Cardiac arrhythmias on 24-h ambulatory electrocardiography in older women and men: the Cardiovascular Health Study.

OBJECTIVES: This study describes the prevalence and correlates of cardiac arrhythmias in older persons. BACKGROUND: Cardiac arrhythmias are frequent in selected samples of elderly persons, but their prevalence and association with cardiovascular disease and its risk factors have not been examined in a large population-based sample. METHODS: In 1,372 participants in the Cardiovascular Health Study, a population-based study of cardiovascular disease risk factors, 24-h ambulatory electrocardiography was performed. RESULTS: Serious arrhythmias, such as sustained ventricular tachycardia and complete atrioventricular block, were uncommon, but brief episodes of ventricular tachycardia (> or = 3 consecutive ventricular depolarizations) were detected in 4.3% of women and 10.3% of men. Ventricular arrhythmias as a group (excluding ectopic beats < 15/h) were more common in men than in women but were not significantly associated with age. The same patterns were true for bradycardia/conduction blocks. Supraventricular arrhythmias as a group (excluding ectopic beats < 15/h), in contrast, did not differ by gender but were strongly associated with increased age. Multivariate analyses showed associations with arrhythmias to differ by gender, with only one association (increased age and supraventricular arrhythmias) present in both women and men. Ventricular arrhythmias, particularly in men, were associated with a higher prevalence of cardiovascular disease and its risk factors and with subclinical disease, as measured by increased left ventricular mass and impaired left ventricular function. CONCLUSIONS: Arrhythmias are common in the elderly, and their association with cardiovascular disease differs by gender. Although risk related to arrhythmias can only be determined by prospective study, such studies should have adequate power to examine potential gender differences in associations.

Is thiazide-produced uric acid elevation harmful? Analysis of data from the Hypertension Detection and Follow-up Program.

Interaction of thiazide diuretics and the serum uric acid and creatinine levels was studied in 3693 stepped care participants in the Hypertension Detection and Follow-up Program not receiving treatment at baseline. Among men grouped into quartiles by their level of uric acid at baseline, the upper quartile (average uric acid, 7.7 mg/dL [458 mumol/L]) had an average serum creatinine level of 1.2 mg/dL (106 mumol/L) and the lowest quartile (uric acid, 4.9 mg/dL [291 mumol/L]) had an average serum creatinine level of 1.1 mg/dL (97 mumol/L). Similar findings were present in women. Therapy with chlorthalidone or other thiazide-type diuretics tended to increase levels of uric acid and creatinine, but the increase in both was less in the upper quartile than in the lower quartile. Among individuals who were prescribed uric acid-lowering drugs, the level of serum creatinine increased just as much as in those whose uric acid level was not pharmacologically lowered. Baseline uric acid level was a weak predictor of mortality in men; the introduction of an interaction term for creatinine suggested that this effect was primarily restricted to those with elevated levels of both uric acid and creatinine at baseline. Change in uric acid level at one year after therapy was inversely correlated with mortality in men. There were few episodes of gout (only 15 recorded in five years among 3693 participants at risk). These results suggest that neither the baseline uric acid level nor the change in uric acid level produced by therapy injures the kidney. These results suggest no reason to lower uric acid levels pharmacologically in the treated hypertensive patient who is not gouty. They leave unanswered whether there is a predictive value to baseline uric acid level not explainable by other correlated cardiovascular risk factors.

Properties of the random zero sphygmomanometer.

The random zero sphygmomanometer is widely used in studies involving blood pressure measurement because it is believed to eliminate digit preference and reduce measurement error. We performed blood pressure measurements sequentially using random zero and standard sphygmomanometers in random order in 1,356 participants in the Cardiovascular Health Study. Despite adherence to the manufacturer's instructions, we observed a substantially nonuniform distribution of zero levels generated by the random zero sphygmomanometer and a disturbing correlation between the zero level and blood pressures taken with the standard sphygmomanometer. With the random zero device, the pooled estimated slopes for the regression of standard systolic and diastolic pressures on the zero level were -0.71 and -0.17, respectively (both p < 0.0001). The only plausible explanation for this relation between the random zero device and the standard device is that by some unknown mechanism the subject's blood pressure is influencing the zero level. Systolic and diastolic blood pressures measured with the random zero device were, respectively, 1.65 and 1.84 mm Hg lower (both p < 0.0001) than standard blood pressures. Digit preference was detectable in the uncorrected blood pressure and zero level measured with the random zero device but was eliminated after calculation of the corrected blood pressure. For most epidemiological studies, the random zero sphygmomanometer offers no significant advantage over the standard sphygmomanometer. It may still be useful in those epidemiological studies and clinical trials where blinding is important.

Reserpine and breast cancer in the Hypertension Detection and Follow-Up Program.

Recent reports on the efficacy of pharmacological management in reducing mortality associated with mild hypertension have enhanced the importance of increasing our knowledge about drug toxicity. The Hypertension Detection and Follow-Up Program (HDFP) provides a convenient setting in which to examine the association reported between reserpine usage and breast cancer. In the intensively treated and followed group (Stepped Care [SC]), the relative breast cancer experience of those who did take reserpine and those who did not was examined. Of 2529 females in SC, 1036 received reserpine, with an average exposure of 1.97 years during 5 years of follow-up. Through extensive investigation, 21 cases of breast cancer were identified. using a life table regression method of analysis to adjust for actual time of reserpine exposure, race, sex, and medication status at entry, and comparing those who took reserpine with those who did not, the author's calculated a risk ratio of 1.28, with a confidence interval of 0.58 to 2.80. Adjustment for a number of other variables known to have relationships to breast cancer did not appreciably change the results. Thus, with certain precautions, the authors conclude that in this setting there is no indication of the recently postulated association of reserpine and the short-term enhancement of breast tumor growth.

Combinations of potassium, calcium, and magnesium supplements in hypertension.

Dietary intakes of potassium, calcium, and magnesium have each been reported to lower blood pressure, but the extent of blood pressure reduction in epidemiological studies and clinical trials has tended to be small and inconsistent. We hypothesized that combinations of these mineral supplements would lower blood pressure and that the reductions would be greater than that usually reported in studies of each cation alone. One hundred twenty-five patients 82 men and 43 women) with untreated mild or borderline hypertension were randomly assigned to daily treatment with one of the following four regimens: 60 mmol potassium and 25 mmol (1000 mg) calcium, 60 mmol potassium and 15 mmol (360 mg) magnesium, calcium and magnesium, or placebo. Standardized clinic blood pressure measurements were obtained on 3 days at baseline and after 3 and 6 months of treatment. At baseline, systolic and diastolic blood pressures (mean +/- SD) were 139 +/- 12 and 90 +/- 4 mm Hg, respectively, and dietary intakes of potassium, calcium, and magnesium were 77 +/- 32, 19 +/- 13, and 12 +/- 52 mmol/d, respectively. The mean differences (with 95% confidence intervals) of the changes in systolic and diastolic blood pressures between the treatment and placebo groups were not significant: -0.7 (-4.3 to +2.9) and -0.4 (-2.9 to +2.1) for potassium and calcium, -1.3 (-4.4 to +1.8) and 0.4 (-2.5 to +3.3) for potassium and magnesium, and +2.1 (-1.8 to +6.0) and +2.2 (-1.0 to +5.4) for calcium and magnesium. In conclusion, this trial provides little evidence of an important role of combinations of cation supplements in the treatment of mild or borderline hypertension.

Utility of new electrocardiographic models for left ventricular mass in older adults. The Cardiovascular Health Study Collaborative Research Group.

Several multivariate statistical models have recently been introduced for estimation of left ventricular mass from standard 12-lead electrocardiographic measurements. The validity of these algorithms has not been adequately evaluated. The objective of this investigation was to compare the associations between echocardiographic and electrocardiographic left ventricular mass values with clinical and subclinical indexes of coronary heart disease. The evaluation was performed with participants of the Cardiovascular Health Study, a population-based sample of 5201 men and women aged 65 years and older. Echocardiographic M-mode measurements of left ventricular mass were performed from videotape recordings with the use of a strictly standardized protocol. Electrocardiographic algorithms of the Novacode program and new algorithms derived from the Cardiovascular Health Study population were used for left ventricular mass prediction. Echocardiographic and electrocardiographic determinations of left ventricular mass were technically successful in 3410 (65.6%) and 5013 (96.4%) participants, respectively. The Novacode model overestimated echocardiographic left ventricular mass. Compared with the Novacode model, the new Cardiovascular Health Study electrocardiographic model, which includes adjustment for body weight, eliminated left ventricular mass prediction bias and improved the correlation between echocardiographic and electrocardiographic left ventricular mass from .33 to .54 in women and from .46 to .51 in men. Echocardiographic and electrocardiographic models both demonstrated similar and about equally strong associations with overt and subclinical disease and with risk factors for left ventricular hypertrophy. These observations demonstrate the potential utility of electrocardiographic models for left ventricular mass estimation.

Impact of hypertension information on high blood pressure control between 1973 and 1978.

To determine whether the immense multifocal efforts in the United States over the past 7 years to detect and treat high blood pressure (BP), had affected the status of hypertension, data from a national household survey in 1973-74 were compared with data obtained in 1977-78 from a second non-overlapping population in the same three communities. The impact of hypertension programs was measured by assessing change over the 5-year period in BP distribution, degree of awareness, and level of treatment in the population. Our data show that a substantial improvement in the status of high BP detection, treatment, and control has occurred since the early 1970s for all age, sex, and race groups studied.

Correlates of blood pressure in community-dwelling older adults. The Cardiovascular Health Study. Cardiovascular Health Study (CHS) Collaborative Research Group.

Although elevated blood pressure is an important predictor of cardiovascular disease and stroke in the elderly, little information exists on the distribution and risk factor correlates of blood pressure in this group. As part of the Cardiovascular Health Study, a population-based cohort study of 5201 men and women aged 65 to 101 years, we investigated correlates of systolic and diastolic blood pressure. Multiple regression analyses were conducted for all participants and a subgroup of 2482 without coronary heart disease and not on antihypertensive therapy (the "healthier" subgroup). In the total group, independent predictors of diastolic blood pressure included heart rate, aortic root dimension, creatinine, hematocrit, alcohol use, and black race (positive associations) and internal carotid artery wall thickness, mitral early/late peak flow velocity, white blood cell count, cigarette smoking, and age (negative associations). Positive predictors of systolic blood pressure included mitral late peak flow velocity, left ventricular mass, common carotid artery wall thickness, serum albumin, factor VII, diabetes, alcohol use, and age; negative predictors were coronary heart disease, uric acid, height, and smoking. In the healthier subgroup, positive predictors of diastolic blood pressure included heart rate, hematocrit, serum albumin, creatinine, and body weight, whereas mitral early/late peak flow velocity, serum potassium, smoking, and age inversely related to diastolic pressure. For the same group, common carotid artery wall thickness, left ventricular mass, serum albumin, factor VII, high-density lipoprotein cholesterol, and age were directly related to systolic blood pressure, whereas serum potassium was inversely related. Both systolic and diastolic pressures varied considerably by geographic site.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of acute caloric restriction on cholesterol metabolism in man.

The effects of acute caloric restriction on cholesterol balance and kinetics of plasma cholesterol specific activity were investigated in five hyperlipemic subjects with varying degrees of obesity. Caloric restriction decreased plasma triglycerides by 41 +/- 12%, plasma cholesterol by 11 +/- 9%, and the ratio of esterified to free cholesterol by 12 +/- 7+. Immediately on institution of caloric restriction there appeared to be an influx of tissue cholesterol into plasma and a reduction in endogenous synthesis of cholesterol. The cholesterol balance decreased from 1,469 +/- 441 to 1,212 +/- 349 mg/day and the rate of decay of plasma cholesterol specific activity decreased 62 +/- 3%. The effect of caloric restriction on hepatic synthesis of bile acids was also very prompt. The total fecal bile acids were reduced immediately by 36 +/- 7%. Because the effect on fecal excretion of deoxycholic acid was greater than that on fecal lithocholic acid, it was suggested that hepatic synthesis of cholic acid was reduced more than the synthesis of chenodeoxycholic acid. Caloric restriction did not cause any change in the percentage of absorption of dietary cholesterol (40 +/- 2% versus 42 +/- 3%). These observations are in accord with our model relating cholesterol metabolism with the metabolism of plasma lipoproteins in man.

Effect of dietary sodium restriction on taste responses to sodium chloride: a longitudinal study.

Normotensive adults on low-sodium, weight-loss, and control diets recorded preferences and perceived saltiness for sodium chloride (NaCl) added to cream soup at intervals over 1 yr. Reduction in sodium intake and excretion accompanied a shift in preference toward less salt: preferred concentrations by ad libitum salting declined from 0.72% at the onset to 0.33% NaCl at week 24; hedonic scores for high concentrations of NaCl decreased significantly while scores for low concentrations increased. After 3 mo of sodium restriction, NaCl preferences readjusted to a lower level: ad libitum additions of NaCl were similar after 13, 24, and 52 wk. Less hedonic variation was observed among controls than among Na-restricted groups. The weight-loss group showed increased liking for mid-range NaCl levels. Mechanisms underlying preference changes, including physiological, behavioral, and context effects, may provide insights into maintenance of low-sodium diets for treatment and prevention of hypertension.

Efficacy of nonpharmacologic interventions in adults with high-normal blood pressure: results from phase 1 of the Trials of Hypertension Prevention. Trials of Hypertension Prevention Collaborative Research Group.

Phase 1 of the Trials of Hypertension Prevention was conducted in 2182 adults, aged 35-54 y, with diastolic blood pressure of 80-89 mm Hg to test the feasibility and blood pressure-lowering effects of seven nonpharmacologic interventions (weight loss, sodium reduction, stress management, and supplementation with calcium, magnesium, potassium, and fish oil). At 6 and 18 mo, weight loss and sodium reduction were well-tolerated and produced significant declines in systolic and diastolic blood pressures (-2.9/-2.4 and -2.1/-1.2 mm Hg for weight loss and sodium reduction, respectively, at 18 mo). None of the other interventions lowered blood pressure significantly at either the 6- or 18-mo follow-up visits. These results suggest that both weight loss and sodium reduction provide an effective means to prevent hypertension. The long-term effects of both of these interventions are being tested in phase 2 of the trial.

University of California, Davis, conference: Mild hypertension.

Prevalence of "higher than normal" blood pressures in a community is inversely related to the magnitude of the elevation; the milder grades of elevation are far more prevalent. A multifactorially inherited tendency to develop hypertension is modulated by multiple environmental influences. Autonomic nervous and behavioral factors plausibly appear to contribute to the initiating mechanisms of hypertension; the associated hemodynamic changes and the resulting cardiovascular structural changes interact to perpetuate the process. The complex interaction of hypertension and atherosclerosis is further complicated by direct as well as secondary effects of antihypertensive drugs on atherogenesis. Attributable cardiovascular risk is generally proportional to the degree of hypertension across the entire range of elevated blood pressure; this kind of relationship holds also for normal versus subnormal blood pressure values. Pharmacologic lowering of blood pressure, however, does not confer proportional benefit. Thus, such lowering of blood pressure to normotensive levels does not reduce the risk level to that in the normotensive population. Therapeutic outcome is influenced by the interaction of blood pressure lowering, type of antihypertensive agents used, existing risk factors, and target organ damage. Benefits of lowering blood pressure in established mild hypertension (diastolic blood pressure greater than 95 mm Hg) are confirmed. Drug treatment of patients with lower diastolic blood pressure or with isolated elevations of systolic blood pressures continues to be controversial as does the choice of initial therapeutic agent(s). The large-scale experience of clinical trials encompassing the long-term risks and benefits of the drug treatment of mild hypertension is limited to the use of diuretics and adrenergic beta blockers. A variety of new and promising therapeutic agents for use as alternate choices for initial therapy needs to undergo comparative evaluation.

Short report: the effect of fish oil on blood pressure and high-density lipoprotein-cholesterol levels in phase I of the Trials of Hypertension Prevention.

OBJECTIVE: To study the effects of moderate doses of fish oil on blood pressure and high-density lipoprotein (HDL)-cholesterol. METHODS: The participants were 350 normotensive men and women aged 30-54 years who were enrolled from seven academic medical centers in phase I of the Trials of Hypertension Prevention. They were randomly assigned to receive placebo or 6 g purified fish oil once a day, which supplied 3 g n-3 polyunsaturated fatty acids for 6 months. RESULTS: Baseline blood pressure was (mean +/- SD) 123 +/- 9/81 +/- 5 mmHg. The mean differences in the blood pressure changes between the fish oil and placebo groups were not statistically significant. There was no tendency for fish oil to reduce blood pressure more in subjects with baseline blood pressure in the upper versus the lower quartile (132/87 versus 114/75 mmHg), low habitual fish consumption (0.4 versus 2.9 times a week) or low baseline plasma levels of n-3 fatty acids. Fish oil increased HDL2-cholesterol significantly compared with the placebo group. Subgroup analysis showed this effect to be significant in the women but not in the men. Increases in serum phospholipid n-3 fatty acids were significantly correlated with increases in HDL2-cholesterol and decreases in systolic blood pressure. CONCLUSION: Moderate amounts of fish oil (6 g/day) are unlikely to lower blood pressure in normotensive persons, but may increase HDL2-cholesterol, particularly in women.

Left ventricular hypertrophy, arrhythmias and sudden death in systemic hypertension.

Left ventricular (LV) hypertrophy is a significant independent risk factor for mortality from coronary heart disease, including sudden death. The proportion of sudden death to total death due to coronary heart disease remains high, at about 50% to 60%, despite the continuing downward trend in coronary heart disease mortality observed in the U.S. during the last decade. Prevalence of LV hypertrophy, determined by electrocardiogram in hypertensive patients (diastolic blood pressure greater than or equal to 90 mm Hg), including tall R wave and evidence of repolarization abnormality, is around 5%. The prevalence of LV hypertrophy by echocardiography is estimated at 44% to 48%. LV hypertrophy on electrocardiogram underestimates the magnitude of the problem of LV hypertrophy in hypertensive patients. Its overall sensitivity is less than 60%. The incidence of LV hypertrophy in hypertensive patients is lower when hypertension is treated successfully than when the patient is left either untreated or inadequately treated; successful treatment of hypertension causes regression of LV hypertrophy. However, hypertensive patients with LV hypertrophy have a poor prognosis despite treatment. The available evidence derived from the results of large clinical trials suggests that hypertensive patients should be treated before there is electrocardiographic evidence of LV hypertrophy.

Primary prevention of coronary heart disease: a critique.

The question is whether alteration of risk factors will aid primary and secondary prevention of coronary heart disease. Critical review of available evidence indicates that inferences have been made about the beneficial effects of risk factor modification without an adequate test of the hypothesis. Trial interventions to assess the efficacy of serum cholesterol-lowering measures have had negative or equivocal results. It remains to be seen whether the findings of clinical trials on hypertension can be applied toward primary prevention of coronary heart disease in the community. The cigarette smoking habit seems to be unique among coronary heart disease risk factors. The evidence appears sufficient to justify serious consideration of a strategy of preventing the smoking habit now, persuading patients to stop and encouraging teenagers not to start.

Major electrocardiographic abnormalities in persons aged 65 years and older (the Cardiovascular Health Study). Cardiovascular Health Study Collaborative Research Group.

Electrocardiographic abnormalities are often found in older patients, but their prevalence in free-living elderly populations is not well-defined. In addition, the clinical significance of many of these abnormalities is uncertain. The prevalence of major electrocardiographic abnormalities was determined in 5,150 adults aged greater than or equal to 65 years from the Cardiovascular Health Study--a study of risk factors for stroke and coronary heart disease in the elderly. Ventricular conduction defects, major Q/QS waves, left ventricular hypertrophy, isolated major ST-T-wave abnormalities, atrial fibrillation and first-degree atrioventricular block were collectively categorized as major electrocardiographic abnormalities. Prevalence of any major electrocardiographic abnormality was 29% in the entire cohort, 19% among 2,413 participants who reported no history of coronary artery disease or systemic hypertension, and 37% among 2,737 participants with a history of coronary artery disease or hypertension. Prevalence of major electrocardiographic abnormalities was higher in men than in women regardless of history, and tended to increase with age. Major Q/QS waves were found in 5.2%, and more than half were in those who did not report a previous myocardial infarction. Major electrocardiographic abnormalities are common in elderly men and women irrespective of the history of heart disease.

The effect of chlorthalidone on ventricular ectopic activity in patients with isolated systolic hypertension. The SHEP Study Group.

In an ancillary study of the Systolic Hypertension in the Elderly Program (SHEP), the effects of diuretics on ventricular ectopic activity were investigated in 186 patients with isolated systolic hypertension. Ventricular premature complexes (VPCs) were examined as the number of VPCs/24 hours, presence of > or = 1 VPC, presence of > or = 10 VPCs/24 hours, and presence of VPC pairs or ventricular tachycardia. Significant changes in VPCs were not observed either in the 92 patients randomized to chlorthalidone stepped-care (12.5 and 25 mg/day) or in the 94 placebo-treated patients (p > 0.1 for all VPC definitions and both groups). Serum potassium decreased from 4.4 +/- 0.5 to 4.1 +/- 0.5 mEq/liter (p = 0.002) in the chlorthalidone group and did not change (4.4 +/- 0.5 to 4.5 +/- 0.4 mEq/liter) in the placebo group. Potassium was prescribed routinely for confirmed hypokalemia < 3.5 mEq/liter. A relation between serum potassium and VPC or change in serum potassium and change in VPC was not observed in the chlorthalidone group. In summary, in patients with isolated systolic hypertension, chlorthalidone in doses that are effective in decreasing stroke and cardiovascular event rates (12.5 or 25 mg/day), did not increase VPCs.

Validation of reported myocardial infarction mortality in blacks and whites. A report from the Community Cardiovascular Surveillance Program.

It is widely believed that blacks experience a higher mortality due to coronary heart disease (CHD) than do whites. To determine whether this reported difference in mortality between blacks and whites is real, we studied the question in the context of the Community Cardiovascular Surveillance Program (CCSP). Fatal and nonfatal cases of CHD were reviewed in 12 US communities. Standardized criteria were applied to classify these cases as possible CHD, definite CHD, possible myocardial infarction (MI), or definite MI. The annual age-adjusted mortality rate per 100,000 ascribed to definite MI by the CCSP criteria was higher in blacks than in whites: 47 in white men (95% confidence interval, 36 to 58), 18 in white women (95% confidence interval, 8 to 28), 95 in black men (95% confidence interval, 10 to 180), and 41 for black women (95% confidence interval, 0 to 99). The proportion of definite MI to all fatal CHD events was higher in blacks (16%) than in whites (12%). For nonfatal events, however, the rate of definite MI was higher in whites than in blacks: 322 in white men (95% confidence interval, 293 to 351), 225 in black men (95% confidence interval, 160 to 290), 82 in black women (95% confidence interval, 43 to 121), and 103 in white women (95% confidence interval, 88 to 118). The proportion of definite MI to all nonfatal CHD events was lower in blacks (16%) than in whites (30%). Thus, the overall rate for fatal and nonfatal definite MI was lower in blacks (215/100,000) than in whites (244/100,000). These observations suggest that a combination of high case-fatality ratio and misclassification of cause and death may contribute to the reported higher rate of CHD mortality among blacks.

Baseline characteristics of participants in phase II of the Trials of Hypertension Prevention (TOHP II). Trials of Hypertension Prevention (TOHP) Collaborative Research Group.

Phase II of the Trials of Hypertension Prevention (TOHP II) is a multicenter, controlled clinical trial designed to test whether weight loss, a reduced sodium intake, or a combination of weight loss and a reduced sodium intake will lower blood pressure (BP) and prevent the occurrence of hypertension. The study population consists of middle-aged, moderately overweight individuals with a diastolic BP between 83 and 89 mm Hg. Of the 2382 randomized participants, 816 (34%) are female and 494 (21%) are from a racial or ethnic minority background. At baseline, mean dietary intakes of sodium, based on measurements of 24-hour urinary excretion, were 199 mmol/d in men and 154 mmol/d in women. The average body mass index was 30.9 kg/m2. Across the four randomized groups, there was no substantial imbalance in the distribution of baseline variables; however, the mean age in the four groups was slightly but significantly different (range: 43.2 to 44.2 years, P = 0.02). A comparison of baseline characteristics of TOHP II participants with those of participants in three other primary prevention trials reveals a high level of mean dietary sodium intake in each study. Data reported in this article indicate that any subsequent differences in BP among the randomized groups are unlikely to result from maldistribution of known confounding variables at baseline. Finally, because of the high prevalence of overweight and excessive sodium intake in the United States, results from TOHP II should be broadly applicable to the general population.

Recruitment of adults 65 years and older as participants in the Cardiovascular Health Study.

Few large-scale epidemiologic studies have enrolled older adults; hence, little is known about the feasibility of recruiting this group for long-term population-based studies. In this article we present the recruitment experience of the Cardiovascular Health Study (CHS), a population-based, longitudinal study of cardiovascular diseases in adults 65 years and older. Participants were sampled from the Health Care Financing Administration's (HCFA) Medicare eligibility lists in four US communities. Letters were mailed to 11,955 sampled individuals. Persons recruited were required to complete an extensive home interview and then a 4-hour in-clinic examination. Excluded were persons who were expected to be able to complete the baseline examination and who were not expected to return for the 3-year follow-up. Some 3654 participants were recruited from those randomly selected from the Medicare sampling frame. In addition, 1547 other age-eligible persons living in the household with the sampled individuals also participated, yielding a total of 5201 participants. Of those who were contacted, 9.6% were ineligible and 34.9% refused participation. Among those eligible, 38.6% refused and 57.3% were enrolled (the remaining did not refuse but were not enrolled before the recruitment ended). Data from a subsample indicate that compared to those who were ineligible or who refused, enrolled participants were younger, more highly educated, more likely to be married, and less likely to report limitations in activity. Compared to those who were eligible but refused, enrolled participants were less likely to have high blood pressure and stroke and more likely to have quit smoking and to perceive their health status as very good or excellent.(ABSTRACT TRUNCATED AT 250 WORDS)