiba SSA-270A (Toshiba American Medical Systems, Tustin, CA) equipped with a 5.0 MHz transducer. Two-dimensional brightness mode (B-mode) imaging was used to detect thickening of the arterial wall, disruption of normal wall interfaces, and development of focal plaques bilaterally. Images obtained included a single lateral view of the distal common carotid artery showing the near and far walls, and three views of the carotid bulb or proximal internal carotid artery centered on the site of maximum wall thickening. Single Doppler pulsed wave, continuous wave, and color Doppler images were recorded at the site of maximum acceleration of blood flow. Images were stored on super VHS videotape and on optical disc, and were interpreted at the CHS Ultrasound Reading Center. Analytic measurements included average wall thickness, discrete maximal wall thickness, minimal residual lumen, Doppler velocities, an image quality score, a lesion morphology score, and a lesion surface characteristic score. Quality of data was assessed through periodic duplicate studies to assess intra- and inter-technician reproducibility of Field Center and Reading Center technicians. M-mode, two-dimensional, and Doppler echocardiographic examinations were performed with the Toshiba SSH-160A (Toshiba Medical Systems, Tustin, CA), equipped with 2.5 MHz and 3.75 MHz transducers. Imaging was performed with the highest frequency transducer that provided adequate penetration. Doppler recordings

270

Fried et al. THE CARDIOVASCULAR

HEALTH STUDY: DESIGN AND RATIONALE

AEP Vol. 1, No. 3 Febnmy 1991: 263-276

were generally performed using the 2.5 MHz transducer. Images obtained included parastemal long and short axis, apical four-chamber and two-chamber views. Images were recorded on super VHS video tape, and were interpreted at the CHS Echocardiography Reading Center using a Dextra D-200 Image Analysis System (Dextra Medical Systems, Inc., Lakewood, CA). Analytic measurements included M-mode left ventricular wall thickness and dimensions; left atria1 and aortic root dimensions; twodimensional left ventricular volumes and wall motion; and Doppler mitral flow velocity. From these measurements, estimates of left ventricular mass, ejection fraction, stroke volume, end-systolic stress, regional and segmental wall motion, and percent fractional shortening were calculated. Quality of data was assessed through periodic duplicate studies to assess intra- and inter-technician reproducibility of Field Center and Reading Center technicians, as well as review and consensus sessions among echocardiography readers. At the end of the clinic visit, an exit interview was held with each participant to review important findings, answer questions, and provide medical referral to those who needed it. Criteria for notification and referral were established during protocol development, and were limited to urgent situations such as unstable angina, worsening congestive failure, uncontrolled hypertension (greater than 220 mmHg systolic or 115 mmHg diastolic), or new onset transient ischemic attacks. Participants and their physicians received a complete report of examination findings. The duration of the clinic examination was approximately four to five hours, with frequent breaks for rest and snacks.

Data Management and Quality Control for the Baseline Examination CHS Field Centers used a network of microcomputers with data-checking capabilities for entry and editing of data. Data were collected by a combination of paper-first forms (primarily used during the home visit), and direct (on-line) data entry. All paper-first forms were entered in duplicate. Data entry was completed before participants left the clinic, permitting generation of an exit summary for each participant of information collected during the examination. Data were transmitted weekly to the Coordinating Center. Duplicate measurements for quality control purposes were kept to a minimum, in order to decrease burden on the participants. Masked duplicate blood specimens were drawn on 5% of participants to determine reproducibility within the Central Blood Analysis Laboratory. Repeat echocardiography and ultrasound examinations were performed on a subset of 20 participants per Field Center to assess intra- and intertechnician variability. Masked duplicate readings were performed in 5% of ultrasound examinations and 10% of echocardiography examinations to assess intra- and interreader variability. Masked duplicate resting ECGs were performed on 20 participants per Field Center, and masked duplicate readings of ambulatory ECGs were performed on 25 participants. Reproducibility data on repeat pulmonary function maneuvers were transmitted and reviewed weekly at the Pulmonary Function Reading center. Key variables from each procedure were analyzed monthly for differences by technician, Field Center or time. Significant differences were referred to a study-wide Quality Control Committee for immediate investigation and remedial action as necessary.

AEP Vol. 1, No. 3 February 1991: 263-276

THE CARDIOVASCULAR

HEALTH

Fried et al. STUDY: DESIGN AND RATIONALE

271

Interim Follow-up Between the two extensive clinical examinations, five semi-annual contacts are scheduled in order to ascertain interim cardiovascular disease events, assess changes in major risk factors and functional status, and maintain contact with participants. These contacts consist of alternating telephone calls and clinic visits. The three interim telephone calls, of approximately ten minutes’ duration, serve primarily to identify hospitalizations and major life events in the preceding six months. The two interim clinic visits, of approximately 90 minutes duration, serve to collect more complete information on medical diagnoses, medications, life events, social support and networks, and physical and cognitive functioning. Sitting blood pressure and weight are measured and the resting ECG will be repeated at these visits. The schedule for followup measurements of components of the baseline examination is given in Table 2. Between contacts, participants and their physicians are encouraged to report all hospitalizations and major illnesses to the CHS Field Center. At entry, each participant was given a wallet-sized card with the telephone number and address of the Field Center, and individuals close to the participants (relative, friend) were asked to contact the Field Center in the event of hospitalization or illness. CHS also performs periodic searches of the HCFA Medicare Utilization (MEDPAR) files, to identify hospitalizations of cohort members that may have been missed by other methods. Discharge coding of all hospitalizations identified by these methods will be reviewed, and detailed data will be abstracted for potential cardiovascular diagnoses (Table 3).

Classification

of Cardiovascular Events

The major cardiovascular events in CHS are myocardial infarction, angina pectoris, congestive heart failure, peripheral arterial disease, stroke, and transient ischemic attack (Table 4). The CHS definition of myocardial infarction is identical to that used in the Atherosclerosis Risk in Communities (ARIC) Study (36). Criteria for angina pectoris, congestive heart failure and peripheral arterial disease rely on physician diagnosis, but also include information from diagnostic and surgical procedures if these were performed. Criteria for stroke and transient ischemic attack are similar to those used in the Systolic Hypertension in the Elderly Program (SHEP) (37). Provisional diagnoses are assigned by the Field Center principal investigator and reviewed by the Coordinating Center. Provisional diagnoses not related to cardiovascular disease or stroke are accepted without further review. Provisional diagnoses of cardiovascular disease or stroke are reviewed and adjudicated at periodic meetings of a study-wide Morbidity Review Committee. All fatal events are reviewed by a study-wide Mortality Review Committee, and

TABLE 3 Code 250 401-405 410-414.9 424-425 427-429.9 430-438 440.2-443.9

ICD-9CM

codes investigated

for possible incident

events Disease

Diabetes mellitus Hypertensive heart disease Ischemic heart disease Cardiomyopathy and valvular heart disease Cardiac dysrhythmias, heart failure (cause unknown) Cerebrovascular disease Claudication and peripheral vascular disease

272

Fried et al. THE CARDIOVASCULAR

TABLE 4 Myocardial Angina

HEALTH

Definition Infarction

Pectoris

Congestive

Peripheral

Heart Failure

Arterial

Disease

Stroke

Transient

Ischemic Attack

STUDY: DESIGN AND RATIONALE

of major

cardiovascular

AEP Vol. I, No. 3 Februnry 1991: 263-276

events

Evolving Q-wave MI; or cardiac pain plus abnormal enzymes and either an evolving ST-T pattern or new left bundle branch block. Coronary artery bypass graft surgery; or 2 70% obstruction of any coronary artery; or ST depression > 1 mm on exercise testing plus positive Rose questionnaire; OTangina diagnosed by physician plus receiving medical treatment for angina (nitrates, beta-blockers or calcium-channel blockers). Cardiomegaly and pulmonary edema on chest X-ray; or dilated ventricle and wail-motion abnormalities by echocardiography or contrast ventriculography; or congestive failure diagnosed by physician, plus receiving medical treatment (diuretic plus either digitalis, vasodilator or angiotensin converting enzyme inhibitor). Ultrasonographically or angiographically demonstrated obstruction or ulcerated plaque; UTabsence of doppler pulse in any major vessel; or positive exercise test for claudication; or bypass surgery, angioplasty or thrombolysis for peripheral arterial disease; or exertional leg pain relieved by rest plus either claudication diagnosed by physician or ankle-arm systolic ratio of 5 0.8. Abrupt onset of new neurologic deficit lasting at least 24 hours, with specific locahzing findings confirmed by unequivocal physical examination or laboratory data and without evidence for underlying nonvascular cause. Rapid onset of focal neurologic deficit lasting less than 24 hours, assessed to be due to ischemia, without evidence for underlying nonvascular cause.

classified by specific cause of death. Information for classification of death are obtained from death certificates, autopsy and coroners’ forms (if available), hospital records, and interviews with attending physicians, next-of-kin, and witnesses. Because the CHS design does not exclude those with cardiovascular disease at entry, participants were classified, at the end of the baseline examination, as to the presence or absence of the six major cardiovascular diseases. This distinction is important because risk factors for new disease may differ from those for existing or recurrent disease. In add;*ion, recurrent events are less likely to be misclassified than new events, so data collection efforts were designed to maximize information obtained Confirmation of a participant’s about occurrence of new disease (i.e., incidence). report of prevalent myocardial infarction, angina pectoris, congestive failure, and peripheral arterial disease are sought from objective information collected at entry and from the review of hospital and physicians’ records. Stroke and transient ischemic attack reported at entry are not confirmed further. Criteria for confirmation of prevalent disease are presented in Table 5.

CONCLUSION Although the mortality rate from cardiovascular diseases is declining, the combined effects of population aging and improved survival after CVD events in younger age groups are expected to lead to a more than 40% increase in the incidence, prevalence, and cost of coronary heart disease by the year 2010 (38). The majority of this increase

AEP Vol. 1, No. 3 February 1991: 263-276

THE CARDIOVASCULAR

TABLE

5

Criteria for confirmation

Myocardial Infarction

Angina

Pectoris

Congestive

Peripheral

Heart Failure

Arterial

’ Confirmation

Disease

of reported

HEALTH

Fried et al. STUDY: DESIGN AND RATIONALE

of prevalent

273

disease at baseline*

Old MI on ECG (Minnesota codes l-1-1 through l-2-5 plus l-2-1 (43)); OT segmental wall-motion abnormality on echocardiography; OThospital discharge or physician diagnosis of MI. Use of nitroglycerin or nitrates; ur use of beta-blocker or calcium-channel blocker plus no history of hypertension; OThistory of coronary bypass surgery or angioplasty; OThospital discharge or physician diagnosis of angina. Use of diuretic plus either digitalis or vasodilator; or dilated ventricle plus wall-motion abnormality plus decreased systolic function on echocardiography; or hospital discharge or physician diagnosis of congestive failure Ankle-arm systolic ratio 5 0.8; or history of bypass surgery or angioplasty for peripheral arterial disease; or absence of lower limb; OThospital or physician diagnosis of peripheral arterial disease stroke

or transient

ischemic

attack

(TIA)

at entry

was not

sought

will occur in those aged 65 and older; this is, ironically, the group for which risk factor information is least complete and consistent. The extensive examinations performed in CHS will permit evaluation of a variety of conventional and hypothesized CVD risk factors in older adults, particularly in groups previously under-represented in epidemiologic studies, such as women and the very old. Emphasis will be placed on absolute, as well as relative, numbers of excess events associated with given risk factors (i.e., attributable risk vs. relative risk) (39). Inclusion of measures of subclinical disease will allow more precise classification into groups with and without cardiovascular disease, a distinction that is especially important in a cohort expected to have a high rate of underlying atherosclerosis. In addition, use of subclinical disease measures will permit evaluation of these measures as possible intermediate end points, for use in future clinical trials and intervention studies in this age group. Measurement of risk factors believed to be important initiators of the atherosclerotic process (such as lipids and blood pressure) and possible precipitants of overt events (such as hemostatic factors and adverse life events) will allow assessment of either or both in relation to the development, progression and course of CVD in this age group. Finally, identification of possible protective factors, associated with preservation of a vigorous and active life style at advanced stages, may hold promise for extending healthy and independent living beyond the age at which it is commonly assumed to be possible. The results of CHS are expected to help address these important epidemiologic and public health issues in anticipation of the dramatic increase in the number of persons at risk for, and suffering from, coronary heart disease and stroke.

REFERENCES 1. Kannel WB, Vokonas PS. Primary risk factors for coronary heart disease in the elderly: The Framingham Study. In: Wenger NK, Furberg D, Pitt E eds. Coronary Heart Disease in the Elderly. New York: Elsevier; 1986:60-95. 2. Rose G, Shipley M. Plasma cholesterol concentration and death from coronary heart disease: 10 year results of the Whitehall study. Br Med J. 1986;293:306-7. 3. Seltzer CC. Smoking and coronary heart disease in the elderly. Am J Med Sci. 1975;269:309-15.

274

Fried et al. THE CARDIOVASCULAR

4.

Langer RD, Ganiats TG,

high blood pressure. 5.

AEP Vol. 1, No. 3 Febmaq 1991: 263-276

HEALTH STUDY: DESIGN AND RATIONALE

Mattila

Barrett-Connor

E. Paradoxical

survival of elderly men with

Br Med J. 1989;298:1356-8.

K, Haavisto

M, Rajala

S, Heikinheimo

R. Blood pressure

and five year

survival in the very old. Br Med J. 1988;296:887-9. 6.

Feinleib

M, Gillum RI: Coronary

problem in the United States. in the Elderly. 7. States

New York: Elsevier;

Spencer

G. U.S.

D.C.:

Government

8.Wenger current Coronary

Printing

NK, Furberg CD,

knowledge

and research

MacDonald

Age Ageing. 10.

of the population

Population

Reports,

Gfiice,

May, 1984.

Pitt E. Coronary

of the United

Series P-25,

No. 952,

heart disease in the elderly: Review

In: Wenger

New York: Elsevier;

JB. Presentation

NK, Furberg CD,

of

Pitt E eds.

1986:1-7.

of acute myocardial infarction

in the elderly-a

review.

1984;13:196.

Gerstenblith

Bierman

Projections

Current

recommendations.

Heart Disease in the Elderly,

9.

of the

1986:60-95.

Bureau of the Census,

by age, sex and race: 1983-2020.

Washington,

heart disease in the elderly: The magnitude

In: Wenger NK, Furberg CD, Pitt E eds. Coronary Heart Disease

G, Weisfeldt

E, Hazard

ML, Lakatta

W eds. Principles

EG. Disorders of the heart.

of Geriatric

Medicine.

New York:

In: Andres R, McGraw

Hill;

1985:104-19. 11.

CHS Collaborative

no. 2, Field Center Center,

University

12. naires.

I, Newell

Health

Psychology.

14.

CA,

JD-30,

no. 4, Quality

Control.

No.

CHS

1, protocol; Coordinating

1107 N.E. 45th St., Suite 530, Seattle, health:

WA 98105.

A guide to rating scales and question-

Press; 1987:296-8.

Kaplan RM.

Assessment

of methods for measuring

social support.

1988;7:75-109.

Lubben JE. Assessing

Maint.

CHS Manuals of Operation:

C. Measuring

New York: Oxford University Heitsmann

Group.

no. 3, Events;

of Washington,

McDowell

13.

social networks

among elderly populations.

J Health

Promo

1988; 11:42-52.

15.

Wells JA. Chronic

Variables

in Epidemiologic

Department 16.

of Health

life situations

and life change events.

Studies of Cardiovascular

and Human Services.

17.

studies depression

Taylor

HL, Jacobs

leisure time physical 18. 19.

alumni.

National

Center

20.

Office,

21.

of the center

for

1986;42:28-33.

B, et al. A questionnaire

for the assessment

of

Dis. 1978;31:741-55. 1978;108:161-75.

for Health Statistics,

JE Fitti and MG Kovar. The Supplement

Health Interview Survey. Vital and Health Statistics, 87-1323.

Oct.

Public Health

Service,

Washington,

on

Series 1, No.

D.C.:

U.S.

Gov-

1987.

Rose G, McCartney

pain and intermittent

validity

U.S.

1985.

RS Jr, Wing AL, Hyde RT. Physical activity as an index of heart attack

Pub. No. (PHS)

Printing

and indiscriminate

Psychosocial

of a Workshop.

No. 85-2770,

scale. J Clin Psychol.

Schucker

J Chronic

Am J Epidemiol.

Aging to the 1984 National 21, DHHS

DR,

activities.

Paffenbarger

risk in college

(CES-D)

In: Measuring

Disease: Proceedings

NIH Publication

Orme J, Reis J, Herz E. Factorial

epidemiological

ernment

Research

Procedures;

P, Reid DD. Self-administration

claudication.

of a questionnaire

on chest

Brit J Prev Sot Med. 1977;31:42-8.

Ferris BG Jr. Epidemiology

standardization

project.

Am Rev Respir Dis. 1978;118(6,

Part 2): l-88. 22. Aging.

National

Institute

on Aging.

Manual of Operation,

23.

Chumlea

WC,

Established

1988. Jack Guralnick,

Roche AF,

Steinbaugh

persons 60 to 90 years of age. J Am Geriatr 24.

Folstein

MF, Folstein

grading the cognitive 25.

Salthouse,

performance. 26.

Wilkinson

Conference

M.D.,

for Epidemiologic

Bethesda,

ML. Estimating

WE,

Studies

of

MD 20892.

stature from knee height for

1985;33:116-20.

PR. Mini-Mental

for the clinician.

State:

J Psychiat Res.

PA. The role of memory in the age decline

J Gerontol.

toms: A predictor

SE, McHugh

state of patients

Sot.

Populations

A practical

method for

1975;12:189.

in digit-symbol

substitution

1978;33:232-8. Heyman A, Pfeffer RI, Burch JG. A questionnaire

of subsequent

on Cerebrovascular

stroke. Disease.

In: Reivich

for TIA

M, Hurtig HI eds. Thirteenth

New York: Raven Press, 1983:77-92.

symp-

Princeton

AEP Vol. I, No. 3 February 1991: 263-276

THE CARDIOVASCULAR

Fried et al. HEALTH STUDY: DESIGN AND RATIONALE

275

27. Block G, Hartman AM, Dresser CM, Carrel MD, Gannon J, Gardner L. A databased approach to diet questionnaire design and testing. Am J Epidemiol. 1986;124:453-69. 28. Wolf H, MacInnis PJ, Stock S, Helppi RK, Rautaharju PM. The Dalhousie Program: A comprehensive analysis program for rest and exercise electrocardiograms. In: Zywietz C, Schnieder B eds. Computer Application on ECG and VCG Analysis. Amsterdam-London: North Holland Publishing, 1973:231-2. 29. Clark KW, McLear PW, Kortas RG, Mead CN, Thomas LJ. Argus/ZH detection of ST-Segment Changes in Ambulatory ECG Recordings. Proceedings of Computers in Cardiology, Institute of Electrical and Electronics Engineers (IEEE) Computer Society, Long Beach, CA; 1980;27-31. 30. Gardner RM, Hankinson JL, Clausen JL, Crapo RO, Johnson RL, Epler GR. Standardization of spirometry-1987 update. Official statement of the American Thoracic Society. Am Rev Respir Dis. 1987; 136: 1285-98. 3 1. Crapo RO, Morris AH, Gardner RM. Reference spirometric values using techniques and equipment that meet ATS recommendations. Amer Rev Respir Dis. 1981;123:659-664. 32. Black LF, Hyatt RE. Maximal static respiratory pressures: Normal values and relationship to age and sex. Am Rev Respir Dis. 1969;99:696-702. 33. National Diabetes Data Group. Classification and diagnosis of diabetes and other categories of glucose intolerance. Diabetes. 1979;28:1039-57. 34. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499-501. 35. von Clauss A. Gerinnungsphysiologische Schnellmethode zur Bestimmmung des Fibrinogens. Acta Haemat. 1957;17:237-46. 36. ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) Study: Design and objectives. Am J Epidemiol. 1989;129:687-702. 37. Systolic Hypertension in the Elderly Program (SHEP) Cooperative Research Group. Rationale and design of a randomized clinical trial on prevention of stroke in isolated systolic hypertension. J Clin Epidemiol. 1988;41:1197-208. 38. Goldman L, Cook EF. Reasons for the decline in coronary heart disease mortality: Medical interventions versus life-style changes. In: Higgins MW, Luepker RV eds. Trends in Coronary Heart Disease Mortality. New York: Oxford University Press; 1988:67-75. 39. Lilienfeld AM, Lilienfeld DE. Foundations of Epidemiology, second edition. New York: Oxford University Press; 1980:217-8.

APPENDIX Participating

Institutions

and Principal

Staff

Field Centers-Forsyth County, NC: Bowman Gray School of Medicine of Wake Forest University-Drs. Curt D. Furberg (Principal Investigator), M. Eugene Bond, Walter Ettinger, Gerard0 Heiss, Sidney Klopfenstein, Mary Lyles, Maurice Mittelmark, Grethe S. Tell, James F. Toole and Marie Cody, Glenda Garner, Gale Cruise; Sacramento County, CA: University of California, Davis-Drs. Nemat 0. Borhani (Principal Investigator), William Bommer, Marshall Lee, Marc B. Schenker, C. J. Tupper, Philip Weiler, and Terry Himmelmann, Frances LaBaw, Jackie Kay, Patricia Borhani; Washington County, MD: The Johns Hopkins University-Drs. Linda P. Fried (Principal Investigator), Trudy L. Bush, George W. Comstock, Pearl S. German, Steven J. Kittner, Shiriki Kumanyika, Thomas R. Price, Robert C. Rock, Roger C. Sanders, Sheila Sheth, Moyses Szklo, Bernard Tabatznik, Melvyn S. Tockman and Joel G. Hill, Joyce B. Chabot; Allegheny County, PA: University of Pittsburgh-Drs. Lewis H. Kuller (Principal Investigator), Jane Cauley, Karen Matthews, Anne Newman, Trevor J. Orchard, Glae H. Rutan, Richard Schulz, Vivenne

276

Fried et al. THE CARDIOVASCULAR

HEALTH STUDY: DESIGN AND RATIONALE

AEPVol. I, No. 3 February 1991: 263-276

Smith, Sidney K. Wolfson and Peg Meyer, Bett McLaughlin, Janet Bonk; Echocardiography Reading Center-University of California, Irvine: Drs. Julius M. Gardin (Principal Investigator), Hoda Anton-Culver, Walter L. Henry, Slawomir Lobodzinski, Nathan Wong and Peggy Knoll, Jason Conner; Ultrasound Reading CenterNew England Deaconess Hospital: Drs. Daniel H. O’Leary (Principal Investigator), Joseph F. Polak, Hugh G. Wheeler and Jeffrey Potter; Central Blood Analysis Laboratory--University of Vermont: Drs. Russel P. Tracy (Principal Investigator), Edwin Bovill and Elaine Cornell; Pulmonary Function Reading Center-Mayo Clinic and Foundation: Dr. Paul Enright (Principal Investigator), and Sheila Toogood; Electracardiography Reading Center-University of Alberta: Dr. Pentti Rautaharju (Principal Investigator), Farioa Rautaharju, Patty Montague, and Harry Calhoun; Coordinating Center-University of Washington, Seattle: Drs. Richard A. Kronmal (Principal Investigator), Bruce M. Psaty, David Siscovick, Patricia Wahl and Gwen Glaefke, Bonnie Hermanson; NHLBI Project Office-Drs. Teri A. Manolio (Scientific Project Officer), Diane E. Bild, Elaine M. Eaker, Millicent W. Higgins, Petter J. Savage, A. Richey Sharrett and Dorothy Tyler.