Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 May 1.
Published in final edited form as: Int J Psychophysiol. 2012 Feb 11;84(2):172–179. doi: 10.1016/j.ijpsycho.2012.02.003

Hostile behavior links negative childhood family relationships to heart rate reactivity and recovery in young adulthood

Linda J Luecken a,1, Danielle S Roubinov a
PMCID: PMC3340447  NIHMSID: NIHMS356715  PMID: 22331058

Abstract

Prior research has demonstrated that vulnerability to stress is influenced by early life experiences. This study evaluates the impact of negative childhood family relationships on cardiovascular stress reactivity in young adulthood. Participants (age 18–22) from families characterized by negative (n = 39) or positive relationships (n = 36) engaged in a role-play conflict task. Hostile/aggressive verbal behaviors during the task were observed, and blood pressure (BP) and heart rate (HR) responses were measured before, during, and after the task. Participants from negative families engaged in more hostile/aggressive verbal behavior during the task and showed attenuated HR reactivity. Hostile/aggressive verbal behavior predicted attenuated HR reactivity and recovery. Path analyses linked negative family relationships to more hostile verbal behavior during the task, and attenuated HR reactivity and recovery. These results support the development of hostile/aggressive behavior in social situations as a pathway linking childhood adversity to stress vulnerability across the lifespan.

Keywords: childhood adversity, hostility, cardiovascular reactivity

1. Introduction

Exposure to adverse childhood family environments increases the risk for negative psychological and physical health outcomes over the lifespan (Weich et al., 2009). A recent 45-year epidemiological study found that psychopathological consequences of childhood adversity can persist through adolescence, young adulthood, and into middle-age (Clark et al., 2010). Reports emanating from the Adverse Childhood Experiences study link early life adversity to a wide range of psychological and physical health outcomes, including obesity, depression, smoking, substance use, chronic obstructive pulmonary disease, chronic bronchitis and emphysema, and ischemic heart disease (Anda et al., 2006). The strength and frequency of these associations has prompted research into the mechanistic pathways by which early family risk factors exert their influence. An understanding of these processes may offer important insight into intervention strategies to reduce the long-term health impact of adverse childhood experiences.

Research into patterns of physiological stress reactivity has proven informative as one possible pathway linking childhood adversity to negative mental and physical health outcomes. Although the majority of research has focused on the hypothalamic-pituitary-adrenal (HPA) stress response system, childhood adversity has also been associated with dysregulation of the sympathetic nervous system. For example, dysregulated patterns of cardiovascular activity (Davies et al., 2009), elevated blood pressure (Ballard et al., 1993), and higher skin conductance level reactivity (El-Sheikh, 2005) have been observed among children from homes characterized by high conflict. Notably, these relations may persist later in life; poor cardiovascular stress reactivity and recovery patterns have been observed among young adults reporting poor family relationships in childhood (Luecken, 1998; Luecken et al., 2005). Elevated blood pressure and greater cardiovascular disease risk have also been observed among adults reared in early family environments characterized by low socioeconomic status (SES) and conflict (Lehman et al., 2009) and those maltreated in childhood (Batten et al., 2004).

Little is known about mechanisms underlying the relation between childhood adversity and cardiovascular stress responses. Lovallo and Gerin (2003) suggest that specific patterns of physiological stress reactivity may arise from individual differences at a cognitive-emotional level, as people evaluate events and formulate habitual behavioral responses. Thus, characteristic patterns of appraising and responding to stressful situations represent an important pathway linking stress to long-term disease outcomes. Bearing associations with both negative childhood family environments and physiological reactivity, hostility and hostile/aggressive behavioral responses to stress represent a particularly promising pathway for exploration.

Smith (1994) defines hostility as “a devaluation of the worth and motives of others, an expectation that others are likely sources of wrongdoing, a relational view of being in opposition toward others, and a desire to inflict harm or see others harmed” (p. 26). Negative parental behaviors, including rejection, low affection, conflict, and strict control, are associated with trait hostility in children and young adults (Matthews et al., 1996). The behavioral component of hostility, aggression, has been observed among children exposed to poor quality family environments (Andreas and Watson, 2009) and “destructive” conflicts (Cummings et al., 2004). Modeling of aggressive tactics in the home has been hypothesized to provide children with a framework by which to respond to other conflicts, and studies have observed direct effects of parental marital aggression on children’s aggressive behaviors at home and school (Erath and Bierman, 2006). Researchers have also theorized that negative interactions with caregivers disrupt children’s ability to self-regulate their emotions, leading to aggressive behavior (Cummings et al., 2004). Importantly, conditions of conflict early in life may contribute to a persistent hostile/aggressive interpersonal style, setting the stage for more hostile interpretations and aggressive behavioral responses to later experiences of stress. Larkin and colleagues (1998) provide evidence that adults reared in an early family environment characterized by low cohesion and adaptability utilize more negative verbal behavior during a stressful role-play task.

A hostile and aggressive orientation towards others has been associated with long-term adverse cardiovascular outcomes, including cardiovascular disease, hypertension, and abnormal cholesterol levels (Miller, Chen, & Cole, 2009). It is commonly theorized that exaggerated or prolonged cardiovascular reactivity is a mechanism linking hostility to long-term health outcomes (Das and O’Keefe, 2006; Rozanski, Blumenthal, & Kaplan, 1999). However, the literature is marked by inconsistency regarding the impact of self-reported trait hostility on the cardiovascular stress response among adults. Positive associations between adult hostility/aggression and blood pressure and heart rate reactivity to stress have been reported (Chida and Hamer, 2008), as well as negative associations (Kline et al., 2008), and contrasting effects of hostility on blood pressure and heart rate (Hernandez et al., 2009). High hostility has been associated with delayed recovery of blood pressure after stress (Brydon et al., 2010; Anderson et al., 2005), while others have found no differences in recovery between participants high and low in hostility (Vella and Friedman, 2007).

Similarly, existing literature on childhood adversity, hostility/aggression and cardiovascular reactivity presents a complex picture of hyper-reactivity in some contexts and hypo-reactivity in others. Although it seems somewhat consistent that children demonstrate elevated blood pressure and heart rate in the context of adverse family environments, the implications for stress responsivity in young adulthood are less clear, as both exaggerated and blunted patterns of cardiovascular activity have been reported. Further, as the aforementioned studies indicate, empirical evidence links hostile/aggressive states and behaviors among adults to both higher and lower cardiovascular reactivity. A plausible conclusion to be reached from these seemingly contradictory findings is that departures from normative cardiovascular activity and reactivity represent dysregulation that is potentially health-damaging. Such an interpretation is consistent with theories of allostatic load, suggesting that chronic stress can promote dysregulated physiological reactivity, as evidenced by responses that are exaggerated, prolonged, or blunted, contributing over time to the etiology of heart disease and infectious illnesses (McEwen and Wingfield, 2003). Exaggerated or prolonged cardiovascular arousal has been associated with the development of hypertension and preclinical and clinical cardiovascular disease (Treiber et al., 2003), while blunted cardiovascular reactivity has been linked to poorer self-rated health, depression, addictions, and obesity (Phillips, 2011).

Although theoretical models have linked early life stress to health and disease via a pathway of hostility that influences behavior and biological responsivity (Cohen et al., 2007; Repetti et al., 2002), empirical evidence has mainly evaluated discrete associations between individual components in such a pathway; comprehensive models remain to be tested (Matthews and Gallo, 2011). Moreover, prior studies of adult hostility have relied primarily upon limited self-report measures. The assessment of observable hostile behavior provides a different perspective that may advance understanding of the relations between hostility and cardiovascular activity. The current study extends prior literature by testing an integrative model of childhood adversity, hostile verbal behavior, and cardiovascular stress responses. It was hypothesized that young adults from negative childhood family environments would display more hostile/aggressive verbal behavior during a social stress interaction. It was further predicted that those from negative families would have poor cardiovascular recovery after the interaction. Given the preliminary state of this research area, we take an empirical approach to the evaluation of cardiovascular reactivity, and do not put forth a directional hypothesis. Finally, a path model was hypothesized suggesting that group differences in cardiovascular reactivity and recovery could be explained by hostile/aggressive verbal behavior during the task.

2. Material and methods

2.1. Participants

Participants were 75 students recruited from Introductory Psychology classes after completing a large screening survey that included the Family Relationships subscales (FR; conflict, cohesion, expressiveness) of the Moos Family Environment Scale (FES; Moos and Moos, 1994; α = .90). Respondents were asked to complete the FES in reference to their family environment prior to age 16. The three subscales were combined and scored such that higher FR scores reflect more positive relationships. Respondents age 18–22, raised in continuously married families by both biological parents, and who scored in the highest (positive relationships) or lowest (negative relationships) quartiles of FR on the screening survey were invited to participate. Respondents who experienced parental death or divorce were not eligible. Eighty-one potential participants were invited to a lab session 1–3 months after completing the screening survey, at which time they again completed the FES. Only participants who scored within the same positive or negative group on both administrations were included in analyses (n = 75; 39 from the lowest quartile and 36 from the highest quartile). Test-retest reliability on the FES was high, R = .91. Sample characteristics are displayed in Table 1.

Table 1.

Sample demographics and study variablesa

Total sample Neg FR Pos FR
Age (M, SD) 18.9 (.98) 18.8 (.90) 19.0 (1.1)
Gender (N, %)
 Male 38 (50%) 19 (49%) 19 (51%)
 Female 37 (50%) 20 (51%) 17 (49%)
Family Income (N, %)
 $0 – 29,999 3 (4%) 3 (8%) 0
 $30,000 – 44,999 8 (11%) 3 (8%) 4 (11%)
 $45,000 – 59,999 6 (8%) 4 (10%) 2 (5%)
 $60,000 – 79,999 7 (9%) 4 (10%) 3 (8%)
 $80,000 – 99,999 16 (21%) 9 (23%) 7 (20%)
 $100,000 + 34 (45%) 16 (41%) 18 (51%)
 No answer 2 (3%) 0 2 (5%)
Ethnicity (N, %)
 White non-Hispanic 57 (76%) 28 (72%) 29 (80%)
 Hispanic 12 (16%) 7 (18%) 5 (14%)
 African-American 2 (3%) 2 (5%) 0
 Asian 2 (3%) 1 (3%) 1 (3%)
 Other 2 (3%) 1 (3%) 1 (3%)
BMI (M, SD) 23.5 (3.8) 23.8 (4.1) 23.2 (3.4)
FR (M, SD)** 16.1 (6.7) 10.6 (3.7) 22.2 (2.6)
Anxiety symptoms (M, SD)* 7.8 (7.1) 9.4 (8.2) 5.9 (5.1)
Depressive symptoms (M, SD)** 8.2 (5.6) 10.1 (6.0) 5.9 (4.2)
Hostile/aggressive behavior (M, SD)** 5.9 (6.3) 8.1 (7.9) 3.9 (3.3)
Cardiovascular measures (M, SD)b
 SBP
  Pre-task 115.5 (10.7) 117.3 (11.8) 113.7 (9.4)
  Task 127.3 (12.2) 127.2 (12.6) 127.4 (11.9)
  Recovery 118.0 (10.3) 118.0 (10.5) 118.0 (10.4)
 DBP
  Pre-task 76.2 (6.3) 76.7 (6.9) 76.8 (5.7)
  Task 86.9 (9.7) 86.0 (10.3) 87.7 (9.2)
  Recovery 78.3 (7.3) 78.0 (6.7) 78.6 (7.9)
 HR
  Pre-task 91.2 (14.3) 94.6 (16.6) 87.8 (11.3)
  Task 94.9 (12.9) 96.4 (14.4) 93.4 (11.3)
  Recovery 92.9 (14.0) 95.4 (16.6) 90.4 (10.3)
Open in a new tab
a

“Neg FR” = negative family relationships group; “Pos FR” = positive family relationships group

b

unadjusted means

*

group difference p < .05

**

group difference p < .01

2.2. Procedure

Participation occurred between 1–5 PM, Monday-Friday. After providing informed consent and measuring height and weight, participants stood in the location of the upcoming role-play task while 10 minutes of pre-task cardiovascular measures were taken. Participants were instructed on and completed a 10-minute role-play task, after which 10 more minutes of cardiovascular readings were taken while participants remained standing. Participants completed questionnaires after the role-play.

2.2.1. Role-play task

The role-play task was chosen to reflect a naturalistic interpersonal stressor and has elicited angry emotional and behavioral responses in prior research (Frazer et al., 2002; Semenchuk and Larkin, 1993). Conflictual role-play tasks are believed to enhance the generalizability of cardiovascular responses to stress observed in the laboratory (Waldstein et al., 1998), especially among individuals high in hostility (Suls and Wan, 1993). For 10 minutes, participants role-played a challenging social situation (requesting a neighbor to turn down loud music so he/she can study for an important exam) with a same-sex research assistant. The interaction was videotaped, and the “neighbor” maintained a neutral expression and posture while following an ordered series of scripted responses indicating a refusal to cooperate. Experimenters and research assistants were blind to the participant’s family relationship history.

2.3. Hostile/aggressive verbal behavioral responses

Participant verbalizations during the role-play task were coded for spoken expressions of hostility/aggression. Categories of behavioral hostility/aggression were chosen in accordance with Smith’s (1994) definition, and included Patronize/Sarcasm/Mock (e.g., mimicking the research assistant), Blame/Insult (e.g., calling the research assistant “stubborn” or “inconsiderate”), Demand/Insist (e.g., “Shut the music off now”), Threat of Revenge (e.g., “Next time you have a big exam I’m going to turn my music up”), and Threat of Violence (e.g., “I’m going to make both you and your music be quiet”). Each interaction was divided into 30-second intervals and coded for the absence/presence (coded as 0 or 1) of each strategy in each interval. Two members of the research team, blind to the participant’s family relationship history, coded a small subset of videotapes collaboratively until multiple examples of each strategy were observed and agreed upon. Coders then worked independently until 10% of the videotapes had been coded consecutively with a reliability (kappa) ≥ 0.85. The remaining videotapes were then divided among coders for completion, with a random 10% of the remaining videos coded collaboratively to verify continued interrater reliability. Interrater reliability was high for all categories, ranging from 0.85 to 0.98. A composite hostile/aggressive verbal behavior variable was created by summing the frequency of occurrence of each type of behavior.

Confirmatory analysis of the composite variable was conducted with structural equation modeling (SEM; Amos 18.0; Maximum Likelihood Estimation), by forming a latent variable with the 5 categories (patronize, blame, demand, revenge, violence) as indicators. The model represented a good fit to the data, χ2 (5) = 2.3, p = .80, CFI = 1.0, RMSEA = 0.0, {0.0; 0.098}. All factor loadings were significant at p < .05.

2.4. Cardiovascular Measures

Systolic blood pressure (SBP), diastolic blood pressure (DBP) and heart rate (HR) were measured using an Omega 5600 adult blood pressure monitor (Invivo Research; Orlando, FL), with the cuff positioned on the participant’s non-dominant arm. The Omega 5600 uses an oscillometric method of measurement to provide SBP, DBP, and HR. Readings were taken every two minutes during ten minute pre-task, task, and recovery periods, and were then averaged for the pre-task, task, and recovery experimental periods to create three repeated measures.

2.5. Anxiety and depressive symptoms

Because of the potential confounding effects of current anxiety or depressive symptoms on the relation between family relationship history and current cardiovascular activity, participant reports of recent symptoms of anxiety were assessed with the Beck Anxiety Inventory (BAI: Beck, 1990; α = .91), and recent symptoms of depression were assessed with the Beck Depression Inventory II (BDI: Beck, 1996; α = .84).

2.6. Data Analyses

Repeated measures GLM were used to evaluate the impact of childhood family relationships and hostile/aggressive behavior on SBP, DBP, and HR responses to the role-play task. Greenhouse-Geisser corrections were used to correct for sphericity. Family relationships group (coded with the negative relationships group = ‘0’ and the positive relationships group = ‘1’) or hostile/aggressive verbal behavior (entered as a continuous variable) served as independent variables, and three repeated cardiovascular readings (pre-task, task, and recovery) served as dependent variables. Participant sex and BMI were covariates in models predicting SBP and DBP (see preliminary analyses for details of covariate selection). Models predicting HR included BMI and anxiety as covariates. Analyses predicting hostile/aggressive verbal behavior included participant sex as a covariate. BMI and anxiety were centered at the sample mean. Following significant omnibus tests of differences in the overall pattern of responses, planned within-subjects contrasts evaluated reactivity (the change from pre-task to task measures) and recovery (the change from task to recovery measures). Partial eta-squared (ηp2) is reported as a measure of effect size.

AMOS 18.0 was used to evaluate the proposed structural model shown in Figure 1. Standardized residualized change scores were calculated to represent reactivity and recovery. The residualized change score, which represents the difference between the observed HR and the predicted HR, is used as an alternative to calculating difference scores because it adjusts for baseline but avoids some of the reliability concerns with difference scores (MacKinnon, 2008). This strategy does not fully remove baseline differences between the groups, but is used to minimize their effect on change scores. A standardized residualized change score for reactivity was created by predicting task HR from pre-task HR. A standardized residualized change score for recovery was created by predicting recovery HR from task HR. Anxiety symptoms were included as a covariate in the path model. Because men exhibited more hostile/aggressive verbal behavior, sex was also included as a covariate. Maximum Likelihood Estimation was used, and model fit was examined using Chi-square Goodness of Fit Test, as well as Cumulative Fit Index (CFI) and Root Mean Square Error of Approximation (RMSEA), as recommended for smaller sample sizes (Fan et al., 1999).

Figure 1.

Figure 1

Open in a new tab

A path model linking childhood family relationships to HR responses to a conflict task via hostile/aggressive verbal behaviora

aAll paths significant at p < .05. Regression weights and standard errors shown. Higher scores on Childhood Family Relations indicate more positive relationships. HR reactivity and recovery are represented by residualized change scores. Standardized error terms not shown.

3. Results

3.1. Preliminary Analyses

3.1.1. Demographic comparisons

The family relationships groups were compared for equivalence on demographic variables and covariates potentially associated with cardiovascular activity. Chi-square and t-tests found no group differences in sex (p = .73), ethnicity (p = .72), family income (p = .36), age (p = .36), body mass index (BMI; p = .51), waist/hip ratio (p = .60), hormonal contraceptive use (p = .42), medications (p = .38), or smoking status (p = .21). On the day of testing, the groups did not differ on the time of day of participation (p = .84), number of cigarettes smoked (p = .46), or caffeine/energy drinks (p = .39). Higher BMI predicted elevated overall SBP (p < .01) and SBP reactivity (p = .049). BMI did not predict overall DBP (p = .09), DBP reactivity (p = .65), overall HR (p = .91), or HR reactivity (p = .09). Because BMI predicted SBP and was near-significant in the prediction of overall DBP and HR reactivity, it was included as a covariate in all statistical models predicting cardiovascular outcomes.

3.1.2. Current anxiety and depression

Participants from families characterized by negative relationships reported elevated symptoms of anxiety, t(71) = 2.2, p = .032, and depression, t(71)=3.4, p = .001, relative to those from families characterized by positive relationships (see Table 1). Anxiety predicted elevated HR reactivity to the challenge task, F(2, 126) = 3.4, p = .050, but did not predict SBP or DBP reactivity, p’s ≥ .26, or overall HR, SBP, or DBP, p’s ≥ .29. Depressive symptoms did not predict HR, SBP, or DBP reactivity, p’s ≥ .46, or overall HR, SBP, or DBP, p’s ≥ .28. Therefore, anxiety was included as a control variable in statistical analyses of HR. Neither anxiety (p = .59) nor depressive symptoms (p = .30) were significantly correlated with hostile behavior during the task.

3.1.3. Gender differences

Women had lower SBP and DBP across the task than men (SBP p < .001; DBP p = .017). Women did not differ in SBP reactivity (p =.25), but did show less DBP reactivity than men (p = .001). Because of these gender differences, participant sex was a covariate in models predicting SBP or DBP. Men and women did not differ in overall HR (p = 14) or HR reactivity (p = .57). Men and women did not differ on reports of family relationships (p = .87). Men displayed more hostile/aggressive verbal behavior (M = 7.5, SD = 7.4) than women (M = 4.3, SD = 4.6; p = .04).

3.1.4. Reactivity to the role-play task

Significant reactivity (change from pre-task to task level) to the role-play task was observed across the sample for SBP, F(1,67) = 140.7, p < .001, DBP, F(1,66) = 145.4, p < .001, and HR, F(1,67) = 14.0, p < .001. Significant recovery (change from task to recovery level) was also observed across the entire sample for SBP, F(1,67) = 86.2, p < .001, DBP, F(1,66) = 66.4, p < .001, and HR, F(1,67) = 9.5, p = .01.

3.1.5. Examination of outliers

Outliers were examined using the measures of influence DFFITS and DFBETAS (SPSS 17.0; Chicago, IL). For small to moderate sample sizes, DFFITS and DFBETAS greater than an absolute value of 1 suggest an influential case and require further analyses (Cohen et al., 2003). Three cases were identified as potentially influential (2 from the negative family group, 1 from the positive family group). Upon closer inspection, all three reported elevated anxiety (M = 26.3; sample M = 7.0). When analyses were repeated excluding those 3 data points there were no significant changes in the pattern or significance of findings, with the exception that anxiety symptoms were no longer predictive of HR reactivity, p = .55. Thus all cases were retained for analyses.

3.2. Family relationships and cardiovascular responses to the task

We first evaluated the influence of a negative childhood family environment on SBP and DBP responses to the roleplay task. A repeated measures GLM predicting SBP from family relationships group, controlling for sex and BMI was not significant in the prediction of SBP response to the task (p = .31) or mean SBP across the task (p = .88). Similarly, family relationships group was not significant in the prediction of DBP response to the task (p = .24) or mean DBP across the task (p = .63).

We next evaluated the influence of a negative childhood family environment on HR responses. Family relationships group was a significant predictor of HR response to the task, F(2, 124) = 4.2, p = .018 (Greenhouse-Geisser adjusted p = .027), ηp2 = .063 (see Figure 2). Tests of within-subjects contrasts were significant for the change from pre-task to task levels, F(1,62)=6.4, p =.014, ηp2 = .093, such that the negative relationships group had attenuated reactivity. The change from task to recovery did not significantly differ between the groups, p = 0.20. Although the groups did not differ in overall HR (p = .24), the negative relationships group showed a non-significant trend towards higher pre-task HR than the positive relationships group, F(1,66)=3.6, p = .063.

Figure 2.

Figure 2

Open in a new tab

HR responses to roleplay task by the quality of childhood family relationshipsa

a“Neg FR” = negative family relationships group; “Pos FR” = positive family relationships group; Error bars are standard errors.

3.3. Family relationships and hostile/aggressive behavior

Analyses comparing family relationships groups on hostile/aggressive verbal behavior during the task found that the negative relationships group displayed significantly more hostility/aggression during the task than the positive relationships group, F(1,55)=8.0, p = .007 (see Table 1).

3.4. Hostile/aggressive behavior and cardiovascular responses to the task

Next, we evaluated if hostile/aggressive verbal behavior during the task would be associated with cardiovascular responses. Hostile/aggressive behavior was not a significant predictor of SBP responses to the task (p = .11), DBP responses to the task (p = .69), overall SBP (p = .33) or overall DBP (p = .59). However, a repeated measures GLM predicting HR from hostile/aggressive behaviors was significant, F(2,108) = 4.07, p = 0.020 (Greenhouse-Geisser adjusted p = .029), ηp2 = .070 (see Figure 3). Analyses of reactivity were significant for the change in HR from pre-task to task levels, F(1,54) = 5.1, p = 0.028, ηp2 = .087, showing attenuated reactivity among those displaying more hostile/aggressive verbal behaviors. The change from task to recovery was also significant, F(1,54) = 4.5, p = 0.038, ηp2 =.078, such that more hostile/aggressive participants had attenuated HR recovery. Hostile/aggressive behaviors did not predict overall HR (p = .91).

Figure 3.

Figure 3

Open in a new tab

Hostile/aggressive behavioral responses to the task and HR reactivitya

a“low aggr/ho”: set at 1 SD below the mean on hostile/aggressive responses during the task.

“high aggr/ho”: set at 1 SD above the mean on hostile/aggressive responses during the task.

3.5. Path model

The proposed path model shown in Figure 1 evaluated a path linking negative childhood family relationships to more hostile/aggressive verbal behavior during the conflict task, which then would predict attenuated reactivity and recovery. Because the prior analyses were only significant for HR, the model was only evaluated predicting HR reactivity and recovery. The model represented a good fit to the data (criteria by Hu and Bentler, 1999): χ2 (6) = 2.8, p = .84, CFI = 1.0, RMSEA=0.00, {0.0; 0.089}, AIC=44.8, all paths significant at p < .05. More negative family relationships predicted higher hostile/aggressive behavior, and higher hostile/aggressive behavior predicted attenuated reactivity and attenuated recovery. Higher anxiety predicted higher reactivity and stronger recovery. A comparison model evaluating if elevated pre-task HR may have contributed to hostile/aggressive behavior included a path from pre-task HR to hostile verbal behavior. This model was a poor fit to the data, χ2 (9) = 33.0, p < .01, CFI = .697, RMSEA=0.19, {0.12; 0.26}, AIC=85.0, and the path from pre-task HR to hostile verbal behavior was not significant, p =.39. Results suggest that the more parsimonious original model is a better fit to the observed data.

4. Discussion

The present study examined the influence of negative family relationships in childhood on behavioral and cardiovascular responses to a conflict interaction task in young adulthood, and evaluated a behavioral pathway linking childhood adversity to cardiovascular reactivity in young adulthood. Attenuated HR reactivity to the task was observed among young adults reporting more negative family relationships relative to those who experienced more positive relationships. This finding supports a growing research literature linking early adversity to attenuated biological stress response systems (Miller et al., 2007). Exposure to chronic stress, as is the nature of consistently negative family relationships, has been theorized to lead to reduced physiological reactivity to new stressors over time (Musante et al., 2000). Existing literature has primarily assessed neuroendocrine systems; however attenuation may also be apparent in sympathetic nervous system responses to stress. For example, childhood adversity has been associated with lower vascular resistance in childhood (Gump et al., 1999), as well as lower daily blood pressure (Luecken et al., 2009) in young adulthood. Consistent with the results of the current study, attenuated heart rate reactivity has also been observed among young adult men reared in stressful childhood family environments (Torres et al., 2001). Although research often focuses on the negative health consequences of increased reactivity, attenuated HR reactivity to stress may also put individuals at higher long-term risk for poor health. Responses that do not sufficiently prepare the body to respond to stress or facilitate recovery are recognized as maladaptive, and have been associated with risk for addiction, obesity, cardiovascular disease, poor self-rated health, and depression (Carroll et al., 2009; Phillips, 2011; Salomen et al., 2009).

The current study also evaluated the relation between negative childhood family relationships and behavioral concomitants of hostility. As theorized, we observed greater hostile verbal behavior during the role-play task among young adults reared in negative early family environments. Developmental contextual approaches to social learning describe the family environment as a “training ground” through which parents model and support social skills that influence children’s interpersonal relationships (Cui et al., 2002). Perceptions of the primary caregivers as unsupportive, argumentative, and uncaring may contribute to the development of the hostile behaviors observed among individuals exposed to negative family relationships in childhood. A general interpretive style characterized by heightened vigilance and suspicion of others may develop in the context of stressful and unpredictable childhood environments (Chen et al., 2004); our findings suggest that such an interpersonal orientation may persist later in life. In a recent study of implicit affect, attributions of anger and fear to one’s early childhood family environment were associated a greater tendency to perceived threat in response to ambiguous cues among adults (Chan et al., 2011).

Our results further suggest that hostile behaviors during interpersonal stressors can significantly impact cardiovascular activity. More hostile/aggressive verbal behavior during the role-play task predicted attenuated HR reactivity and recovery. Discrete findings linking childhood adversity to hostile behavior and attenuated HR responses support prior research, however the current study advances the literature by evaluating hostile behavior as a potential mechanism linking adversity to attenuation. A path analytic model of the relations between childhood adversity, contemporaneous hostile/aggressive behavior, and HR stress responses provided a good fit to the data, and suggested that negative family-of-origin relationships can exert a lasting effect on heart rate responses to stress into adulthood via a pathway of hostile verbal behavior during challenging interpersonal interactions. These results are partially supportive of findings by Larkin et al. (2010), who similarly observed more negative verbal behavior during a stressful role-play task among young adults reared in negative family environments. However, their findings of elevated DBP reactivity in response to stress diverged from the current study which found no effects of hostile verbal behavior on DBP reactivity or recovery.

One strength of this study is in the use of observational measures of hostility and aggression during a simulated social conflict task. Behavioral assessments of hostility have been primarily conducted among children; within adult samples, self-reports have predominated. Smith (2003) recommended that studies inducing stress responses use an experimental stressor relevant to the population or psychosocial risk factor being studied. The construct of hostility is, in part, understood in terms of its consequences on social interactions and social relationships, suggesting that an interpersonal conflict stressor may provide an ecologically valid approach to the study of the impact of childhood adversity and hostility on cardiovascular reactivity. By observing and coding verbal behavior during the role-play, we assessed hostile responses as they occurred, minimizing potential biases in self-reports, and more immediately tying hostile behavior to cardiovascular responses. The observed pattern of aggressive verbal behavior and cardiovascular consequences in participants from negative families may simulate the responses that occur in a variety of social interactions in participants’ natural environments, increasing the potential long-term health risks for those exposed to early family adversity. Although self-reported trait hostility in adulthood has been associated with patterns of higher and lower cardiovascular activity, a meta-analysis across 40 studies related childhood aggressive behavior to lower heart rate during a stressful task (Ortiz & Raine, 2004). Thus the method through which hostility is assessed, as well as developmental stage, may be important considerations when evaluating the relation of hostility to cardiovascular activity.

The finding that hostile/aggressive behavior predicted attenuated HR recovery further emphasizes the importance of considering recovery from stress, in addition to reactivity, in gaining a comprehensive understanding of the physiological responses of young adults exposed to adverse childhood family relationships. Delayed cardiovascular recovery after the termination of a stressor may emerge after prolonged stress, and has been linked to future hypertension risk (Steptoe and Marmot, 2005). Prolonged cardiovascular activity resulting from poor recovery may predispose individuals to heart problems in addition to the risks imposed by elevated reactivity (Curtis and O’Keefe, 2002; Stewart and France, 2001).

Although significant reactivity to the task was observed across the sample for SBP, DBP, and HR measures, the relations among family environment, hostility/aggression, and attenuated reactivity and recovery responses found in the present study were only observed among measures of HR. Prior studies of negative family environments have observed an effect on measures of HR (Torres et al., 2003) or BP (Larkin et al., 2010) only, without significant findings on other indices of cardiovascular functioning. These nuanced patterns of stress responsivity highlight the importance of examining multiple physiological indices through which adverse childhood environments may influence health. Individual differences in appraisals of the task were not assessed in the current study, but may help future studies understand differences between HR and SBP findings. For example, “active” stressors are expected to elicit elevations in both SBP and HR, while “passive” stressors may elicit elevated SBP but minimal HR responses (Gregg et al., 1999). Passive and withdrawn orientations toward stressors may develop among individuals consistently exposed to family adversity early in life (Santiago and Wadsworth, 2009; Repetti et al., 2002). Thus, a speculative explanation for our findings is that participants from negative families may have been more likely to approach the conflict task in a passive manner, resulting in similar SBP reactivity but a blunted HR response relative to those from positive families. However, it should be noted that heart rate and blood pressure responses may appear discrepant when examining any one association within a discrete measurement period, but bear greater relation when multiple measurements are aggregated across tasks or settings (Kamarck and Lovallo, 2003).

Because participants from more negative family environments also reported higher current symptoms of anxiety and depression, it was important to consider whether the heart rate findings were explained by current distress. Higher anxiety predicted greater reactivity to the role-play task across the sample, but the relation of negative family relationships to reduced heart rate reactivity remained significant after controlling for anxiety. Although inclusion in the statistical model does not fully remove the effect of anxiety, it is suggests anxiety does not explain the effects of family group on heart rate reactivity. Hostile behavior was uncorrelated with anxiety, and was not predicted by pre-task HR. Further, adjusting for anxiety did not alter the observed behavioral pathway linking adversity to HR reactivity. Nonetheless, additional time for relaxation and acclimation to a novel laboratory environment may help minimize effects of anxiety on cardiovascular reactivity. Levels of depressive symptoms were not significantly related to any of the cardiovascular measures. In short, we find little evidence that current distress explains the relation of a negative childhood family environment to heart rate reactivity in young adulthood.

Results of the current study must be evaluated in light of its limitations. Results may not generalize to adults in a different developmental stage or to the general population. The characteristics of childhood family relationships were assessed retrospectively, and although accuracy cannot be independently verified, high test-retest reliability of the measure was obtained. In addition, experimenters and coders were blind to participants’ family environments, and we considered current distress in our analyses. All participants were raised by two continuously married biological parents. The sample may not have captured individuals exposed to more severe family relations or adversity for whom the model of hostility, aggression, and stress reactivity may be especially relevant. Previous research has highlighted the multidimensional nature of aggression (e.g., proactive, reactive, overt, relational, indirect, etc.; Vaillancourt et al., 2007), a distinction not made by the current study. Our observational measure included a composite of multiple types of hostile verbal behavior (e.g., patronizing, blaming, demanding, threatening). It would be interesting for future studies to attempt to link specific types of hostile behavior to particular patterns of HR arousal and recovery.

The measured heart rates in the sample were higher than typical resting heart rates among young adults, potentially because participants were standing and the room was warm. The heart rates in the sample should not be taken to represent clinical measures of resting blood pressure/heart rate. One effect of an adverse childhood environment on HR may be via negative emotionality; it is possible that the negative family group reacted with more arousal upon entering the novel environment or reading the Informed Consent. Chronically elevated HR or HR increases upon entering the lab could contribute to a ceiling effect in reactivity to the interaction task, potentially contributing to reduced reactivity in the negative family group. The groups did not significantly differ in pre-task or overall HR, but it is possible that the sample size limited power to find significant differences. Future studies of HR in natural environments will be important in order to address these possibilities.

There are several of limitations associated with path modeling. Results from SEM analyses support the theorized path model, but the non-experimental design does not allow for conclusions about causality. Alternative models may also fit the data, but the model is strengthened by theoretical support for causal relations and temporal precedence in the ordering of the variables. An alternative model in which pre-task HR predicted hostile verbal behavior was a poor fit to the data and the path from pre-task HR to hostile behavior was not statistically significant. Finally, the sample size was relatively small for SEM. We used model evaluation indices appropriate for small sample sizes but the results should be considered exploratory until replicated with a larger sample size.

The present study advances research on the health impact of childhood adversity by evaluating a potential behavioral mechanism linking adversity to cardiovascular activity in young adulthood. Negative childhood family relationships were associated with more hostile/aggressive verbal behavioral responses to stress in young adulthood as compared to participants reporting positive childhood family relationships. Additionally, previously observed relations between hostility/aggression and attenuated heart rate reactivity and recovery were extended to young adults with a history of family adversity. In combination, support was found for a path model in which negative family relationships in childhood contributed to behavioral aggression in young adulthood in the context of a challenging interpersonal interaction, which predicted attenuated heart rate reactivity and recovery. Over time, aggressive behavioral responses to recurring interpersonal conflict may contribute to maladaptive cardiovascular activity, increasing long-term health risk. These results provide evidence for the development of hostility as an important psychosocial mechanism of risk for dysregulated cardiovascular responses to stress in young adults exposed to an adverse childhood family environment.

Highlights.

  • We examined the impact of early family adversity on cardiovascular stress reactivity in young adulthood.

  • Young adults from negative families had attenuated heart rate reactivity during a conflict role-play task.

  • Young adults from negative families displayed more hostile behavior during the conflict task.

  • Hostility during the task was associated with attenuated heart rate reactivity and recovery.

  • Negative childhood family environments may impact adult cardiovascular functioning via a pathway of hostile responses to stress.

Acknowledgments

Role of the funding source

This research was supported by NIMH R03 MH069804-1 Luecken (PI). The funding source had no role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

ABBREVIATIONS

BMI

body mass index

BP

blood pressure

FR

family relationships

HPA

hypothalamic-pituitary-adrenal

HR

heart rate

SAM

sympathetic-adrenomedullary

SBP

systolic blood pressure

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Anda RF, Felitti VJ, Bremner JD, Walker JD, Whitfield C, Perry BD, Dube SR, Giles WH. The enduring effects of abuse and related adverse experiences in childhood: A convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatry Clin Neurosci. 2006;256:174–186. doi: 10.1007/s00406-005-0624-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anderson JC, Linden W, Habra ME. The importance of examining blood pressure reactivity and recovery in anger provocation research. Int J Psychophysiol. 2005;57:159–163. doi: 10.1016/j.ijpsycho.2004.12.011. [DOI] [PubMed] [Google Scholar]
  3. Andreas JB, Watson MW. Moderating effects of family environment on the association between children’s aggressive beliefs and their aggression trajectories from childhood to adolescence. Dev Psychopathol. 2009;21:189–205. doi: 10.1017/S0954579409000121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ballard ME, Cummings EM, Larkin K. Emotional and cardiovascular responses to adults’ angry behavior and to challenging tasks in children of hypertensive and normotensive parents. Child Dev. 1993;64:500–515. [PubMed] [Google Scholar]
  5. Batten SV, Aslan M, Maciejewski PK, Mazure CM. Childhood maltreatment as a risk factor for adult cardiovascular disease and depression. J Clin Psychiatry. 2004;65:249–254. doi: 10.4088/jcp.v65n0217. [DOI] [PubMed] [Google Scholar]
  6. Beck AT, Steer RA. The Beck Anxiety Inventory Manual. The Psychological Corporation; 1990. [Google Scholar]
  7. Beck AT. The Beck Depression Inventory II. Harcourt Brace & Co; San Antonio: 1996. [Google Scholar]
  8. Brydon L, Strike PC, Bhattacharyya MR, Whitehead DL, McEwan J, Zacchary I, Steptoe A. Hostility and physiological response to laboratory stress in acute coronary syndrome patients. J Psychosom Res. 2010;68:109–116. doi: 10.1016/j.jpsychores.2009.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carroll D, Lovallo WR, Phillips AC. Are large physiological reactions to acute psychological stress always bad for health? Soc Personal Psychol Compass. 2009;3:725–743. [Google Scholar]
  10. Chan M, Chen E, Hibbert SA, Wong JHK, Miller GE. Implicit measures of early-life family conditions: Relationships to psychosocial characteristics and cardiovascular disease risk in adulthood. Health Psychol. 2011;30:570–578. doi: 10.1037/a0024210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen E, Langer DA, Raphaelson YE, Matthews KA. Socioeconomic status and health in adolescents: The role of stress interpretations. Child Dev. 2004;75:1039–1052. doi: 10.1111/j.1467-8624.2004.00724.x. [DOI] [PubMed] [Google Scholar]
  12. Chida Y, Hamer M. Chronic psychosocial factors and acute physiological responses to laboratory-induced stress in healthy populations: A quantitative review of 30 years of investigations. Psychol Bull. 2008;134:829–885. doi: 10.1037/a0013342. [DOI] [PubMed] [Google Scholar]
  13. Clark C, Caldwell T, Power C, Stansfeld S. Does the influence of childhood adversity on psychopathology persist across the lifecourse? A 45-year prospective epidemiologic study. Ann Epidemiol. 2010;20:385–394. doi: 10.1016/j.annepidem.2010.02.008. [DOI] [PubMed] [Google Scholar]
  14. Cohen J, Cohen P, West SG, Aiken LS. Applied multiple regression/correlation analysis for the behavioral sciences. 3. Lawrence Erlbaum Associates, Inc; New Jersey: 2003. [Google Scholar]
  15. Cohen S, Janicki-Deverts D, Miller GE. Psychological stress and disease. JAMA. 2007;298:1685–1687. doi: 10.1001/jama.298.14.1685. [DOI] [PubMed] [Google Scholar]
  16. Cui M, Conger RD, Bryant CM, Elder GH. Parental behavior and the quality of adolescent friendships: A social contextual perspective. J Marriage Fam. 2002;64:676–689. [Google Scholar]
  17. Cummings EM, Goeke-Morey MC, Papp LM. Everyday marital conflict and child aggression. J Abnorm Child Psychol. 2004;32:191–202. doi: 10.1023/b:jacp.0000019770.13216.be. [DOI] [PubMed] [Google Scholar]
  18. Curtis BM, O’Keefe JH. Autonomic tone as a cardiovascular risk factor: The dangers of chronic flight or fight. Mayo Clin Proc. 2002;77:45–54. doi: 10.4065/77.1.45. [DOI] [PubMed] [Google Scholar]
  19. Das S, O’Keefe JH. Behavioral cardiology: Recognizing and addressing the profound impact of psychosocial stress on cardiovascular health. Curr Atheroscler Rep. 2006;8:111–118. doi: 10.1007/s11883-006-0048-2. [DOI] [PubMed] [Google Scholar]
  20. Davies PT, Sturge-Apple ML, Cicchetti D, Manning LG, Zale E. Children’s patterns of emotional reactivity to conflict as explanatory mechanisms in links between interpartner aggression and child physiological functioning. J Child Psychol Psychiatry. 2009;50:1384–1391. doi: 10.1111/j.1469-7610.2009.02154.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. El-Sheik M. The role of emotional responses and physiological reactivity in the marital- conflict-child functioning link. J Child Psychol Psychiatry. 2005;46:1191–1199. doi: 10.1111/j.1469-7610.2005.00418.x. [DOI] [PubMed] [Google Scholar]
  22. Erath SA, Bierman KL the Conduct Problems Prevention Research Group. Aggressive marital conflict, maternal harsh punishment, and child aggressive-disruptive behavior: Evidence for direct and mediated relations. J Fam Psychol. 2006;20:217–226. doi: 10.1037/0893-3200.20.2.217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Fan X, Thompson B, Wang L. Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes. Struct Equ Modeling. 1999;6:56–83. [Google Scholar]
  24. Frazer NL, Larkin KT, Goodie JL. Do behavioral responses mediate or moderate the relation between cardiovascular reactivity to stress and parental history of hypertension? Health Psychol. 2002;21:244–253. [PubMed] [Google Scholar]
  25. Gregg ME, James JE, Matyas TA, Thorsteinsson EB. Hemodynamic profile of stress-induced anticipation and recovery. Int J Psychophysiol. 1999;34:147–162. doi: 10.1016/s0167-8760(99)00074-4. [DOI] [PubMed] [Google Scholar]
  26. Gump BB, Matthews KA, Räikkönen K. Modeling relationships among socioeconomic status, hostility, cardiovascular reactivity, and left ventricular mass in African American and White children. Health Psychol. 1999;18:140–150. doi: 10.1037//0278-6133.18.2.140. [DOI] [PubMed] [Google Scholar]
  27. Hernandez DH, Larkin KT, Whited MC. Cardiovascular response to interpersonal provocation and mental arithmetic among high and low hostile young adult males. Appl Psychophysiol Biofeedback. 2009;34:27–35. doi: 10.1007/s10484-009-9076-3. [DOI] [PubMed] [Google Scholar]
  28. Hu L, Bentler PM. Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Modeling. 1999;6:1–55. [Google Scholar]
  29. Kamarck TW, Lovallo WR. Cardiovascular reactivity to psychological challenge: conceptual and measurement considerations. Psychosom Med. 2003;65:9–21. doi: 10.1097/01.psy.0000030390.34416.3e. [DOI] [PubMed] [Google Scholar]
  30. Kline KA, Fekete EM, Sears CM. Hostility, emotional expression, and hemodynamic responses to laboratory stressors: Reactivity attenuating effects of a tendency to express emotion interpersonally. Int J Psychophysiol. 2008;68:177–185. doi: 10.1016/j.ijpsycho.2008.01.005. [DOI] [PubMed] [Google Scholar]
  31. Larkin KT, Frazer NL, Wheat AL. Responses to interpersonal conflict among young adults: Influence of family of origin. Pers Relationship. 2010;17:1–11. [Google Scholar]
  32. Larkin KT, Semenchuk EM, Frazer NL, Suchday S, Taylor RL. Cardiovascular and behavioral response to social confrontation: Measuring real-life stress in the laboratory. Ann Behav Med. 1998;20:294–301. doi: 10.1007/BF03356737. [DOI] [PubMed] [Google Scholar]
  33. Lehman BJ, Taylor SE, Kiefe CI, Seeman TE. Relationship of early life stress and physiological functioning to blood pressure in the CARDIA study. Health Psychol. 2009;28:338–346. doi: 10.1037/a0013785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lovallo WR, Gerin W. Psychophysiological reactivity: Mechanisms and pathways to cardiovascular disease. Psychosom Med. 2003;65:36–45. doi: 10.1097/01.psy.0000033128.44101.c1. [DOI] [PubMed] [Google Scholar]
  35. Luecken LJ. Childhood attachment and loss experiences affect adult cardiovascular and cortisol function. Psychosom Med. 1998;60:765–772. doi: 10.1097/00006842-199811000-00021. [DOI] [PubMed] [Google Scholar]
  36. Luecken LJ, Kraft A, Appelhans BM, Enders C. Emotional and cardiovascular sensitization to daily stress following childhood parental loss. Dev Psychol. 2009;45:296–302. doi: 10.1037/a0013888. [DOI] [PubMed] [Google Scholar]
  37. Luecken LJ, Rodriquez A, Appelhans BM. Cardiovascular stress responses in young adulthood associated with family-of-origin relationships. Psychosom Med. 2005;67:514–521. doi: 10.1097/01.psy.0000160466.10397.18. [DOI] [PubMed] [Google Scholar]
  38. MacKinnon DP. Introduction to Statistical Mediation Analysis. Lawrence Erlbaum Associates, Inc; New Jersey: 2008. [Google Scholar]
  39. Matthews KA, Gallo LC. Psychological perspectives on pathways linking socioeconomic status and physical health. Annu Rev Psychol. 2011;62:501–530. doi: 10.1146/annurev.psych.031809.130711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Matthews KA, Woodall KL, Kenyon K, Jacob T. Negative family environment as a predictor of boy’s future status on measures of hostile attitudes, interview behavior, and anger expression. Health Psychol. 1996;15:30–37. doi: 10.1037//0278-6133.15.1.30. [DOI] [PubMed] [Google Scholar]
  41. McEwen BS, Wingfield JC. The concept of allostasis in biology and biomedicine. Horm Behav. 2003;43:2–15. doi: 10.1016/s0018-506x(02)00024-7. [DOI] [PubMed] [Google Scholar]
  42. Miller G, Chen E, Cole SW. Health psychology: Developing biologically plausible models linking the social world and physical health. Annu Rev Psychol. 2009;60:501–524. doi: 10.1146/annurev.psych.60.110707.163551. [DOI] [PubMed] [Google Scholar]
  43. Miller GE, Chen E, Zhou ES. If it goes up, must it come down? Chronic stress and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol Bull. 2007;133:25–45. doi: 10.1037/0033-2909.133.1.25. [DOI] [PubMed] [Google Scholar]
  44. Moos RH, Moos BS. Family Environment Scale Manual. Consulting Psychologists Press; Palo Alto, CA: 1994. [Google Scholar]
  45. Musante L, Treiber FA, Kapuku G, Moore D, Davis H, Strong WB. The effects of life events on cardiovascular reactivity to behavioral stressors as a function of socioeconomic status, ethnicity, and sex. Psychosom Med. 2000;62:760–767. doi: 10.1097/00006842-200011000-00004. [DOI] [PubMed] [Google Scholar]
  46. Ortiz J, Raine A. Heart rate level and antisocial behavior in children and adolescents: A meta-analysis. J Am Acad Child Psy. 2004;43:154–162. doi: 10.1097/00004583-200402000-00010. [DOI] [PubMed] [Google Scholar]
  47. Phillips AC. Blunted cardiovascular reactivity relates to depression, obesity, and self- reported health. Biol Psychol. 2011;86:106–113. doi: 10.1016/j.biopsycho.2010.03.016. [DOI] [PubMed] [Google Scholar]
  48. Repetti RL, Taylor SE, Seeman TE. Risky families: Family social environments and the mental and physical health of offspring. Health Psychol. 2002;128:330–366. [PubMed] [Google Scholar]
  49. Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation. 1999;99:2192–2217. doi: 10.1161/01.cir.99.16.2192. [DOI] [PubMed] [Google Scholar]
  50. Salomen K, Clift A, Karlsdóttir M, Rottenberg J. Major depressive disorder is associated with attenuated cardiovascular reactivity and impaired recovery among those free of cardiovascular disease. Health Psychol. 2009;28:157–165. doi: 10.1037/a0013001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Santiago CD, Wadsworth ME. Coping with family conflict: What’s helpful and what’s for low-income adolescents. J Child Fam Stud. 2009;18:912–202. [Google Scholar]
  52. Semenchuk EM, Larkin KT. Behavioral and cardiovascular responses to interpersonal challenges among male offspring of essential hypertensives. Health Psychol. 1993;12:416–419. doi: 10.1037//0278-6133.12.5.416. [DOI] [PubMed] [Google Scholar]
  53. Smith TW. Health Psychology. In: Schinka JA, Velicer W, Weiner IB, editors. Handbook of Psychology: Volume 2: Research Methods in Psychology. Jon Wiley & Sons; New Jersey: 2003. pp. 241–270. [Google Scholar]
  54. Smith TW. Concepts and methods in the study of anger, hostility, and health. In: Siegman AW, Smith TW, editors. Anger, Hostility and the Heart. Lawrence Erlbaum Associates, Inc; New Jersey: 1994. pp. 23–42. [Google Scholar]
  55. Steptoe A, Marmot M. Impaired cardiovascular recovery following stress predicts 3- year increases in blood pressure. J Hypertens. 2005;23:529–36. doi: 10.1097/01.hjh.0000160208.66405.a8. [DOI] [PubMed] [Google Scholar]
  56. Stewart JC, France CR. Cardiovascular recovery from stress predicts longitudinal changes in blood pressure. Biol Psychol. 2001;58:105–120. doi: 10.1016/s0301-0511(01)00105-3. [DOI] [PubMed] [Google Scholar]
  57. Suls J, Wan CK. The relationship between trait hostility and cardiovascular reactivity: A quantitative review and analysis. Psychophysiology. 1993;30:615–626. doi: 10.1111/j.1469-8986.1993.tb02087.x. [DOI] [PubMed] [Google Scholar]
  58. Torres A, Evans WD, Pathak S, Vancil C. Family structure influences cardiovascular reactivity in college students. J Divorce Remarriage. 2001;34:161–177. [Google Scholar]
  59. Treiber FA, Kamarck T, Schneiderman N, Sheffield D, Kapuku G, Taylor T. Cardiovascular reactivity and development of preclinical and clinical disease states. Psychosom Med. 2003;65:46–62. doi: 10.1097/00006842-200301000-00007. [DOI] [PubMed] [Google Scholar]
  60. Vaillancourt T, Miller JL, Fagbemi J, Côté S, Tremblay RE. Trajectories and predictors of indirect aggression: Results from a nationally representative longitudinal study of Canadian children aged 2–10. Aggress Behav. 2007;33:314–326. doi: 10.1002/ab.20202. [DOI] [PubMed] [Google Scholar]
  61. Vella EJ, Friedman BH. Autonomic characteristics of defensive hostility: Reactivity and recovery to active and passive stressors. Int J Psychophysiol. 2007;66:95–101. doi: 10.1016/j.ijpsycho.2007.03.014. [DOI] [PubMed] [Google Scholar]
  62. Waldstein SR, Neumann SA, Burns HO, Maier KJ. Role-played interpersonal interaction: Ecological validity and cardiovascular reactivity. Ann Behav Med. 1998;20:302–309. doi: 10.1007/BF02886379. [DOI] [PubMed] [Google Scholar]
  63. Weich S, Patterson J, Shaw R, Stewart-Brown S. Family relationships in childhood and common psychiatric disorders in later life: systematic review of prospective studies. Br J Psychiatry. 2009;194:392–398. doi: 10.1192/bjp.bp.107.042515. [DOI] [PubMed] [Google Scholar]

RESOURCES