(Circulation. 1996;93:60-66.) 1996 American Heart Association, Inc.

Articles

Blood Pressure in Young Blacks and Whites: Relevance of Obesity and Lifestyle Factors in Determining Differences
The CARDIA Study
Kiang Liu, PhD; Karen J. Ruth, MS; John M. Flack, MD; Rhonda Jones-Webb, DrPH; Gregory Burke, MD; Peter J. Savage, MD; Stephen B. Hulley, MD
From the Department of Preventive Medicine, Northwestern University Medical School, Chicago, Ill (K.L., K.J.R.); the Hypertension Center, Bowman Gray School of Medicine, Winston-Salem, NC (J.M.F.); the Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis (R.J.-W.); the Department of Public Health Sciences, Bowman Gray School of Medicine, Winston-Salem, NC (G.B.); the Division of Epidemiology and Clinical Applications, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md (P.J.S.); and the Department of Epidemiology and Biostatistics, University of California at San Francisco (S.B.H.). Correspondence to Kiang Liu, PhD, Department of Preventive Medicine, Northwestern University Medical School, 680 N Lake Shore Dr, Suite 1102, Chicago, IL 60611.

Abstract
Top Abstract Introduction Methods Results Discussion References

Background Middle-aged black men and women have higher blood pressure, on average, than whites. However, this pattern is inconsistent in children and adolescents. This study explores how differences in lifestyle factors in young adulthood may influence blood pressure patterns in the two races.

Methods and Results The Coronary Artery Risk Development in Young Adults (CARDIA) study is an ongoing collaborative investigation of lifestyle and the evolution of cardiovascular disease risk factors in a random sample of young adults ages 18 to 30 years at baseline (1985 to 1986). Data from four examinations over 7 years were analyzed with the use of a method that simultaneously examined cross-sectional and longitudinal relationships of lifestyle factors and blood pressure. This study included 1154 black women, 853 black men, 1126 white women, and 1013 white men. Blacks had higher systolic blood pressure and diastolic blood pressure than whites at every examination. Racial differences were much greater in women than in men and increased over time. Within each sex-race group, average diastolic blood pressure over four

examinations was positively associated with baseline age, body mass index, and alcohol intake and negatively associated with physical activity, cigarette use, and intake of potassium and protein. Longitudinal change in diastolic blood pressure was positively associated with changes in body mass index and alcohol intake. After adjustment for obesity and other lifestyle factors, black-white diastolic blood pressure differences were reduced substantially: 21% to 75% for men and 49% to 129% for women. Results for systolic blood pressure were similar. Conclusions Differences in obesity and other lifestyle factors in young adults largely explain the higher baseline blood pressure and greater increase over time of blacks relative to whites. Key Words: blood pressure lifestyle obesity race

Introduction

Hypertension is a major public health problem in the United States, especially in blacks. Many studies have demonstrated that for middle-aged men and women, blacks have higher mean blood pressure and prevalence of hypertension than whites.1 2 3 4 5 For example, data from the second National Health and Nutrition Examination Survey indicated that the prevalence Top rates of definite hypertension (systolic blood pressure [SBP] 160 mm Hg, Abstract diastolic blood pressure [DBP] 95 mm Hg and/or taking antihypertensive Introduction medication) in persons aged 55 to 64 years were 60% and 46% for black women Methods Results and men and 34% and 31% for white women and men, respectively.3 However, 6 7 8 9 10 11 12 13 Discussion for children and adolescents, the results are inconsistent. One References plausible hypothesis is that during young adulthood, obesity, various aspects of lifestyle, and psychosocial characteristics may differ between blacks and whites and subsequently contribute to evolving racial patterns of blood pressure. This study presents an examination of this hypothesis using 7-year follow-up data from the CARDIA study. Specifically, data were analyzed for the following purposes: (1) to describe the black-white differences in 7-year blood pressure change in young adults, (2) to determine factors that are associated with blood pressure change over time, and (3) to examine the degree to which the observed black-white differences in blood pressure change among young adults can be explained by the differences in these factors.

Methods

The CARDIA study is a multicenter, longitudinal study on lifestyle and evolution of cardiovascular disease risk factors. The cohort consisted of 5115 black and white men and women aged 18 to 30 years at baseline with varying levels of socioeconomic status. Four clinical centers conducted the baseline (1985 to 1986), year 2 (1987 to 1988), year 5 (1990 to 1991), and year 7 (1992 to 1993) examinations. For the centers in Birmingham, Ala; Chicago, Ill; and Minneapolis, Minn, participants were randomly recruited from the total community or from selected census tracts in the community. For the center in Top Abstract Oakland, Calif, participants were randomly recruited from the KaiserIntroduction Permanente health plan membership. During the first 7 years of follow-up, 41 Methods participants died. Overall retention of the surviving cohort at the year 7 Results examination was 80%. The study sample used in the analyses included 1154 Discussion black women, 853 black men, 1126 white women, and 1013 white men who had References completed the baseline and at least two of three subsequent examinations. Among them, 1043 black women, 752 black men, 1058 white women, and 956 white men had completed all four examinations. Within each sex-race group, comparisons of baseline characteristics between those who were in the study sample and those who were not indicated that the two groups were similar with respect to blood pressure, body mass index (BMI), physical activity score, alcohol intake, and energy intake. However, for each sex-race group, those who were not in the study sample tended to be slightly younger (differences of 0.6 to 0.8 years), slightly less educated (differences of 0.4 to 0.6 years), and were more likely to smoke cigarettes. Detailed descriptions of the design of the CARDIA study and baseline blood pressure differences between blacks and whites have been published elsewhere.14 15 All measurements were taken by trained and certified technicians according to the CARDIA manual of operations. Before each examination, participants were asked to fast for 12 hours and to avoid smoking and heavy physical activity for 2 hours. After a 5-minute rest in a quiet room, three systolic and fifth-phase diastolic blood pressures were measured at 1-minute intervals on the participant's right arm with the use of a random zero sphygmomanometer. The average of the second and the third blood pressure measurements was used in the analyses. Height and weight were measured with the participant wearing light clothing and no shoes. Height was recorded to the nearest 0.5 cm and weight to the nearest 0.2 lb. BMI was calculated as weight (kg) divided by height squared (m2). Age at baseline was computed from the reported birth date. Sex, race, years of education, and number of cigarettes smoked per day were self-reported. Alcohol intake (mL/d) was computed from the self-reported frequency of beer, wine, and liquor consumed per week.16 Physical activity score was derived from the CARDIA Physical Activity History, a modified version of the Minnesota Leisure Time Physical Activity Questionnaire.17 A hostility score was calculated as the sum of the `hostile' responses on the 50-item Cook-Medley subscale of the Minnesota Multiphasic Personality Inventory (MMPI).18 Dietary protein (% kcal), calcium (mg/1000 kcal), and potassium (mg/1000 kcal) were derived from the baseline and year 7 CARDIA dietary history, an interviewer-administered quantitative food frequency questionnaire.19 These variables, reflecting aspects of overweight, lifestyle, and psychosocial characteristics, were selected because they have been reported to be associated with blood

pressure.15 16 20 21 22 23 24 25 26 27 28 29 30 31 32 Quality of the data collection was monitored by the Coordinating Center and the CARDIA Quality Control Committee throughout the four examination periods. A more detailed description of the methods was published in the CARDIA baseline monograph.33 The longitudinal analyses were based on the Generalized Estimating Equation method developed by Liang and Zeger.34 35 This method simultaneously examines the cross-sectional relationship between each of the independent variables and blood pressure and the relationship between changes in these variables and changes in blood pressure. For each sex-race group, a typical model is

where for t=0, 2, 5, 7, Yit is the SBP for the ith person at year t; Ui is a time-independent covariate (baseline age); Xit is a time-dependent covariate for the ith person at time t (for example, for t=0, 2, 5, 7, Xit=1 if the ith person is on antihypertensive medication, and Xit=0 otherwise); Zi0 is the baseline value of a time-dependent covariate (baseline weight), Zit=Zit-Zi0 (that is, the change in weight between year t and baseline for the ith person), and eit is the error term. The coefficient 1 measures the average annual change in SBP adjusting for the covariates. The coefficient 2 measures the association between the SBP (averaged over time) and baseline age adjusting for other covariates. The coefficient 3 measures the average difference in SBP between those who are on antihypertensive medication and those who are not on medication. Similarly, the coefficient 4 measures the relationship between the average SBP and baseline weight adjusting for other covariates. The coefficient 5 measures the relationship between changes in weight and changes in SBP over time. Within each sex group, in addition to the race-specific model, one can use three dummy variables to separate the four time points and one dummy variable to separate the two race groups. The three cross-product terms between the variables for time and race provide the different racial patterns of changes in SBP over time.

Results

Comparison of Baseline Characteristics The baseline characteristics of the participants are provided in Table 1 . Blacks were on average 1 year younger than whites. The average education level was lower in blacks than in whites. For both men and women, average SBP was higher in blacks than in whites; this difference was fairly large in women, reaching 3.4 mm Hg. On the other hand, Top Abstract the average DBP level was similar for black and white men and was slightly but Introduction significantly higher in black women than in white women. Few people in the Methods four sex-race groups reported antihypertensive drug treatment (ranging from Results 0.4% in white women to 1.0% in black women and white men). The baseline Discussion average BMI was comparable between black and white men; however, black References women were on average much more obese than white women. The average physical activity score was similar for black and white men but was much lower in black women than white women. The prevalence of alcohol drinking was higher in whites than in blacks; however, for drinkers, the intake per day was similar for the two racial groups. The prevalence of cigarette smoking was higher in blacks than in whites; however, for smokers, the number of cigarettes smoked per day was higher in whites than in blacks. Blacks had higher hostility scores and lower potassium, calcium, and protein intakes than whites.

View this table: Table 1. Baseline Characteristics of the Cohort by Sex and Race [in this window] [in a new window] Black-White Difference in Blood Pressure Over Time Fig 1 presents the black-white difference in blood pressure over time for men and women. For three of four sex and race groups, the average DBP decreased somewhat at year 2 and then increased at years 5 and 7. The decreases at year 2 were presumably due to the adaptation effect. For men, the black-white differences in average DBP were -0.2, 1.4, 1.9, and 1.6 mm Hg at baseline, year 2, year 5, and year 7, respectively. For women, the black-white differences increased from 1.1 mm Hg at baseline to 2.0 mm Hg at year 2, 4.1 mm Hg at year 5, and 4.6 mm Hg at year 7. For black men, white men, and white women, the average SBP continued to decrease even at year 5 and then increased at year 7 (Fig 1 ). For black women, the average SBP decreased at year 2 and then increased at years 5 and 7. Similar to DBP, the black-white difference in SBP increased over time. For men, the difference increased from 1.6 mm Hg at baseline to 3.8 mm Hg at year 7. For women, the difference increased from 3.4 to 6.6 mm Hg.

Figure 1. Plot of average systolic blood pressure (SBP) and diastolic blood pressure (DBP) (mm Hg) at each CARDIA exam (years 0, 2, 5, and 7) for black and white men (left) and black and white women (right).

View larger version (24K): [in this window] [in a new window] Factors Associated With Blood Pressure and Changes in Blood Pressure Table 2 presents the results of the longitudinal analyses for DBP. Each sex-race group was analyzed separately; each column represents a separate analysis. Within each sex-race group, the three types of variables (baseline, time-dependent, and change) were included simultaneously in the same model. The estimated coefficient between a baseline variable and blood pressure can be viewed as the difference in DBP averaged over 7 years corresponding to per unit difference in this variable after adjusting for the longitudinal relationship. Baseline age and BMI were significantly positively associated and baseline cigarette smoking was significantly negatively associated with DBP in all sex-race groups. Baseline alcohol intake was positively associated with DBP in all groups; however, the relationship was significant in black men only. Baseline physical activity score was significantly negatively related to DBP in men but not in women. Baseline potassium intake was significantly negatively associated with DBP in white women. No consistent associations were observed between baseline education, hostility score, calcium intake, or protein intake and DBP.

View this table: [in this window] [in a new window]

Table 2. Factors Associated With DBP and Changes in DBP: Regression Coefficient From Generalized Estimated Equation Analyses

The coefficient of time reflects the average change in DBP per year (Table 2 ). For blacks, the positive coefficient indicates that on average DBP increased slightly over the 7-year period. For whites, average DBP decreased slightly over this period. Antihypertensive drug treatment was a time-dependent variable. The coefficient measures the average difference in DBP between persons with treatment and persons without treatment (including normotensive) at the four time points. The results indicate that except for white women, persons on antihypertensive drug treatment had significantly higher DBP than persons not on treatment. Thus, for hypertensive

subjects on treatment, blood pressure was not normalized to a level similar to that of normotensive subjects. Change in BMI was significantly related to change in DBP in every sex-race group (Table 2 ). Change in alcohol intake was significantly positively associated with change in DBP in white men and women and black women. On the other hand, changes in the number of cigarettes smoked per day and physical activity score were not associated with change in DBP. Since the dietary variables and hostility score were not collected at every examination, the longitudinal associations could not be determined. However, in a separate analysis to assess the relationships between 7-year changes in potassium, protein, or calcium intake and changes in DBP or SBP, none of the correlations was statistically significant (data not shown). For most variables, the associations with SBP were generally similar to those with DBP (Table 3 ). The only exceptions were that the baseline number of cigarettes smoked per day was unrelated to SBP, the baseline hostility score was inversely associated with average SBP in white men and women, and age was related to average SBP in black women only.

View this table: [in this window] [in a new window]

Table 3. Factors Associated With SBP and Changes in SBP: Regression Coefficient From Generalized Estimated Equation Analyses

Impact of Obesity and Lifestyle Factors on Black-White Differences in Blood Pressure With adjustment for baseline age, antihypertensive treatment status, obesity, and lifestyle factors listed in Tables 2 and 3 , black-white differences in DBP were reduced for both men and women as compared with the differences in DBP adjusting for baseline age and antihypertensive treatment status only (Figs 2 and 3 ). For men, the black-white differences in DBP were reduced from 0.4 to 0.1 mm Hg (-75%) at baseline, 1.9 to 1.5 (-21%) at year 2, 2.3 to 1.6 (-30%) at year 5, and 2.0 to 1.2 (-40%) at year 7. For women, the reductions were from 1.4 to -0.4 mm Hg (-129%) at baseline, 2.4 to 0.3 (-88%) at year 2, 4.3 to 1.9 (-56%) at year 5, and 4.9 to 2.5 (49%) at year 7 (Fig 3 ). Thus, 49% of the observed black-white difference in DBP at year 7 in women and 40% in men were explained statistically by black-white differences in obesity and lifestyle factors. Similar results also were observed for SBP (data not shown).

View larger version (22K): [in this window] [in a new window]

Figure 2. Line graph shows adjusted diastolic blood pressure (DBP) (mm Hg) in black (B) and white (W) men at year 0, 2, 5, and 7 CARDIA examinations (left axis). Bar graph shows the black-white difference in adjusted DBP (mm Hg) at each examination (right axis). Dark line and dark bars indicate adjustment for age and antihypertensive medication use. Light line and light bars indicate further adjustment for age, education, body mass index, physical activity, alcohol, cigarettes, hostility, potassium intake, calcium intake, and protein.

View larger version (20K): [in this window] [in a new window]

Figure 3. Line graph shows adjusted DBP (mm Hg) in black and white women at year 0, 2, 5, and 7 CARDIA examinations (left axis). Bar graph shows the black-white difference in adjusted DBP (mm Hg) at each examination (right axis). Dark line and dark bars indicate adjustment for age and antihypertension medication use. Light line and light bars indicate further adjustment for age, education, body mass index, physical activity, alcohol, cigarettes, hostility, potassium intake, calcium intake, and protein. See Fig 2 for abbreviations.

These data were further analyzed with additional adjustment for center. Even though the average blood pressure differed among the four centers, adjusting for center did not change the results for the black-white differences (data not shown).

Discussion
Top Abstract Introduction Methods Results Discussion References

This study examined the black-white difference in blood pressure change during young adulthood. For both men and women, the blood pressure difference was very small at baseline and gradually increased over time. BMI and BMI change were strongly associated with blood pressure and blood pressure change, respectively, in all four sex-race groups. In addition, certain lifestyle factors and changes in these factors were also related to blood pressure or changes in blood

pressure. With adjustment of BMI and lifestyle variables, the black-white difference in blood pressure was substantially reduced. Patterns of Black-White Differences in Blood Pressure Previous studies on the difference in black-white blood pressure levels suggest that the pattern varies with age. Reported differences are inconsistent among children and adolescents.6 7 8 9 10 11 12 13 However, studies on middle-aged adults have clearly indicated that blacks have a higher average blood pressure and prevalence of hypertension than whites.1 2 3 4 5 Previous reports of the CARDIA baseline cross-sectional data suggest that the average blood pressure levels of blacks and whites start to diverge during young adulthood and that the difference increases with age.15 This report presents the longitudinal data to confirm that there were, at most, very small differences in blood pressure when participants were young. The differences increased over time as participants aged. Given that approximately 45% of the black-white differences in blood pressure at year 7 were explained by major environmental factors, these results suggest that black-white differences in obesity and lifestyle factors during young adulthood impact on the divergence of blood pressure over time. Factors Related to Blood Pressure Consistent with results from other studies, BMI was significantly associated with average DBP in every sex-race group20 ; in addition, BMI change was strongly associated with DBP change in every sex-race group, an association that was stronger in whites than in blacks. For example, for an increase in BMI for one unit, the average change in DBP was 0.5 mm Hg for black men, 0.3 mm Hg for black women, 1.0 mm Hg for white men, and 0.6 mm Hg for white women. These results suggest that for prevention of the age-related rise in blood pressure, and ultimately of hypertension, it is important to prevent weight gain in young adulthood while blood pressure levels are still mostly in the normal range and participants are not yet significantly obese. Alcohol consumption has been shown by many studies to be a risk factor for hypertension.16 21 22 In CARDIA, the baseline alcohol intake was only weakly associated with the average blood pressure during young adulthood. However, the longitudinal findings indicated that for young adults, blood pressure increased (or decreased) as alcohol intake increased (or decreased). This suggests that alcohol intake should be minimized, among other reasons, for prevention of hypertension. An inverse cross-sectional relationship between cigarette smoking and blood pressure independent of weight or BMI has been reported by many studies.23 24 In general, smokers tend to be leaner. It is possible that the effect of obesity on blood pressure is not completely removed by statistical adjustment for weight or BMI. It has also been hypothesized that the inverse relationship may be explained by depressor effects of cotinine.23 The results in CARDIA confirmed the cross-sectional relationship between cigarette smoking and blood pressure; however, longitudinally, there was no relationship between changes in the number of cigarettes smoked per day and the changes in blood pressure. Even if smoking does lower blood pressure slightly, it should be avoided for its well-known adverse effects on health. Habitual physical activity has been inversely associated with blood pressure in other studies.25 In CARDIA, the baseline physical activity score was inversely related to blood pressure in men but not in women, and there was no relationship between change in physical activity score and change in blood

pressure. The lack of relationship may in part be due to attenuation caused by the inaccuracy of the self-reported physical activity data. Several dietary factors including calcium, potassium, sodium, and protein have been found or hypothesized to be associated with blood pressure.26 27 28 29 30 Due to practical difficulties, in CARDIA, urinary sodium excretion was collected only in a subgroup of participants and therefore could not be included in the analysis. Baseline calcium intake was not related to baseline blood pressure, nor was it related to the 7-year average blood pressure.15 In addition, the 7-year change in calcium intake was not correlated with the 7-year change in blood pressure. These results are consistent with other studies that have shown no relationship between calcium intake and blood pressure.27 Many cross-sectional studies have reported an inverse relationship between potassium intake and blood pressure.28 29 In CARDIA, baseline potassium intake was significantly inversely related to average blood pressure only in white women and (borderline significant) in black men. A recent finding from the INTERSALT study suggests that protein intake is inversely associated with blood pressure.30 In CARDIA, although protein intake was inversely associated with average blood pressure in every sex-race group, none of the coefficients attained statistical significance. The psychosocial variables used in the analyses were education and hostility score. Unlike the results of many other studies, in CARDIA, baseline education was not associated with blood pressure.31 In this young adult cohort, baseline education is a less precise indicator of socioeconomic status because some participants were still going to school and others might go back to school in the future. Furthermore, similar to the findings of another study, the CARDIA data do not support the hypothesis that hostility is associated with hypertension.32 Impact of Obesity and Lifestyle on the Black-White Difference in Blood Pressure The data from this study indicate that obesity and certain lifestyle factors, for example, alcohol drinking and lack of physical activity, are responsible for a large proportion of the black-white difference in blood pressure. However, a residual black-white difference in blood pressure remains after adjusting for these factors. There are several hypothetical explanations for this difference. First, most of the lifestyle factors used in this study are based on self-reported data and may have large measurement errors. Measurement errors would attenuate the association between lifestyle factors and blood pressure and thus underestimate the reduction of a blackwhite difference in lifestyle factoradjusted blood pressure. Second, due to lack of data, several important factors (for example, sodium intake and certain psychosocial characteristics such as psychosocial stress or anxiety), were not measured and could not be included in the analyses. Third, race may independently predict blood pressure change. Certain stressors associated with minority status (for example, racial discrimination) may be related to the residual black-white difference in blood pressure. Several studies have shown that blacks with darker skin tend to have higher blood pressure.36 37 38 Blacks with darker skin also may be exposed to greater discrimination and poverty.38 Fourth, the black-white difference also may be influenced by genetic factors. It is unlikely that the reduction in the black-white difference was biased by the exclusion of participants from the original cohort. Approximately 19% of the participants were excluded from the analyses if they had missed two or more examinations. Comparisons of baseline

characteristics indicated that within each sex-race group, those who were included and those who were excluded were similar with respect to all important variables such as SBP, DBP, BMI, alcohol intake, and physical activity. For age, education, and cigarette smoking, small differences were observed for both blacks and whites. The relatively weak associations between these variables and blood pressure suggest these differences are unlikely to bias the results. Despite these problems, after adjusting for obesity and lifestyle factors, the black-white difference was reduced both in men and women; for women, the difference was reduced by more than half. These results suggest that it is important to establish a healthy lifestyle during young adulthood including regular exercise, limited or no alcohol drinking, and most important, avoidance of weight gain.

Acknowledgments
This study was supported by contracts N01-HC-48047, N01-HC-48048, N01-HC-48049, N01HC-48050, and N01-HC-95095 from the National Heart, Lung, and Blood Institute. The authors would like to thank Dr Philip Greenland for his helpful suggestions. Received May 16, 1995; revision received August 7, 1995; accepted August 14, 1995.

References
Top Abstract Introduction Methods Results Discussion References

1. Hypertension Detection and Follow-up Program Cooperative Group. Race, education and prevalence of hypertension. Am J Epidemiol. 1977;106:351-361. [Abstract/Free Full Text]

2. Hypertension prevalence and the status of awareness, treatment, and control in the United States. Hypertension. 1985;7:457-468. [Abstract/Free Full Text] 3. Blood pressure level in persons 18-74 years of age in 1976-80, and trends in blood pressure from 1960 to 1980 in the United States: data from the National Health Survey series 11, No. 234. US Department of Health and Human Services publication No. (PHS) 86-1684, 1986.

4. Boyle E, Griffey WP, Nichaman MZ, Talbert CR. An epidemiologic study of hypertension among racial groups of Charleston County, South Carolina: the Charleston Heart Study, phase II. In: Stamler J, Stamler R, Pullman TN, eds. The Epidemiology of Hypertension. New York, NY: Grune & Stratton; 1967:193-203. 5. Stamler J, Stamler R, Riedlinger WF, Algera G, Roberts RH. Hypertension screening of one million Americans: Community Hypertension Evaluation Clinic (CHEC) Program, 1973 through 1975. JAMA. 1976;235:2299-2306. [Abstract/Free Full Text] 6. Londe S, Gollub SW, Goldring D. Blood pressure in black and in white children. J Pediatr. 1977;90:93-95. [Medline] [Order article via Infotrieve] 7. Webber LS, Harsha DW, Phillips GT, Srinivasan SR, Simpson JW, Berenson GS. Cardiovascular risk factors in Hispanic, white, and black children: the Brooks County and Bogalusa Heart Studies. Am J Epidemiol. 1991;133:704-714. [Abstract/Free Full Text] 8. Harlan WR, Cornoni-Huntley J, Leaverton PE. Blood pressure in childhood: the National Health Examination Survey. Hypertension. 1979;1:559-565. [Abstract/Free Full Text] 9. Reed WL. Racial differences in blood pressure level of adolescents. Am J Public Health. 1981;71:1165-1167. [Abstract/Free Full Text] 10. Levinson S, Liu K, Stamler J, Stamler R, Whipple I, Ausbrook D, Berkson D. Ethnic differences in blood pressure and heart rate of Chicago school children. Am J Epidemiol. 1985;122:366-377. [Abstract/Free Full Text] 11. Morrison JA, Khoury P, Kelly K. Studies of blood pressure in school children (ages 6-19) and their parents in an integrated suburban school district. Am J Epidemiol. 1980;111:156-165. [Abstract/Free Full Text] 12. Baron AE, Freyer B, Fixler DE. Longitudinal blood pressures in blacks, whites, and Mexican Americans during adolescence and early adulthood. Am J Epidemiol. 1986;123:809-817. [Abstract/Free Full Text] 13. Manatunga AK, Jones JJ, Pratt JH. Longitudinal assessment of blood pressures in black and white children. Hypertension. 1993;22:84-89. [Abstract/Free Full Text] 14. Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR, Liu K, Savage P. CARDIA: study design, recruitment, and some characteristics of the examined subjects. J Clin Epidemiol. 1988;41:1105-1116. [Medline] [Order article via Infotrieve] 15. Liu K, Ballew C, Jacobs DR Jr, Sidney S, Savage PJ, Dyer A, Hughes G, Blanton MM, and the CARDIA Study Group. Ethnic differences in blood pressure, pulse rate, and related characteristics in young adults: the CARDIA Study. Hypertension. 1989;14:218-226. [Abstract]

16. Dyer AR, Cutter G, Liu K, Armstrong MA, Friedman GD, Hughes GH, Dolce JJ, Raczynski J, Burke G, Manolio T. Alcohol intake and blood pressure in young adults: the CARDIA Study. J Clin Epidemiol. 1990;43:1-13. [Medline] [Order article via Infotrieve] 17. Jacobs DR Jr, Hahn LP, Haskell WL, Pirie P, Sidney S. Validity and reliability of short physical activity history: CARDIA and the Minnesota Heart Health Program. J Cardiopulmonary Rehabil. 1989;9:448-459. 18. Cook WW, Medley DM. Proposed hostility and pharisiac-virtue scales for the MMPI. J Appl Psychol. 1954;38:414-418. 19. McDonald A, Van Horn L, Slattery M, Hilner J, Bragg C, Caan B, Jacobs DR Jr, Liu K, Hurbert H, Gernhoffer N, Betz E, Havlik D. The CARDIA dietary history: development and implementation. J Am Diet Assoc. 1991;91:1104-1112. [Medline] [Order article via Infotrieve] 20. Stamler R, Stamler J, Riedlinger WF, Algera G, Roberts RH. Weight and blood pressure: findings in hypertension screening of 1 million Americans. JAMA. 1978;240:1607-1610. [Abstract/Free Full Text] 21. Criqui MH, Wallace RB, Mishkel M, Barrett-Connor E, Heiss G. Alcohol consumption and blood pressure: the Lipid Research Clinics Prevalence Study. Hypertension. 1981;3:557-565. [Abstract/Free Full Text] 22. MacMahon SW. Alcohol consumption and hypertension. Hypertension. 1987;9:111-121. [Abstract/Free Full Text] 23. Benowitz NL, Sharp DS. Inverse relation between serum cotinine concentration and blood pressure in cigarette smokers. Circulation. 1989;80:1309-1312. [Abstract/Free Full Text] 24. Green MS, Jucha E, Luz Y. Blood pressure in smokers and nonsmokers: epidemiologic findings. Am Heart J. 1986;111:932-940. [Medline] [Order article via Infotrieve] 25. American College of Sports Medicine Position Stand. Physical activity, physical fitness, and hypertension. Med Sci Sports Exerc. 1993;25:1-10. [Medline] [Order article via Infotrieve] 26. McCarron DA, Morris CD, Cole C. Dietary calcium in human hypertension. Science. 1982;217:267-269. [Abstract/Free Full Text] 27. Cutler JA, Brittain E. Calcium and blood pressure. an epidemiologic prospective. Am J Hypertens. 1990;3:137S-146S. [Medline] [Order article via Infotrieve] 28. Khaw KT, Barrett-Conner E. Dietary potassium and blood pressure in a population. Am J Clin Nutr. 1984;39:963-968. [Abstract/Free Full Text]

29. The INTERSALT Co-operative Research Group. INTERSALT: an international study of electrolyte excretion and blood pressure: results for 24-hour urinary sodium and potassium. Br Med J. 1988;297:319-328. 30. Elliott P, Kesteloot H, Dyer A, Freeman J, Shipley M, Stamler J, Rose G, Marmot M, Stamler R. 24-Hour urinary nitrogen excretion and blood pressure: INTERSALT findings: abstract from the 64th Scientific Sessions of the American Heart Association, Anaheim, Calif, November 11-14, 1991. Circulation. 1991;84(suppl II):II-698. Abstract. 31. Liu K, Cedres LB, Stamler J, Dyer A, Stamler R, Nanas S, Berkson DM, Paul O, Lepper M, Lindberg HA, Marquardt J, Stevens E, Schoenberger JA, Shekelle RB, Collette P, Shekelle S, Garside D. Relationship of education to major risk factors and death from coronary heart disease, cardiovascular diseases and all causes: findings of three Chicago epidemiologic studies. Circulation. 1982;66:1308-1314. [Free Full Text] 32. Musante L, Treiber FA, Davis H, Strong WB, Levy M. Hostility: relationship to lifestyle behaviors and physical risk factors. Behav Med. 1992;18:21-26. [Medline] [Order article via Infotrieve] 33. Cutter GR, Burke GL, Dyer AR, Friedman GD, Hilner JE, Hughes GH, Hulley SB, Jacobs DR Jr, Liu K, Manolio TA, Oberman A, Perkins LL, Savage PF, Serwitz JR, Sidney S, Wagenknecht LE. Cardiovascular risk factor in young adults: the CARDIA baseline monograph. Control Clin Trials. 1991;12:1S-13S. [Medline] [Order article via Infotrieve] 34. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13-22. [Abstract/Free Full Text] 35. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121-130. [Medline] [Order article via Infotrieve] 36. Harburg W, Gleiberman L, Roeper P, Schork MA, Schull WJ. Skin color, ethnicity and blood pressure, I: Detroit blacks. Am J Public Health. 1978;68:1177-1183. [Abstract/Free Full Text] 37. Tyroler HA, James SA. Blood pressure and skin color. Am J Public Health. 1978;68:11701172. [Free Full Text] 38. Anderson NE, Myers HF, Pickering T, Jackson JS. Editorial Review: Hypertension in blacks: psychosocial and biological perspectives. J Hypertens. 1989;7:161-172.[Medline] [Order article via Infotrieve]