anger expression, body fat, and blood pressure in adolescents: project heartbeat!

Anger Expression, Body Fat, and Blood Pressure inAdolescents: Project HeartBeat!

WILLIAM H. MUELLER,* JO ANNE GRUNBAUM, AND DARWIN R. LABARTHEUniversity of Texas—Houston Health Science Center, School of Public Health, Houston, Texas

ABSTRACT This study explores the potential influence of growth, body/composition, and sexualmaturity on the relation of anger expression and blood pressure in adolescents. Baseline data fromProject HeartBeat! (82 boys and 85 girls, 14 years of age) examined the ability of anger expression(STAXI scale) to predict blood pressure, after controlling for the effects of ethnicity (African-American/non-African-American), height, weight, percentage body fat, and sexual maturity. Blood pressureswere unrelated to anger expression in models that included the above developmental variables.However, girls scoring high on healthy anger expression (“anger-control”) had significantly lowerlevels of percentage body fat (P 4 0.015) independent of the above factors. The literature suggeststhat body fat or body mass is often, though not unanimously, associated with unhealthy forms ofanger expression in adolescents. Research is required into the biological, social, and behavioral originsof the association between body fat and anger expression. Height and sexual maturity, virtuallyignored in this literature, should be included in future research. Am. J. Hum. Biol. 13:531–538,2001. © 2001 Wiley-Liss, Inc.

Anger is a normal emotion that one feelswhen one’s desires are unsatisfied. As a nor-mal emotion then, it is neither “bad” nor“good,” but, like hunger, it is a signal thatsomething needs attending. Anger in youthis associated with two health problems“Free-floating anger” can lead to acts of ag-gression and injury in youth who are disaf-fected from society (Canada, 1995; Pro-therow-Stith, 1995) and chronic anger canlead to habitual stimulation of the sympa-thetic–adrenal–medullary system leadingto deterioration of cardiovascular health(Taylor et al., 1997). This paper deals withthe latter, although the two are related(Johnson, 1998).

There are three ways chronic anger canbecome problematic and ultimately affecthealth: (1) There may be an inability to re-spond in a healthy manner when anger isfelt. Habitually, anger is either suppressedor expressed in explosive outbursts (Bare-foot and Lipkus, 1994). As a corollary, angermay be experienced as the only emotion feltor else may be denied as never experienced(Jenkins et al., 1978). Suppressed anger andthe experience of frequent angry feelingshave been associated with cardiovasculardisease (Williams and Williams, 1993). (2)There may be an inability to discern thesituations in which it is helpful to expressanger and when it is not helpful, or to seekless aggressive ways to resolve conflict. Ri-gidity in expressing anger has been related

to elevated lipids (Engebretson and Stoney,1995). (3) There may be negative and mis-trustful beliefs regarding the motivations ofothers (“cynical mistrust”), often accompa-nied by chronic anger and aggression (Costaet al., 1986). These are essential elements ofthe construct “hostility,” a precursor of car-diovascular disease (Williams and Williams,1993).

On balance, the research shows that an-ger and hostility are important predictors ofcardiovascular health of adults (Siegman,1993; Helmers et al., 1994; Barefoot et al.,1995; Miller et al., 1996). Associations be-tween anger and potential for hostility havealso been linked to unfavorable levels of car-diovascular risk variables in children andadolescents (Grunbaum et al., 1997b).These associations are less marked andmore variable in studies of youths than theyare in studies of adults (Mueller et al.,1998). One reason, perhaps, is that adoles-cents vary significantly in physical and sex-

Contract grant sponsor: National Heart, Lung and Blood In-stitute; Contract grant number: Cooperative Agreement U01-HL-41166; Contract grant sponsor: Centers for Disease Controland Prevention, through the Southwest Center for PreventionResearch; Contract grant number: U48/CCU609653; Contractgrant sponsor: Compaq Computer Corporation.

*Correspondence to: William H. Mueller, Ph.D., University ofTexas—Houston Health Science Center, School of Public Health,P.O. Box 20186, Houston TX 77225.E-mail: [email protected]

Received 12 July 2000; Revision received 12 January 2001;Accepted 16 January 2001

AMERICAN JOURNAL OF HUMAN BIOLOGY 13:531–538 (2001)

© 2001 Wiley-Liss, Inc.

PROD #M20053R

ual maturity and in body composition. At noother time in life would such body size andcomposition differences potentially con-found the association of emotions andhealth. Yet little has been said about this inadolescents (Johnson, 1990). In previous re-search, associations have been noted be-tween aspects of anger expression and bodymass in adolescents, and body fat or weighthave affected the relationship between an-ger and blood pressure (Johnson, 1990;Mueller et al., 1998). The current study ex-plores the effects of growth variables onthe relationships between anger expressionand blood pressure in adolescents from Proj-ect HeartBeat! and the adequacy withwhich these variables have been studied inthe literature on adolescent cardiovascularhealth.

METHODS

Project HeartBeat! is a mixed longitudi-nal study of the development of blood pres-sure, cardiac structure and function, bloodlipids, and health behaviors in children8–18 years old residing in a large suburbanarea of southeast Texas (Labarthe et al.,1997). About 20% of the sample are African-American; the rest are mostly European-American, with a very small percentage ofother ethnic groups (Grunbaum et al.,1997a). Subjects were recruited from twoneighboring communities with the assis-tance of local health departments, schooldistricts, churches, and several communityagencies. A field center was established in1991, and data collection began that year.Children were initially enrolled in three co-horts at a baseline age of either 8, 11, or 14years of age. The data presented here arecross-sectional from the baseline cohort. Ini-tially, anger expression was assessed onlyin 14-year olds. The data for analysis thusconsists of 82 boys and 85 girls 14 years ofage.

The State-Trait Anger Expression Inven-tory (STAXI) is a self-report instrumentused to assess components of anger expres-sion and frequency of angry feelings (Spiel-berger et al., 1985). “State anger” assessesangry feelings in the present time. “Traitanger” measures the frequency with whicha subject experiences angry feelings daily(Spielberger et al., 1983). The variable “an-ger expression” is a summary constructcomprising three latent variables: “anger-in” is the tendency to withhold expression of

angry feelings; “anger-out” is the habit ofletting people know one is angry, includingboth verbal and nonverbal actions; and “an-ger-control” is the tendency to handle angryfeelings in a reflective or thoughtful way,such as forgiving others and not resorting toexcessive self-blame for misunderstandings.The scale has acceptable reliability and va-lidity and is widely used with adolescents(Spielberger et al., 1988; Johnson, 1989).The anger assessments were made close to,but not simultaneously with, the blood pres-sure measurements. Therefore the vari-ables that reflect the “usual” manner of ex-pressing or experiencing angry feelings aremost relevant for study purposes (trait an-ger, anger-in, anger-out, and anger-control).

Resting blood pressure was measured bytrained and certified staff with a mercurysphygmomanometer (Baumanometer), fol-lowing standard procedures and using cuffsizes based on arm circumference. Tworeadings of each blood pressure were takenon three different days (about one weekapart), and the average of all valid mea-sures was used to characterize a child’sblood pressure. Four growth variablesknown to be related to blood pressure inyouth were selected: height; weight; thebody mass index (BMI 4 weight (kg)/heightsquared (m2)); and percent body fat (PBF)estimated from bioelectrical impedance us-ing the cross-validated equations of Guo etal. (1989). The equations include arm cir-cumference and skinfold thicknesses. Armcircumference was obtained with a plasticnonstretch tape, and skinfolds were mea-sured with a Holtain caliper. Body measure-ments were obtained following standardprocedures (Lohman et al., 1988). Sexualmaturity was rated using breast stages ingirls and genital stages in boys duringphysical examination (Tanner, 1989). Proj-ect HeartBeat! children are similar in bodysize and composition to national referencevalues (Mueller et al., 1999).

AnalysisAll analyses were done with SPSS in a

sexes-stratified design. The level for statis-tical significance was P < 0.05. Descriptivestatistics were computed for all variables bygender and compared by t-test. Pearson cor-relations of blood pressure, body measure-ments, and anger variables were computed.Post hoc regression of blood pressure (sys-tolic and diastolic-phase 5) on anger expres-

532 W.H. MUELLER ET AL.

sion, controlled for the potential confoundingeffects of ethnicity (non-African-Americanand “other” 4 0, African-American 4 1),sexual maturity, and height and weight, orPBF.

By post hoc is meant that variables wereincluded in the models that seemed to bepredictors on inspection of the initial corre-lations and that made biological sense. In-teractions between anger expression andbody size, composition, maturity and ethnic-ity were also included to examine potentialmoderating effects of these variables (Baronand Kenny, 1986). Some anger expressionvariables were correlated initially with bodymeasurements, and these relations werealso explored through regression analysis,controlling for potential confounding effectsof sexual maturity and ethnicity (Mueller etal., 1996). Possible nonlinear associations ofblood pressure and anger expression vari-ables such as those found by Spielberger etal. (1988) were explored by the “curve fit”procedure in SPSS. None were found.

RESULTS

Descriptive analysis

Means and standard deviations of riskfactors are presented in Table 1 for boys andgirls. Boys are significantly taller than girlsand have higher systolic blood pressures (P< 0.01). Girls have a larger BMI and per-centage body fat than boys (P < 0.01). Boysand girls do not differ, on average, for dia-stolic blood pressures or body weight. Traitanger is the only anger measure that differssignificantly between the sexes, and it ishigher in boys than girls (P < 0.05).

Correlates of blood pressure

Correlations among blood pressures, bodymeasurements, anger variables and demo-graphics are shown in Table 2. Systolicblood pressure is mainly dependent onweight, height, and BMI in both sexes. Di-astolic blood pressure is negatively associ-ated with body fat in boys and positivelyassociated with weight and height in girls.Sexual maturity and ethnicity are not re-lated to blood pressure (not shown in table).However, sexual maturity is correlated withweight, BMI, and percentage body fat inboth boys and girls. Earlier maturation isassociated with less fat in boys and more fatin girls. Earlier maturing boys are tallerthan later maturing boys. There are few sig-nificant associations of ethnicity with othervariables. Most notably, African Americanboys have less, and girls have more relativefatness (PBF) than non-African Americans.

There are nine correlations of anger ex-pression with blood pressure, ethnicity andphysical development that are statisticallysignificant (P < 0.05) out of a possible 96(Table 2). None of the relevant anger vari-ables are correlated with blood pressures.However, “trait anger” and “anger-out” arenegatively associated with height and Afri-can American ethnicity is associated withhigher levels of “anger-out.” In girls, per-centage body fat is associated with lowerscores on “anger control”; African Americangirls have lower trait anger scores.

Relationships amonganger-expression variables

Correlations among anger-expressionvariables are shown in Table 3. “Anger con-trol” tends to have low to moderate negativeassociations with other anger variables,particularly “anger-out” and “trait” anger.“Anger-in” and “anger-out” are positively re-lated to “trait” anger.

Regression models

Regression of blood pressures on sexualmaturity, ethnicity, body measurements,and anger expression variables are shownin Table 4. Weight in girls and a combina-tion of body size and maturity in boys ac-count for 23% to 30% of the variation in sys-tolic blood pressure (P < 0.01) (top, Table 4).None of the anger expression variables aresignificant predictors of systolic blood pres-sure. Sexual maturity and ethnicity also are

TABLE 1. Means and standard deviations of riskfactors by gender: Project HeartBeat! 14-year olds

Variable

Boys(n 4 79–82)

Girls(n 4 78–85)

P valueMean SD Mean SDSystolic BP,

mmHg 111.5 8.09 106.0 8.43 .00Diastolic BP-5,

mmHg 59.0 9.81 60.5 7.85 .29BMI, kg/m2 20.7 3.22 22.6 4.86 .00Percentage

body fat 18.0 7.07 27.6 6.47 .00Weight (kg) 58.0 11.56 59.9 13.94 .35Height (cm) 166.8 7.08 162.6 5.51 .00Anger-in 15.4 4.62 16.0 4.54 .38Anger-out 16.1 4.28 17.0 4.90 .20Anger-control 20.1 4.95 20.2 5.52 .91Trait anger 20.3 6.34 18.5 5.33 .05

ANGER EXPRESSION IN ADOLESCENTS 533

not significant predictors of systolic bloodpressure. Diastolic blood pressure is signifi-cantly predicted in a model that includesethnicity, sexual maturity, percentage bodyfat, “state-anger” and “anger-control” inboys (P < 0.01) (bottom, Table 4). The modelexplains 24% of the variance in diastolicblood pressure of boys. However, diastolicblood pressure is not predicted by any of theindependent variables in girls.

Regression of percentage body fat (PBF)on anger expression in girls shows a nega-tive association of PBF with “anger-control”(P < 0.05), after controlling for ethnicity,sexual maturity, “state-anger,” and “anger-in” (Table 5). In the model, 30% of the varia-tion in PBF is accounted for by the abovevariables (P < 0.01). PBF could also be con-sidered a predictor variable for “anger-

control.” In such a model, PBF is a highlysignificant predictor of “anger-control” (P <0.01), independent of sexual maturity, eth-nicity, and their interactions with body fat(bottom, Table 5).

DISCUSSION

Blood pressures in this sample of 14-year-old children from Project HeartBeat! weremore dependent on body size and composi-tion and sexual maturity than on the expe-rience or expression of anger (Table 4).There was an association of anger expres-sion (“anger-control”) with body fat indepen-dent of potential confounding variables(Table 5).

In a previous study there was a positiveassociation of body fat with “anger-in” and“hostility” in girls (Mueller et al., 1998). In

TABLE 2. Intercorrelations among anger expression variables, blood pressures, body measurements, anddemographic variables: Project HeartBeat!*

Blood pressurebody meas.

Anger expression variables Body measurementsState

aTrait

a Anger-in Anger-outAnger

conAnger

exp PBF BMI Weight HeightBoys (n 4 71–82)

SBPa −0.06 −0.03 −0.05 −0.09 −0.09 −0.03 0.05 0.32** 0.42** 0.45**DBP5b 0.26 0.09 −0.10 0.09 −0.21 −0.26 −0.26 −0.09 −0.09 −0.02PBFc 0.05 0.01 0.03 −0.09 0.05 −0.05 0.64** 0.54** 0.09BMId −0.04 0.01 0.05 −0.05 −0.06 0.03 0.92** 0.33**Weight −0.09 −0.08 0.01 −0.14 −0.07 −0.03 0.67**Height −0.12 −0.24 −0.10 −0.29 −0.08 −0.16Genital stage 0.00 0.15 0.10 0.15 −0.22 0.22 −0.33** 0.13 0.30 0.44**Ethnicitye −0.01 0.15 0.13 0.24 0.12 0.11 −0.30** −0.20 −0.21 −0.14

Girls (n 4 71–85)SBPa −0.17 0.08 0.07 0.05 −0.12 0.12 0.07 0.41** 0.47** 0.24DBP5b −0.09 0.04 −0.11 0.06 −0.08 0.02 −0.06 0.18 0.24** 0.26PBFc 0.23 0.15 0.15 0.13 −0.27 0.25 0.83** 0.74** −0.11BMId 0.11 0.19 0.21 0.10 −0.14 0.20 0.96** 0.05Weight 0.10 0.14 0.21 0.07 −0.12 0.18 0.33**Height −0.03 −0.11 0.04 −0.08 0.06 −0.06Breast stage −0.11 0.10 0.03 −0.01 −0.07 0.04 0.43** 0.45** 0.43** 0.05Ethnicitye 0.21 −0.22 −0.07 −0.15 0.03 −0.11 0.25 0.41** 0.35** −0.13

*Bold text signifies P < 0.05.**P < 0.01.aSBP 4 Systolic Blood PressurebDBP5 4 Diastolic Blood PressurecPBF 4 Percent Body Fat from bioelectrical impedance and body measurementsdBMI 4 Body Mass IndexeEthnicity: African-American 4 1, Non-African American 4 0

TABLE 3. Intercorrelation among anger expression variablesa,*

State a Trait a Anger-in Anger-out Anger con Anger expState a — 0.36** 0.24 0.15 −0.06 0.22Trait a 0.31** — 0.60** 0.69** −0.29** 0.75**Anger-in 0.24 0.45** — 0.45** −0.04 0.67**Anger-out 0.23 0.67** 0.21 — −0.30** 0.80**Anger-con −0.05 −0.40** −0.09 −0.46** — −0.65**Anger-exp 0.23 0.70** 0.57** 0.78** −0.77** —

*Bold text signifies P < 0.05.**P < 0.01.aBoys (n 4 78) above and girls (n 4 81) below diagonal.


Project HeartBeat!, percentage body fat wasinversely related to “anger-control” in girlsin a model which included ethnicity, sexualmaturity, and other facets of anger expres-sion. In both instances these relationshipswere strengthened when possible confound-ing variables were held constant (ethnicity,sexual maturity). If “anger-in” can be con-sidered a generally unhelpful manner of ex-pressing anger and “anger-control” ahealthier one, the two studies are consistentin showing an association of higher levels ofbody fat with less healthy modes of angerexpression.

At least 26 reports that included anger orhostility constructs in relation to blood pres-sure or other cardiovascular disease risk

factors in children have appeared since1981 (Siegel and Leitch 1981; Hunter et al.,1982; Siegel et al., 1983; Siegel, 1984; Ea-gelston et al., 1986; Matthews et al., 1986;Murray et al., 1986; Southard et al., 1986;Weidner et al., 1986; Johnson et al.,1987a,b; Hayman et al., 1988; McCann andMatthews, 1988; Treiber et al., 1989;Johnson, 1989, 1990; Raikkonen and Kelti-kangas-Jarvinen, 1991; Engebretson andMatthews, 1992; Vogele and Steptoe, 1993;Ewart and Kolodner, 1994; Groer et al.,1994; Kuczmarski et al., 1994; Ravaja et al.,1996; Siegman and Snow, 1997; Vogele etal., 1997; Mueller et al., 1998). Four of thereports included no measures of physical de-velopment (Eagelston et al., 1986; Southardet al., 1986; Engebretson and Matthews,1992; Siegman and Snow, 1997). Only twostudies included sexual maturity, but onlyin one was it utilized in the analysis (Eagel-ston et al., 1986; Mueller et al., 1998). Of the22 studies with body measurements, 16 con-trolled for the effects of body size in somemanner either in multivariate or categoricalanalyses (Siegel and Leitch, 1981; Hunter etal., 1982; Siegel et al., 1983; Siegel, 1984;Matthews et al., 1986; Murray et al., 1986;Weidner et al., 1986; Johnson et al., 1987a,b; Hayman et al., 1988; Treiber et al.,1989; Johnson, 1990; Ewart and Kolodner,1994; Ravaja et al., 1996; Vogele et al., 1997;Mueller et al., 1998). In all but four studies,the BMI was the only developmental vari-able considered.

TABLE 4. Regression of blood pressure on sexual maturity, ethnicity, body measurements, and anger expressionvariables: Project HeartBeat!

Model forBoys (n 4 68–71) Girls (n 4 71)

Estimate P value Estimate P valueSystolic blood pressure

Constant 59.210 0.025 98.631 0.000Maturitya 0.465 0.662 −1.743 0.633Ethnicityb −4.168 0.174 0.655 0.809Weight 0.170 0.137 0.356 0.000Height 0.247 0.185 −0.038 0.163DR2 0.226 0.002 0.302 0.000

Diastolic blood pressureConstant 62.810 0.000 25.715 0.335Maturitya −1.893 0.131 0.019 0.988Ethnicityb 0.717 0.851 1.294 0.636State anger 1.727 0.007 — —Anger-control −0.376 0.094 — —% Body fat −0.415 0.016 — —Weight — — 0.110 0.169Height — — 0.170 0.313DR2 0.240 0.004 0.096 0.166

aMaturity 4 genital stage in boys, breast stage in girls.bEthnicity: African-American 4 1; non-African-American 4 0.

TABLE 5. Regression analysis of the relationship ofanger expression and body fat: adolescent girls;

Project HeartBeat!

Model for: Estimate P valuePercent body fat (n 4 63)

Constant 17.761 0.008Maturitya 3.317 0.001Ethnicityb 4.414 0.102Anger-control −0.339 0.015Anger-in 0.139 0.416State anger −0.008 0.710DR2 0.309 0.001

Anger control (n 4 63)Constant 25.666 0.000Maturitya 0.695 0.455Ethnicityb 5.086 0.027% Body fat −0.315 0.008DR2 0.161 0.015

aMaturity 4 breast stage.bEthnicity 4 African-American 4 1, non-African-American 4 0.


Body mass was significantly related insome way to anger expression or hostility in11 of the 22 studies that included body mea-surements (Siegel and Leitch, 1981; Siegelet al., 1983; Siegel, 1984; Matthews et al.,1986; Murray et al., 1986; Weidner et al.,1986; Johnson et al., 1987b; Treiber et al.,1989; Johnson, 1990; Ravaja et al., 1996;Mueller et al., 1998). These relationshipswere not necessarily evident in every sub-group of a study. For example, in two inves-tigations, hostility was linked to body massin one sex but not in the other (Johnson,1990; Mueller et al., 1998). In another,whether body measurements or the emo-tions were primary in predicting blood pres-sure, was studied as a function of ethnicity(Johnson et al., 1987b).

In the 16 reports for which body measure-ments were accounted in analysis, the rela-tion of anger/hostility to blood pressure wassignificantly reduced or changed in 6 ofthem (Siegel and Leitch, 1981; Siegel et al.,1983; Matthews et al., 1986; Murray et al.,1986; Treiber et al., 1989; Mueller et al.,1998), and remained unchanged in 8(Hunter et al., 1982; Siegel, 1984; Johnsonet al., 1987a,b; Hayman et al., 1988;Johnson, 1990; Ewart and Kolodner, 1994;Vogele et al., 1997). In the remaining twostudies, anger appeared to be more relatedto body mass and lipids than it did to bloodpressure (Weidner et al., 1986; Ravaja et al.,1996).

One-half of the 22 reports to date havethus found a positive association of bodymass with negative aspects of anger expres-sion, and in some cases this has altered theassociation of blood pressure and anger ex-pression. Few studies measured “body fat”per se. However, the BMI has a moderaterelation to body fat in children as measuredby body density from underwater weighing(Roche et al., 1981). The findings from Proj-ect HeartBeat! and from the literature sug-gest an association of body fat and someforms of anger expression or hostility amongsome groups of children.

Sexual maturity, which signals the initia-tion of puberty and its associated changes inbody size and body composition (Malina andBouchard, 1991), has been little studied inthe literature on blood pressure and angerexpression. Changes in anger expressionwith age or sexual maturity have also notbeen addressed (Liehr et al., 1997; Mein-inger et al., 1998). Some studies of blood

pressure in childhood are now investigatingthe potential importance of sexual maturity(Kozinetz, 1991; Cho et al., 2001). It is atleast as important as body size (Daniels etal., 1996), and blood pressure studies inchildren have long considered expressingnormal variability in blood pressure as afunction of height (Rosner et al., 1993). It isthus a puzzle as to why so few studies of theeffects of the emotions on blood pressure inchildhood should ignore height and sexualmaturity. BMI and height are only moder-ately correlated in youth (Cole, 1986) so thatthere would be no reason not to includethem together as independent predictors ofcardiovascular health in children.

Biopsychosocial models

Models of the pathways linking hostilityand cardiovascular health provide clues asto why there might be associations of angerand hostility with body fat and blood pres-sure (Smith, 1992). Although most of the re-search is based on adults, the data could beuseful in understanding the situation inchildren and adolescents as well since bio-logical and behavioral risk factors forchronic disease have their origins in child-hood (Labarthe, 1998).

Research using the health behavior modelhas shown that hostile adults frequently ex-hibit poorer health habits in general. Theysmoke more and use more alcohol and caf-feine; they often have higher BMI’s andhigher caloric intakes than less hostile per-sons; and they tend to devote less leisuretime to physical activity (Smith, 1992). Re-search using the psychosocial vulnerabilitymodel has consistently shown that personswho score high on hostility have higher lev-els of interpersonal conflict and less socialsupport. This is further supported from re-search guided by the psychophysiologicalvulnerability model, which shows that hos-tile people have greater cardiovascular re-activity while performing tasks in whichthey must interact with others (Smith,1992). Such persons, one could speculate,might find less comfort from having friendsthan from having a relationship with televi-sion programs, books, or food, things thatencourage less physical movement andhigher caloric intake. The above models canhelp us understand why an attitude, such ashostility with its emphasis on angry feelingsand aggression, might be associated with


factors conductive to positive energy bal-ance in adults and children.

It is also possible that being obese or over-weight could influence the way teens ex-press anger. Obese persons may be angryabout being overweight in a society thatsees “thin” as ideal. People who havestruggled with being overweight most oftheir lives have expressed the attitudes ofother Americans toward their condition asthe last acceptable bigotry. Chronic angerfelt as a result of this prejudice might leadto frequent angry feelings and to unhelpfulways of expressing anger. Thus, obese oroverweight individuals might score high on“trait anger” or “anger-in” as they struggleto hide their true feelings for fear of furtherrejection. Having frequent angry feelingsand being unable to express them are char-acteristics associated with poorer cardiovas-cular health (Williams and Williams, 1993).Smith (1992) does not discuss sociologicalmodels of hostility and health. However,such structural/societal models could alsobe critical to the understanding of thehealth consequences of anger and hostilityin the population. At least two large studieshave found associations of hostility withlower social class, non-European-Americanethnicity, male sex, and less education(Williams and Williams, 1993).

Research is required into the biological,social, and behavioral origins of the associa-tion between body fat and anger expression.Height and sexual maturity, virtually ig-nored in the literature on the emotions andadolescent cardiovascular health, should beincluded in future research.

LITERATURE CITED

Barefoot JC, Lipkus IM. 1994. The assessment of angerand hostility. In: Siegman AW, Smith WT, editors.Anger, hostility and the heart. Hillsdale, NJ:Lawrence Erlbaum Associates. p 43–66.

Barefoot JC, Larsen S, von der Leith L, Schroll M. 1995.Hostility and incidence of acute myocardial infarctionand mortality in a sample of older Danish men andwomen. Am J Epidemiol 142:477–484.

Baron RM, Kenny DA. 1986. The moderator–mediatorvariable distinction in social psychological research:conceptual, strategic and statistical considerations. JPers Soc Psychol 51:1173–1182.

Canada G. 1995. Fist, stick, knife, gun. A personal his-tory of violence in America. Boston: Beacon Press.

Cho SD, Mueller WH, Meininger JC, Liehr P, Chan W.2001. Blood pressure and sexual maturity in adoles-cents: The Heartfelt Study. Am J Hum Biol 13:227–234.

Cole TI. 1986. Weight/heightp compared to weight/height2 for assessing adiposity in childhood: influence

of age and bone age on p during puberty. Ann HumBiol 13:433–451.

Costa PT, Zonderman AB, McCrae RR, Williams RB.1986. Cynicism and paranoid alienation in the Cook-Medley hostility scale. Psychosom Med 48:283–285.

Daniels SR, Obarzanek E, Barton BA, Kimm SY, Si-milio SL, Morrison JA. 1996. Sexual maturation andracial differences in blood pressure in girls: The Na-tional Heart, Lung and Blood Institute Growth andHealth Study. J Pediatr 129:208–213.

Eagleston JR, Kirmil-Gray K, Thoresen CE, WeidenfeldSA, Bracke P, Heft L, Arnow B. 1986. Physical healthcorrelates of Type-A behavior pattern in children. JBehav Med 9:341–362.

Engebretson TO, Matthews KA. 1992. Dimensions ofhostility in men, women and boys: relationships topersonality and cardiovascular response to stress.Psychosom Med 54:311–323.

Engebretson TO, Stoney CM 1995. Anger expressionand lipid concentrations. Int J Behav Med 2:281–298.

Ewart CK, Kolodner KB. 1994. Negative affect, genderand expressive style predict elevated ambulatoryblood pressure in adolescents. J Pers Soc Psychol 66:596–605.

Groer M Thomas S, Droppleman P, Younger MS. 1994.Longitudinal study of adolescent blood pressures,health habits, stress and anger. Health Values 18:25–33.

Grunbaum JA, Labarthe DR, Ayars C, Harrist R,Nichaman MZ. 1997a. Recruitment and enrollmentfor Project HeartBeat!: achieving the goals of minorityinclusion. Ethn Dis 6:203–212.

Grunbaum JA, Vernon SW, Clasen CM. 1997b. The as-sociation between anger and hostility and risk factorsfor coronary heart disease in children and adoles-cents: a review. Ann Behav Med 19:179–189.

Guo S, Roche AF, Houtkooper L. 1989. Fat-free mass inchildren and young adults predicted from bioelectricalimpedance and anthropometric variables. Am J ClinNutr 50:435–443.

Hayman LL, Meininger JC, Stashinko EE, GallagherPR, Coates PM. 1988. Type-A behavior and physi-ological cardiovascular risk factors in school-agedtwin children. Nurs Res 37:290–296.

Helmers KF, Posluszny DM, Krantz DS. 1994. Associa-tion of hostility and coronary artery disease: a reviewof studies. In: Siegman AW, Smith TW, editors. An-ger, hostility and the heart. Hillsdale, NJ: LawrenceErlbaum Associates. p 67–96.

Hunter SM, Wolf TM, Sklov MC. 1982. Type-A coro-nary-prone behavior pattern and cardiovascular riskfactor variables in children and adolescents: The Bo-galusa Heart Study. J Chron Dis 35:613–621.

Jenkins CD, Zyzanski SJ, Rosenman RH. 1978. Coro-nary prone behavior: one pattern or several? Psycho-som Med 40:25–43.

Johnson EH, Schork NJ, Spielberger CD. 1987a. Emo-tional and familial determinants of elevated bloodpressure in black and white adolescent females. JPsychosom Res 31:731–741.

Johnson EH, Spielberger CD, Worden TJ, Jacobs GA.1987b. Emotional and familial determinants of el-evated blood pressure in black and white adolescentmales. J Psychosom Res 31:287–300.

Johnson EH. 1989. The role of the experience and ex-pression of anger and anxiety in elevated blood pres-sure among black and white adolescents. J Nat MedAssoc 81:573–584.

Johnson EH. 1990. Interrelationships between psycho-logical factors, overweight and blood pressure in ado-lescents. J Adolesc Health Care 11:310–318.


Johnson EH. 1998. Brothers on the mend. Understand-ing and healing anger for African-American men andwomen. New York: Pocket Books.

Kozinetz CA. 1991. Sexual maturation and blood pres-sure levels of a biracial sample of girls. Am J DisChild 145:142–147.

Kuczmarski RJ, Anderson JJB, Koch GG. 1994. Corre-lates of blood pressure in seventh day adventist (SDA)and non-SDA adolescents. J Am Coll Nutr 13:165–173.

Labarthe DR, Nichaman MZ, Harrist RB, GrunbaumJA, Dai S. 1997. The development of cardiovascularrisk factors from age 8–18 in Project HeartBeat!:study design and patterns of change in total choles-terol concentration. Circulation 95:2636–2642.

Labarthe DR. 1998. Epidemiology and prevention ofcardiovascular diseases: a global challenge. Gaithers-burg, MD: Aspen Publishers.

Liehr P, Meininger JC, Mueller WH, Chandler SP,Chan W. 1997. Blood pressure reactivity in urbanyouth during angry and normal talking. J CardiovascNurs 11:85–94.

Lohman TG, Roche AF, Martorell R (editors). 1988. An-thropometric standardization reference manual.Champaign IL: Human Kinetics.

Malina RM, Bouchard C. 1991. Growth, matruation andphysical activity. Champaign, IL: Human Kinetics.

Matthews KA, Manuck SB, Saab PG. 1986. Cardiovas-cular responses of adolescents during a naturally oc-curring stressor and their behavioral and psycho-physiological predictors. Psychophysiology 23:198–209.

McCann BS, Matthews KA. 1988. Influences of poten-tial for hostility, Type-A behavior and parental his-tory of hypertension on adolescents’ cardiovascularresponses during stress. Psychophysiology 25:503–511.

Meininger JC, Liehr P, Mueller WH, Chan W, ChandlerSP. 1998. Predictors of ambulatory blood pressure:Identification of high risk adolescents. Adv Nurs Sci20:51–65.

Miller TO, Smith TW, Turner CW, Guijarro ML, HalletAJ. 1996. A meta-analytic review of research on hos-tility and physical health. Psychol Bull 119:322–348.

Mueller WH, Meininger JC, Liehr P, Chan W, ChandlerSP. 1996. Demographic moderators of biological vari-ables: lessons from a pilot project of anger and bloodpressure. Riv Anthropol (Roma) 74:139–146.

Mueller WH, Meininger JC, Liehr P, Chandler SP,Chan W. 1998. Adolescent blood pressure, anger ex-pression and hostility: possible links with body fat.Ann Hum Biol 25:295–307.

Mueller WH, Harrist RB, Doyle SR, Ayars CL, LabartheDR. 1999. Body measurement variability, fatness andfat-free mass in children 8, 11 and 14 years of age:Project HeartBeat! Am J Hum Biol 11:69–78.

Murray DM, Matthews KA, Blake SM, Prineas RJ, Gil-lum RF. 1986. Type-A behavior pattern in children:demographic, behavioral and physiologic correlates.Health Psychol 5:159–169.

Prothrow-Stith DB. 1995. The epidemic of youth vio-lence in America: using public health strategies toprevent violence. J Health Care Poor Underserved 6:95–101.

Raikkonen K, Keltikangas-Jarvinen L. 1991. Hostilityand its association with behaviorally induced and so-matic coronary risk indicators in Finnish adolescentsand young adults. Soc Sci Med 33:1171–1178.

Ravaja N, Ketikangas-Jarvinen L, Keskivaara P. 1996.Type-A factors as predictors of changes in the meta-bolic syndrome precursors in adolescents and young

adults—a three-year follow-up study. Health Psychol15:18–29.

Roche AF, Siervogel RM, Chumlea WC, Webb P. 1981.Grading body fatness from limited anthropometricdata. Am J Clin Nutr 34:2831–2838.

Rosner B, Prineas RJ, Loggie J, Daniels SR. 1993. Bloodpressure nomograms for children and adolescents byheight, sex and age for children in the United States.J Pediatr 123:871–886.

Siegel JM, Leitch CJ. 1981. Behavioral factors andblood pressure in adolescence: The Tacoma Study. AmJ Epidemiol 113:171–181.

Siegel JM, Matthews KA, Leitch CJ. 1983. Blood pres-sure variability and the Type-A behavior pattern inadolescence. J Psychosom Res 27:265–272.

Siegel JM. 1984. Anger and cardiovascular risk in ado-lescents. Health Psychol 3:293–313.

Siegman AW. 1993. Cardiovascular consequences of ex-pressing, experiencing and repressing anger. J BehavMed 16:539–569.

Siegman AW, Snow SC. 1997. The outward expressionof anger, the inward experience of anger and CVR: therole of vocal expression. J Behav Med 20:29–45.

Smith TW. 1992. Hostility and health: current status ofa psychosomatic hypothesis. Health Psychol 11:139–150.

Southard DR, Coates TJ, Kolodner K, Parker FC,Padgett NE, Kennedy HL. 1986. Relationship be-tween mood and blood pressure in the natural envi-ronment: an adolescent population. Health Psychol5:469–480.

Spielberger CD, Jacobs G, Russell S, Crane RS. 1983.Assessment of anger: The State-Trait Anger Scale. In:Butcher JN, Spielberger CD, editors. Advances in per-sonality assessment, Vol 2. Hillsdale, NJ: LawrenceErlbaum Associates. p 159–187.

Spielberger CD, Johnson EH, Russell SF, Crane RJ,Jacobs GA, Worden TJ. 1985. The experience and ex-pression of anger: construction and validation of ananger expression scale. In: Chesney MA, andRosenma RH, editors. Anger and hostility in cardio-vascular and behavioral disorders. New York: Hemi-sphere/McGraw-Hill. p 5–30.

Spielberger CD, Krasner SS, Solomon EP. 1988. Theexperience, expression and control of anger. In: Ja-nisse MP, editor. Health psychology: individual differ-ences and stress. New York: Springer Verlag. p 89–108.

Tanner JM. 1989. Foetus Into Man. Physical growthfrom conception to maturity. Cambridge, MA: Har-vard University Press.

Taylor SE, Repetti RL, Seeman T. 1997. Health psychol-ogy: what is an unhealthy environment and how doesit get under the skin? Ann Rev Psychol 48:411–447.

Treiber FA, Musante L, Riley W, Mabe A, Carr T, LevyM, Strong WB. 1989. The relationship between hos-tility and blood pressure in children. Behav Med 15:173–178.

Vogele C, Jarvis A, Cheeseman K. 1997. Anger supres-sion, reactivity and hypertension risk: gender makesa difference. Ann Behav Med 19:61–69.

Vogele C, Steptoe A. 1993. Anger inhibition and familyhistory as modulators of cardiovascular responses tomental stress in adolescent boys. J Psychosom Res37:503–514.

Weidner G, McLellan R, Sexton, Istvan J, Conner S.1986. Type-A behavior and physiologic coronary riskfactors in children of the Family Heart Study: resultsfrom a 1-year follow up. Psychosom Med 48:480–488.

Williams RB, Williams, V. 1993. Anger kills. Seventeenstrategies for controlling the hostility that can harmyour health. New York: Harper Collins Publishers.


anger expression, body fat, and blood pressure in adolescents: project heartbeat!

Documents