Description of Study and of Collected Data

Back to Baseline Monograph

Draft 3: November 12, 1998

1. INTRODUCTION

1.1 Overview

The Cardiovascular Health Study (CHS) is a population-based, longitudinal study of coronary heart disease and stroke in adults aged 65 years and older, funded by the National Heart, Lung, and Blood Institute. The main objective of the study is to identify factors related to the onset and progression of coronary heart disease and stroke. CHS is designed to determine the importance of conventional cardiovascular disease (CVD) risk factors in older adults, and to identify new risk factors in this age group. There is particular interest in identifying participants with sub-clinical CVD and in investigating relationships between risk factors, subclinical disease, and the development of overt CVD. The study design and rationale have been described (1).

Four Field Centers were selected and asked to enroll 1250 men and women each: Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Pittsburgh, Pennsylvania. A Coordinating Center was established at the University of Washington at Seattle, Washington. Central Reading Centers and Laboratories were established at the following sites: Brigham and Women's Hospital, Boston, Massachusetts (ultrasound); University of California at Irvine (echocardiography); Dalhousie University, Halifax, Nova Scotia (ECG); University of Vermont, Burlington (blood laboratory); Mayo Clinic and Foundation, Rochester, Minnesota (pulmonary function).

The purpose of this Monograph is to provide a general description of CHS participants, the methodology involved in collecting the baseline data, and tabulations of key variables. Data are presented by age, gender, and race categories, without interpretation. Several reports providing more detailed analyses of subsets of the baseline data have been published; a complete bibliography of CHS publications is available on this Web page.

1.2 Recruitment

Participants were sampled from the Health Care Financing Administration's (HCFA) Medicare eligibility lists. Letters were mailed to 11,955 sampled individuals. Persons were excluded if they were institutionalized (including nursing homes), in a hospice program or under active treatment for cancer, could not communicate in the language(s) of the interviewer, were confined to a wheelchair in the home, were cognitively unable to sign an informed consent, or did not expect to remain in the community for three years. 3654 participants were recruited from the HCFA sample. In addition, 1547 other age-eligible persons living in households with sampled individuals were enrolled into the study. The recruitment process has been described in detail (2). The study goal of recruiting 5000 adults aged 65 and older into the study in one year was exceeded; 5201 participants were recruited within the allotted year.

The initial CHS cohort included 4926 Caucasian participants (95%), 244 African-American participants, and 31 participants of other races. In order to achieve better representation of the African-American community, three of the Field Centers were asked to recruit an additional 200 African-American participants each between October 1992 and June 1993. All of the sampling procedures used in recruiting the initial cohort were followed in this additional recruitment phase. A total of 687 African-American participants were recruited during this second recruitment phase.

The data shown below for African-American participants combines data for the 244 participants who were part of the original cohort and the 687 participants added in 1992-93. The data for white participants includes data for the 31 participants of races other than black or white.

Data collected during the baseline exam on the original 244 African-American participants were compared to data collected during 1992-93 on the additional 687 African-American participants in order to see if there were any differences related to protocol changes or technician drift. All of the continuous variables were comparable for the two groups, with the following exceptions:

1. Due to laboratory drift, values for cholesterol were not comparable between 1989 and 1992. The values reported here have been adjusted to account for that difference.

    

2. During the original baseline examination, the 30-point Folstein Mini-Mental examination (3) was administered while in subsequent years, the 100-point Teng Modified Mini-Mental examination (4) was administered. Thus the Mini-Mental examination for the original cohort was based on a 30-point scale, and the Mini-Mental examination for the new cohort was based on a 100-point scale. A regression equation was developed to convert the 100-point score into a comparable 30-point score (unpublished data), and this equation was used to convert the scores from the new cohort into comparable 30-point scores. The Mini-Mental scores reported for the black cohort are a combination of scores from the actual 30-point test (n=244) and scores that were converted from the 100-point test into 30-point equivalents (n=687).

    

3. All of the body size measurements except height were larger for the new black cohort than for the original black cohort. Specifically, weight, waist circumference, hip circumference, BMI, and bioelectric reactance were all larger in the new cohort. Because of the consistency between these findings, measured by different technicians using different methods, this seemed to reflect an actual difference in body size between the two cohorts rather than a change in measurement technique or other protocol-related cause. Thus we combined the body size data for the original and new black cohorts and reported it together.

    

Spirometry and Echocardiography procedures were not completed on the African-American cohort during their baseline exam. Spirometry was performed during the 1993-94 exam, and Echocardiography was performed during the 1994-95 exam. These are considered the Abaseline@ values for these procedures in the African-American cohort, and the values are included in the tables.

1.3 Baseline Examination

The baseline examination was completed in two stages. First, an interviewer went to the person's home and conducted the following interviews:

• Eligibility
• Enumeration
• Informed Consent and Medical Release
• Tracking
• Quality of Life, Social Support, and Social Network
• Stressful Life Events and Depression
• Physical Activity
• Assessment of Physical Functioning
• Medications
• Medical History
• Personal History

If the person was eligible and agreed to participate in CHS, an appointment for a clinic examination was made. In addition, any household members who were age-eligible were invited to participate in the study.

The second stage of the baseline examination was a clinic visit which included the following procedures and interviews:

Procedures

• Resting 12-Lead Electrocardiogram
• 2-hour glucose tolerance test
• Phlebotomy
• Seated blood pressure and heart rate
• Orthostatic blood pressure
• Supine Ankle-Arm blood pressure
• Anthropometry
• Echocardiography
• Carotid Ultrasound
• Spirometry
• Physical Exam
• 24-hour ambulatory electrocardiogram

Interviews

• Reception
• Cognitive function
• Nutrition
• Neurologic History
• Exit

Participants were asked to fast for at least 12 hours before the clinic visit and to refrain from smoking for 30 minutes before the clinic visit. People were considered to be enrolled in CHS if they came in for the clinic visit, regardless of the number of procedures completed.

1.4 Training, Certification, and Quality Control

A central training session was held in May 1989, which was attended by clinic staff from all four Field Centers. During this training session, clinic staff members were certified to do interviews and all clinic procedures except phlebotomy, ultrasound, and echocardiography. Separate training sessions for technicians involved in phlebotomy, ultrasound, and echocardiography were held at the respective reading centers and lab, for technicians who would be performing those procedures. At the central training session, a QC Supervisor was identified for each Field Center (usually the Clinic Coordinator). This person was further certified to train and certify local staff members who were unable to attend the central training session. Quality control procedures have been conducted throughout the examination cycles, including observation of technicians by the QC Supervisor, maintenance of equipment, and analysis of data for outlying values. Blinded QC duplicates were included for blood draws, and replicate scans were performed for ultrasound, ecg, and echocardiography.

2. METHODOLOGY

2.1 Seated Blood pressure and heart rate

Seated blood pressure was measured using a Hawksley Random-Zero Sphygmomanometer, Model 7076 (Hawksley and Sons Limited, Sussex, England). The reported blood pressure is the average of two measurements, which were taken after the participant had been sitting quietly for five minutes. First the technician determined the correct cuff size by measuring the arm circumference at the midpoint between the acromion and the olecranon. After applying the appropriate cuff, the maximum inflation level was determined by inflating the cuff until the radial pulse was no longer felt. The maximum inflation level was then determined to be the pulse obliteration pressure plus 30 mm Hg plus the maximum zero level of the instrument.

Blood pressure was measured by inflating the cuff to the maximum inflation level, waiting 5 seconds, then lowering the mercury by 2-3 mm per second. The first and fifth Korotkoff sounds were recorded. At least 30 seconds elapsed between each cuff inflation.

Heart rate was measured by counting the pulse for 30 seconds on the right arm, with the arm resting on a table.

2.2 Orthostatic blood pressure

The orthostatic blood pressure procedure consisted of two supine blood pressure measurements followed by two standing blood pressure measurements. The measurements were taken at the end of the echocardiography procedure, after the participant had been supine for twenty to thirty minutes, using a standard mercury sphygmomanometer (W.A. Baum Co., Inc., Copiague, NY). The procedures used for obtaining the blood pressure measurements, including determining maximal inflation level and actual measurement of the blood pressure, were the same as those described in Section 2.1 above.

The supine blood pressures were measured first, with the participant continuing to lay quietly on the table. The participant was then assisted to sit on the table for up to one minute, then assisted to stand. The participant remained standing for three minutes before the standing blood pressures were measured. The participant was asked if he/she was feeling dizzy or lightheaded, and if the response was positive the standing procedure was aborted.

After the participant stood for three minutes, the heart rate was measured, followed by two blood pressure measurements as described in Section 2.1 above.

2.3 Supine Ankle-Arm blood pressure

Supine ankle and arm blood pressures were measured on all participants except those with venous stasis ulceration, thrombophlebitis, other pathology that precluded placing a blood pressure cuff around the ankle (e.g., open wounds), or rigid arteries such that an occlusion pressure could not be obtained. A standard mercury sphygmomanometer and an 8 MHz Doppler probe attached to a double-headed stethoscope were used for these measurements.

2.4 Anthropometric measurements

Weight was measured in pounds, with the participant wearing the examination gown (which weighed about one pound) but no shoes.

Standing height was measured with the participant standing with his/her heels against the wall next to a vertical ruler which was mounted on the wall. A metal carpenter's square was placed on top of the participant's head, and his/her height in centimeters was read from the ruler.

Sitting height was measured with the participant sitting up straight on a stool with his/her back against the wall, feet not touching the floor. Sitting height was measured as the distance from the floor to the top of the participant's head, minus the height of the stool.

Heel-to-Knee length was measured with the participant in a supine position, with the leg flexed to a 90 degree angle. One blade of a sliding, broad-blade caliper was slid under the left heel, and the other blade was placed over the anterior surface of the left thigh above the femoral condyles and about 2 inches proximal to the patella.

Waist circumference was measured at the level of the umbilicus, under the examination gown, at mid-respiration.

Hip circumference was measured under the examination gown at the maximal protrusion of the gluteal muscles.

Estimates of body composition were obtained using bioelectric impedance measurements of the body's resistance and reactance. The TVI-10 Body Composition Analyzer (Danninger Medical, Columbus, OH) was used for these measurements.

2.5 Blood chemistry measurements

A fasting blood sample of 40 mL was drawn near the beginning of the clinic exam and processed in the clinic. One tube was sent to a local laboratory for analysis of white blood cells, platelets, and hematocrit. The other six tubes were shipped to the Central Blood Analysis Laboratory at the University of Vermont for analysis.

Oral glucose tolerance tests were obtained on all original cohort participants who had fasted for at least 12 hours, except diabetics treated with insulin or oral hypoglycemics. The glucose drink was administered to the participant as soon as the initial blood draw was completed. The glucose drink was a standard 75 gram glucose load in a flavored drink of approximately 7 ounces. The participant was required to consume the drink in 10 minutes or less, with timing of the 2-hour period beginning as soon as the participant began the drink. The second 5mL blood sample was drawn as close possible to two hours later. The glucose tolerance test was not performed on the African-American cohort until their fourth visit, and those data are not included here.

2.6 Resting 12-Lead Electrocardiogram

Resting 12-lead electrocardiograms (ECGs) were obtained using a MAC PC-DT Personal Cardiograph (Marquette Electronics Inc., Milwaukee, WI). Electrode placement was determined with the participant sitting on the examination table with arms relaxed at the sides. The electrode positions were marked on the skin with a felt pen, then the participant laid down on the table and the electrodes were attached. The positions of electrodes V6 and V4 were determined using a Dal-Square to measure the O-E and O-V6 chest measurements.

2.7 24-hour ambulatory electrocardiogram

Randomly selected participants were asked to wear a Holter monitor, and 24-hour ambulatory electrocardiograms were obtained on a sample of 1432 participants who agreed to this procedure. Dynacord Model 420 Cassette Holter Recorders (Del Mar Avionics, Irvine, CA) were used. At the end of the clinic visit, electrodes were placed on the participant, all wires and cables were attached, and a new battery was placed into the Holter monitor. The equipment was tested, and the participant was given instructions including how to reconnect loose wires or electrodes and what to do at the end of the 24-hour recording session. Recording was then initiated. Details of the procedure have been published. (5,6)

2.8 Echocardiography

Echocardiography was performed to assess various parameters of cardiac structure and function. Cardiac structural parameters that were measured at baseline were left ventricular wall thickness and mass, and left ventricular and left atrial dimensions. Cardiac functional parameters that were measured at baseline were left ventricular percent fractional shortening, left ventricular global and regional wall motion, and mitral flow rates. In addition, the technicians who read the echocardiography tapes made subjective assessments of global left ventricular function and wall motion abnormalities. Echocardiography procedures were performed using the Toshiba SSH-160A digital ultrasound system (Toshiba Medical Systems, Tustin, CA), equipped with both 2.5 MHz and 3.75 MHz transducers. Examinations were performed using the highest frequency transducer that provided adequate penetration. M-mode, two-dimensional, and Doppler measurements were performed, using a series of standard imaging views: Parasternal and Apical Long-Axis view; Parasternal (and in some individuals Subcostal) Short-Axis view; and Apical (and in some individuals, Subcostal) Four-Chamber view. The images were recorded on super VHS video tape, and were analyzed at the Echocardiography Reading Center at the University of California at Irvine using a Dextra D-200 Image Analysis System (Dextra Medical Systems, Inc., Lakewood, CA). (7)

During the fifth follow-up visit (1994-95), the echocardiography procedure was repeated on the original cohort and performed for the first time on the African-American cohort. The same ultrasound machines used at baseline were used again for this exam. The cardiac structures and functions that were measured were the same as those measured at baseline. The images were again recorded on super VHS video tape for central processing, although at this exam the Echocardiography Reading Center was located at Georgetown University in Washington, D.C.

Quality control activities were carried out by both the Field Center sonographers and the Echocardiography Reading Center readers. Sonographer quality control included inital central training and certification, retraining during the examination year, and rescanning a random subset of the participants throughout the year. Reader quality control included rereading a random subset of all tapes and periodic review sessions at the reading center. Additional quality control activities compared the processing of a sub-set of baseline tapes at the University of California, Irvine Reading Center and the Georgetown University Reading Center. No systematic differences in processing or interpreting the tapes were found.

2.9 Carotid Ultrasound

Ultrasound imaging of the carotid arteries was performed using the Toshiba SSA-270A system (Toshiba Medical Systems, Tustin, CA) using a 5 MHz transducer. B-mode imaging was used to analyze the structure of the carotid artery near and far walls at the distal common carotid artery and at anterior, lateral, and posterior oblique angles at the proximal interal carotid artery or carotid bulb. Single Doppler pulse wave, color Dopppler, and Doppler continuous wave images were obtained from the site of maximum flow accelaeration in the distal common carotid artery, bulb, or internal cartid artery. Images were recorded on super VHS video tape and optical disc, and were interpreted at the Ultrasound Reading Center. (8) During the third follow-up exam, ultrasound examinations were repeated on the original cohort and were performed for the first time on the African-American cohort.

Quality control activities were carried out by both the Field Center sonographers and the Ultrasound Reading Center readers. Sonographer quality control included inital central training and certification, retraining during the examination year, and rescanning a random subset of the participants throughout the year. Reader quality control included rereading a random subset of all tapes and periodic review sessions at the reading center. Additional quality control activities assessed the comparability of readings of baseline data performed at the Ultrasound Reading Center during baseline, to readings of a subset of baseline tapes during the third follow-up exam. Due to new software systems and new scoring techniques, some systematic differences were discovered. Because of this, the Ultrasound Reading Center re-read the baseline ultrasound tapes, using the same techiniques used to process the third follow-up tapes. The data displayed in the tables for the original cohort reflects this re-read data.

2.10 Spirometry

The spirometry examination consisted of a short questionnaire which asked about recent exposures known to modify spirometry results (e.g., use of inhalers, beta blockers), followed by a test of pulmonary function using a water sealed, Collins Survey II spirometer (WE Collins, Braintree, MA). A software program (PulmoScreen II, S&M Software Co., Doylestown, PA) automatically analyzed quality and reporducibility of effort after each manuever. Each participant tried to produce three good manuevers. All Spirometry Technicians were trained and certified at a central training session prior to the beginning of the examination cycle. Data quality was monitored at the Pulmonary Function Reading Center using American Thoracic Society spirometry standards (9). Spirometry was repeated on the original cohort and performed for the first time on the African-American cohort during 1993-94 (the fourth follow-up visit for the original cohort, and the first follow-up visit for the African-American cohort).

2.11 Physical Exam

Grip strength was measured using a hand-held Jamar Dynamometer (Asimow Engineering Co., Los Angeles, CA). Three measurements were taken for each hand.

Physical performance was measured by a timed 15-foot walk at normal pace, and repeated chair stands from a sitting position with arms folded (10).

2.12 Medications

Information on prescription medication use in the previous two weeks was collected directly from the medication containers. A computer program developed by CHS was used to match the medication names with NDC numbers and then to group medications into analytic variables (e.g. beta blockers, lipid-lowering medications). (11) Additional questions were asked about the use of certain non-prescription medications including aspirin, sleeping pills, and antihistamines.

2.13 Medical History Interviews

Two questionnaires were used to obtain information about participants' medical histories.

History of myocardial infarction, angina, congestive heart failure, claudication, valve disease, cardiovascular surgeries, and other coronary or vascular ailments were obtained by a self-administered questionnaire. Infomation reported in these questions was validated using medical records to establish baseline disease status. This questionnaire also contained questions on history of lung disease, diabetes, pneumonia, Rose angina and claudication (12), shortness of breath, the American Thoracic Society questionnaire on pulmonary symptoms (13), and symptoms experienced in the previous two weeks.

History of neurologic disease -- history of stroke, TIA, and a variety of neurologic symptoms were obtained using an interviewer-administered questionnaire (14).

2.14 Psycho-Social Interviews

Sociodemographic variables including age, gender, race, marital status, income, education, and occupation were collected as part of the Eligibility Form. The number of persons aged 64 and older who lived in the household was collected on the Enumeration Form.

Social Support was measured using a six-item version of the Interpersonal Support Evaluation List (15), and Social Network size was measured using the ten-item Lubben Social Network Scale (16). Two questions on perceived health and two questions on quality of life were also asked (17).

Depression was assessed using the ten-item Center for Epidemiological Studies Depression (CESD) scale (18). Stressful Life Events were measured using a ten-item scale (19) and included events that had occurred in the previous six months.

2.15 Risk Factor Interviews

A self-administered questionnaire assessed weight history, cigarette smoking (current and former), sleep problems, vision and hearing problems, family history of heart attack and stroke, and women's gynecological history.

Usual dietary intake was measured using a modified NCI (Block) food frequency questionnaire (20), administered using picture cards of foods which the participant sorted by frequency of use. Additional questions were asked about alcohol consumption and intake of certain foods including fats, fruits, vegetables, and food from fast food restaurants.

Physical Activities were measured using adaptations of questionnaires from the Minnesota Leisure Time Activities (21) and Paffenbarger (22). Total kilocalories expended in physical activities were estimated from the amount of time spent in the following activities: walking for exercise, household chores, mowing the lawn, raking the lawn, gardening, hiking, jogging, biking, exercise cycle, dancing, aerobics, bowling, golf, tennis, racquetball, calisthenics/general exercise, swimming.

Physical Functioning was assessed using a modified version of the Health Interview Survey Supplement on Aging questionnaire (23), which assessed activities of daily living (ADL) and instrumental ADL.

2.16 Cognitive Function Interview

Cognitive function was measured using the Mini-Mental State examination (3), and the Digit-Symbol Substitution test (24). The African-American cohort enrolled during 1992-93 was administered the Modified Mini-Mental State examination (4), as described in Section 1.2.

2.17 Exit Summary and Reports

At the conclusion of the clinic examination, the computer generated a summary of the clinic information for the participant to take home, including height, weight, and blood pressures. The computer generated another report showing any "alert findings" that were seen, that is, clinic findings that warranted further attention such as uncontrolled hypertension or ECG abnormalities.

After all data were processed at the Reading Centers and Laboratories, a more complete report of results was sent to the participant as well as to the participant's personal physician.

References

1. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH, Manolio TA, Mittelmark MB, Newman A, O'Leary DH, Psaty BM, Rautaharju P, Tracy RP, Weiler PG. The Cardiovascular Health Study: Design and Rationale. AEP 1991;1:263-276.

2. Tell GS, Fried LP, Lind B, Manolio TA, Newman AB, Borhani NO. Recruitment of Adults 65 Years and Older as Participants in the Cardiovascular Health Study. Ann Epidemiol 1993;3:358-366.

3. Folstein MF, Folstein SE, McHugh PR. Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. J Psychiat Res. 1975;12:189.

4. Teng EL, Chui HC: The Modified Mini-Mental State (3MS) Examination. J Clin Psychiatry 48:314-318, 1987.

5. Manolio TA, Borhani NO, Furberg CD, Gardin JM, Newman A, Rautaharju PM, Siscovick D, Tabatznik B: Cardiac Arrhythmias on 24-hour Ambulatory Electrocardiography in Older Women and Men: The Cardiovascular Health Study. (In Press), American College of Cardiology, October 1993.

6. Rautaharju PM, Manolio TA, Borhani NO, Furberg CD, Gardin JM, Newman A, Siscovick D, Tabatznik B: Ischemic Episodes in 24-Hour Ambulatory Electrocardiograms of Elderly Persons: The Cardiovascular Health Study. Submitted for publication, Circulation, August 1993.

7. Gardin JM, Wong ND, Bommer W, Klopfenstein HS, Smith VE, Tabatznik B, Siscovick D, Lobodzinski S, Anton-Culver H, Manolio TA. Echocardiographic Design of a Multi-center Investigation of Free-living Elderly Subjects: The Cardiovascular Health Study. Journal of the American Society of Echocardiography 1992;5:63-72.

8. O'Leary DH, Polak JF, Wolfson Jr. SK, Bond MG, Bommer W, Sheth S, Psaty BM, Sharrett AR, Manolio TA. Use of Sonography to Evaluate Carotid Atherosclerosis in the Elderly: The Cardiovascular Health Study. Stroke 1991;22:1155-1163.

9. 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.

10. National Institute on Aging. Established Populations for Epidemiologic Studies of Aging. Manual of Operation, 1988. Jack Guralnick, M.S., Bethesda, MD 20892.

11. Psaty BM, Lee M, Savage PJ, Rutan GH, German PS, Lyles M. Assessing the Use of Medications in the Elderly: Methods and Initial Experience in the Cardiovascular Health Study. Journal of Clinical Epidemiology 1992;45:683-692.

12. Rose G, McCartney P, Reid DD. Self-administration of a questionnaire on chest pain and intermittent claudication. Brit J Rev Soc Med. 1977;31:42-8.

13. Ferris BG Jr. Epidemiology standardization project. Am Rev Respir Dis. 1978;118(6, Part 2):1-88.

14. Wilkinson WE, Heyman A, Pfeffer RI, Burch JG. A questionnaire for TIA symptoms: A predictor of subsequent stroke. In: Reivich M, Hurtig HI eds. Thirteenth Princeton Conference on Cerebrovascular Disease. New York: Raven Press, 1983:77-92.

15. Heitsmann CA, Kaplan RM. Assessment of methods for measuring social support. Health Psychology. 1988;7:75-109.

16. Lubben JE. Assessing social networks among elderly populations. J Health Promo Maint. 1988;11:42-52.

17. McDowell I, Newell C. Measuring health: A guide to rating scales and questionnaires. New York: Oxford University Press; 1987:296-8.

18. Orme J, Reis J, Herz E. Factorial and indiscriminate validity of the center for epidemiological studies depression (CES-D) scale. J Clin Psychol. 1986;42:28-33.

19. Wells JA. Chronic life situations and life change events. In: Measuring Psychosocial Variables in Epidemiologic Studies of Cardiovascular Disease: Proceedings of a Workshop. U.S. Department of Health and Human Services. NIH Publication No. 85-2770, 1985.

20. Block G, Hartman AM, Dresser CM, Carrol MD, Gannon J, Garner L. A databased approach to diet questionnaire design and testing. Am J Epidemiol. 1986;124:453-69.

21. Taylor HL, Jacobs DR, Schucker B, et al. A questionnaire for the assessment of leisure time physical activities. J Chronic Dis. 1978;31:741-55.

22. Paffenbarger RS Jr, Wing AL, Hyde RT. Physical activity as an index of heart attack risk in college alumni. Am J Epidemiol. 1978;108:161-75.

23. National Center for Health Statistics, JE Fitti and MG Kovar. The Supplement on Aging to the 1984 National Health Interview Survey. Vital and Health Statistics, Series 1, No. 21, DHHS Pub. No. (PHS) 87-1323. Public Health Service, Washington, D.C.: U.S. Government Printing Office, Oct. 1987.

24. Salthouse, PA. The role of memory in the age decline in digit-symbol substitution performance. J Gerontol. 1978;33:232-8.