International Journal of Psychophysiology 88 (2013) 317–324

Contents lists available at SciVerse ScienceDirect

International Journal of Psychophysiology

j ourna l homepage: www.e lsev ie r .com/ locate / i jpsycho

Specific dimensions of perceived support and ambulatory blood pressure:Which support functions appear most beneficial and for whom?☆

Kimberly S. Bowen a,⁎, Wendy Birmingham b, Bert N. Uchino a, McKenzie Carlisle a,Timothy W. Smith a, Kathleen C. Light c

a Department of Psychology and Health Psychology Program, University of Utah, United Statesb Huntsman Cancer Institute, University of Utah, United Statesc Department of Anesthesiology, University of Utah, United States

☆ This research was generously supported by grant RHeart, Lung, and Blood Institute (Bert N. Uchino, PI).⁎ Corresponding author at: University of Utah, Depa

1530 E. Rm. 502, Salt Lake City, Utah 84112-0251, Un3176; fax: +1 801 581 5841.

E-mail address: Kimberly.bowen@psych.utah.edu (K

0167-8760/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.ijpsycho.2012.03.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 20 November 2011Received in revised form 20 February 2012Accepted 20 March 2012Available online 29 March 2012

Keywords:Perceived social supportEmotional supportGenderAmbulatory blood pressurePartner responsivenessSelf-disclosureHealth

Perceived support has been related to lower cardiovascular morbidity and mortality. However, little is knownabout the specific functional components of support responsible for such links. We tested if emotional,informational, tangible, and belonging support predicted ambulatory blood pressure (ABP) and interpersonalinteractions (e.g., responsiveness), and if such links were moderated by gender. In this study, 94 marriedcouples underwent 12 h of ABP monitoring during daily life which included a night at home with theirspouse. They completed a short-form of the interpersonal support evaluation list that provides informationon total (global) support, as well as specific dimensions of support. Results revealed that global supportscores did not predict ABP during daily life. However, separating out distinct support components revealedthat emotional support was a significant predictor of lower ambulatory systolic and diastolic blood pressure,primarily for women. Finally, emotional support predicted greater partner responsiveness and self-disclosure, along with less perceived partner negativity although these results were not moderated bygender. These data are discussed in terms of the importance of considering specific support components andthe contextual processes that might influence such links.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Social support is associated with beneficial health outcomes,including reduced rates of mortality and morbidity (Cassel, 1976;Cobb, 1976; Cohen, 1988; Holt-Lunstad et al., 2010; House et al., 1988;Seeman;, 1996; Uchino, 2004). In themost compelling evidence to date,a meta-analysis of 148 studies and over 300,000 participants found thataspects of social support predicted survival with an effect sizecomparable to or larger than traditional medical predictors, such assmoking and exercise (Holt-Lunstad et al., 2010). These effectswere notmoderated by initial health status, age, or cause of death. Social supportis also specifically linked to lower mortality risk due to cardiovasculardisease (Barth et al., 2010; Berkman et al., 1992; Orth-Gomer et al.,1993; Kawachi et al., 1996) which is important because cardiovasculardisease is the leading cause of death in most industrialized countries(Xu et al., 2010).

01 HL68862 from the National

rtment of Psychology, 380 S.ited States. Tel.: +1 801 581

.S. Bowen).

rights reserved.

Despite this epidemiological evidence, relatively less is known aboutthe mechanisms that might be responsible for such links. Ambulatoryblood pressure (ABP) is one promising biological mechanism that mayexplain links between social support and cardiovascular disease. ABPprotocols typically examine cardiovascular functioning in the “realworld” and hence have relatively high ecological validity. Importantly,ABP is a robust predictor of cardiovascular disease risk even afterconsidering clinic resting blood pressure levels (Bjorklund et al., 2004;Perloff et al., 1983; Pickering et al., 1985).

However, few studies have examined if social support predicts ABPand these studies are heterogeneous in terms of findings. In one study,Linden et al. (1993) found perceived social support to predict lower ABPfor women but not for men, whereas Brownley et al. (1996) found thatperceived support predicted lower ABP primarily in high hostileindividuals (Brownley et al., 1996). Another study found social supportwas related to lower ABP only at night (Steptoe et al., 2000). Finally,several studies report no significant link between global social supportand ABP (Holt-Lunstad et al., 2008; Steptoe, 2000; Vella et al., 2008).

Several important conceptual issues arise when considering thelinks between social support and ABP that may help explain thesediverse findings. First, social support can be differentiated into distinctfunctional components (Barrera, 2000; Cohen et al., 1985). Thesefunctional components consist of emotional (self-esteem), informational(appraisal), tangible (or instrumental), and belonging support (Cohen

Table 1Upper triangle of correlation matrix for ISEL subscales for men and women.⁎.

Variable Men Women

1 2 3 4 1 2 3 4

Informational 1.0 0.35 0.29 0.11 1.0 0.35 0.45 0.22Belonging 1.0 0.53 0.49 1.0 0.65 0.35Tangible 1.0 0.45 1.0 0.34Emotional 1.0 1.0

⁎ All correlations were significant at p=.0001.

1 The current paper focuses on perceived functional support. A focus on the qualityof the specific relationship between spouses and ABP is the subject of other papers(Sanbonmatsu et al., 2011; Birmingham et al., in preparation).

318 K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

et al., 1985; Wills and Shinar, 2000). Emotional support entails empathyand positive feedback regarding self-worth. Informational supportinvolves provision of information or advice, which may also be referredto as appraisal support, because the support might involve provision ofinformation that contributes to re-appraisals of a situation as lessstressful or threatening (Cohen et al., 1985). Tangible support comes inthe form of material aid, whereas belonging support includes opportu-nities for social inclusion and recreation.

Importantly, most prior studies in this area have examined globalsupport. This is a significant issue because it has been argued that somecomponents of support are beneficial across situations and hence maybe more closely linked to ABP (Cohen and McKay, 1984; Cohen andWills, 1985; Cutrona and Russell, 1990; Wills, 1991). In particular,emotional support and informational support are hypothesized to bemore likely to have positive influences on health outcomes becausepeople can always benefit from reassurances of their self-worth oruseful information (Cohen andWills; Cohen, 1988). However, only oneprevious study of which we are aware compared specific functionalcomponents of support and their links to ABP (Brownley et al., 1996).In this particular study, emotional support was not included in themeasurement of functional components, yet we know that emotionalsupport is most often predictive of positive-health related outcomes(Heller et al., 1986; Wills, 1991). For example, King et al. (1993)examined the effects of perceived emotional, informational, andtangible support on recovery from coronary artery bypass grafting(CABG). Their data indicated that emotional support was the only formof social support that proved to be consistently predictive of recoveryfrom CABG over the one year study. In another study, high levels ofemotional support predicted lower mortality in older adults whereastangible support actually predicted greater mortality (Penninx et al.,1997; also see Berkman et al., 1992). Thus, one aim of this study was toexamine if different functional components of support would provide amore sensitive test of links to ABP during daily life.

A second aim of this study was to examine potential genderdifferences in links between functional support components and ABP.Social support processes and outcomes may differ in men and women(Kiecolt-Glaser and Newton, 2001; Shumaker and Hill, 1991; Flahertyand Richman, 1989). More specifically, a broader literature suggeststhat emotional support is a particularly important functional compo-nent of social support for women (Burleson, 2003). Females receivinghigh levels of emotional support from family and peers exhibit lowerblood pressure reactivity during stress and increases in psychologicalwell-being (Slavin and Rainer, 1990; Wilson and Ampey-Thornhill,2001; Wilson et al., 1999). Whether due to socialization, social roles,or biological differences, women tend to prefer, seek out, and respondmore positively to emotional support (Burleson, 2003; Flahertyand Richman, 1989; Gilligan, 1982; Shumaker and Hill, 1991; Wills,1985).

Although not a primary aim due to our emphasis on health-relevantbiological pathways, we also examined if functional componentsof support predicted interpersonal processes when couples were athome together. Such data may provide a “snapshot” into the lives ofindividuals relatively high versus low on social support. Perceivedsupport has been related to positive interpersonal processes such asgreater intimacy through self-disclosure, partner responsiveness, andthe interpretation of social interactions (Laurenceau et al., 1998; Lakeyand Cassady, 1990; Reis, 2007). However, most of this work has notfocused on distinct support components or utilized diary experiencesampling. Importantly, examining these processes in the context offunctional support components furthers our understanding of howperceived social support relates to the interpretation of daily interper-sonal interactions. For example, emotional support may be a particu-larly relevant component as it is characterized by validation, positivity,and responsiveness.

The current study thus expands our understanding of social supportprocesses in a naturalistic setting, with measures of ABP, functional

components of support, gender, and interpersonal processes. Partici-pants in our study completed a one day ABP assessment in which ablood pressure reading was randomly taken once during every 30minute period. Based on the aims of the study, we expected thatindividuals with greater informational and emotional support wouldexhibit lower ABP during daily life and more positive interpersonalexchanges. We further expected that emotional support would beassociated with lower ABP in women compared to men. We alsoexpected that women higher in emotional support would evidencemore positive interpersonal processes.

2. Methods

2.1. Participants

As part of a larger program project, 97 healthy heterosexual coupleswere recruited through advertisements placed in local newspapers,workplace newsletters, and flyers distributed around the community.We used the following criteria to select healthy participants based onour prior work (Cacioppo et al., 1995): no existing hypertension, nocardiovascular prescription medication use, no history of chronicdisease with a cardiovascular component (e.g., diabetes), and no recenthistory of psychological disorder (e.g., major depressive disorder).Participants were all legally married and living together with a meanage of 29.6 (see Table 1). Most were White (83%), college educated(62.4%), and had an income over $40,000 per year (66%). Three coupleswho did not follow the study protocol were eliminated from the study,resulting in a total of 94 couples. Participantswere compensated $75 (orextra course credit) for their time.

2.2. Study protocol

Eligible participants arrived at the laboratory on the morning of atypical work day. Height and weight were assessed using a Health-o-Meter scale in order to calculate body mass index to be used as acovariate. Demographic information was collected including age,income, and education and participants completed the InterpersonalSupport Evaluation List (Cohen et al., 1985).

As part of the larger study protocol, participants completed a oneday ABP assessment, typically from 8 am to 10 pm (M=14.01 h,SD=0.97) which included working hours and an evening at homewith the spouse on the same day.1 The ABP monitor was set to take arandom reading once within every 30 minute window. This randominterval-contingent monitoring procedure minimizes participants'anticipation of a blood pressure assessment that might lead them toalter their activities. Following each ABP assessment, individuals wereinstructed to complete questions (ADR, see below) programmed intoa palm pilot device using the Purdue Momentary Assessment Tool(Weiss et al., 2004). Participants were instructed to complete the ADRwithin 5 min of each cuff inflation.

Table 2Characteristics of study participants.

Variable Mean±SD

Age 29.56±8.56a

BMI 25.61±5.13b

Ethnicity 83.07% Caucasian8.99% Latino/Hispanic6.35% Asian-American1.59% Other

Income 66.40% $40,000+11.11% $30,000–39,99913.76% $20,000–29,9998.73% $19,999 or less

319K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

Participants were fitted with the ABP monitor by a trainedresearch assistant and given detailed instructions on how to use it,including how to remove it at the end of the day. They were alsogiven a palm pilot to record diary entries following each bloodpressure reading and detailed instructions on how to use it. Onereading was obtained before the participants left the lab to insure thatthe monitors were working properly and that participants under-stood how to use the palm pilots and how to correctly complete theADR. An appointment to return the equipment and to receivecompensation on the following day was set and participants weredebriefed at the return appointment and all questionnaires andequipment were returned.

Education 62.4% College degree

a In years.b Body Mass Index = weight (kg)/height (m2).

2.3. Measures

2.3.1. Interpersonal Support Evaluation List (ISEL)We used a short-form of the ISEL that contained 15 questions

(Cohen et al., 1985). It assessed the perceived availability of support,as well as the specific dimensions of informational (appraisal, 4items), emotional (self-esteem, 3 items), belonging (4 items), andtangible (4 items) support. In the present study, the internalconsistency of the scale was adequate: .81 for the total scale, .60 fortangible support, .69 for informational support, .57 for emotionalsupport, and .66 for belonging support.2 For correlations among ISELsubscales for males and females, see Table 2.

2.3.2. Ambulatory blood pressureThe Oscar 2 (Suntech Medical Instruments, Raleigh, NC) was used

to estimate ambulatory systolic blood pressure (SBP) and ambulatorydiastolic blood pressure (DBP). The Oscar was developed to meet thereliability and validity standards of the British Hypertension SocietyProtocol (Goodwin et al., 2007). The cuff was worn under theparticipants' clothing, and only a small control box (approximately5.0×3.5×1.5 in.) attached to the participant's beltwas partially exposed.Outliers associated with artifactual readings were identified using thecriteria by Marler et al. (1988). These included: (a) SBPb70 mmHg or>250 mmHg, (b) DBPb45 mmHg or >150 mmHg, and (c) SBP/DBPb[1.065+(.00125×DBP)] or >3.0. Readings were taken once randomlyduring each 30 minute window. Due to the deletion of artifactualreadings (e.g. cuff inflation problems), the mean is slightly longer thenthe targeted 30 minutes window (M=31.15, SD=7.64).

2.3.3. Ambulatory diary record (ADR)Participants were instructed to complete a series of programmed

questions following each ambulatory cardiovascular assessment. TheADR was designed to be easy to complete (about 2–3 min) in order tomaximize cooperation. It contained information on basic variablesthat might influence ABP (Kamarck et al., 1998). These includedposture (lying down, sitting, standing), activity level (1 = no activity,4 = strenuous activity), location (work, home, other), talking (no,yes), temperature (too cold, comfortable, too hot), prior exercise (no,yes), and prior consumption of nicotine, caffeine, alcohol or a meal(no, yes). Measures of interpersonal processes were adapted fromprior work and included 4 items for perceived partner responsive-ness, 2 items for perceived interaction positivity and negativity withthe spouse (Table 7), and 1 item for self-disclosure (Laurenceau et al.,2005; Reis and Wheeler, 1991). Readings were examined to ensurecompliance and were discarded if not instigated within 5 min of ablood pressure reading.

2 The emotional support subscale had the least amount of items (3) which limited itsinternal consistency. We thus analyzed individual items for this subscale to ensure thatthe pattern was in the same direction for each individual item on ABP. Importantly, thesame pattern of results reported below was evident on all items of the subscale.

2.4. Statistical analyses

We utilized PROC MIXED (SAS institute, Littell et al., 1996) in orderto examine ABP and diary ratings of interpersonal processes (seeSchwartz and Stone, 1998). PROC MIXED uses a random regressionmodel to derive parameter estimates bothwithin and across individuals(Singer, 1998). All factors were treated as fixed (Nezlek, 2008) andPROC MIXED treats the unexplained variation within individuals as arandom factor.

One advantage of PROC MIXED is the ability to model moreaccurate covariance structures for the repeated measure assessments.In the present study, we modeled the covariance structure for the tworepeated measures factors of dyad (i.e., husband, wife) and measure-ment occasion (i.e., reading number). Such nested repeated measuresdesigns can be handled in PROC MIXED by specifying separatecovariance structures for each of the factors (Park and Lee, 2002).More specifically, we modeled the covariance structure betweenindividuals of a dyad within each measurement occasion, as well asthe covariance structure acrossmeasurement occasions using the direct(Kronecker) product (Galecki, 1994; Park and Lee, 2002). This directproduct is a within-subjects covariance profile containing the productof the two separate covariance matrices (Galecki, 1994). PROC MIXEDcurrently allows only a few possible combinations for calculating theKronecker product. Based on the recommendations of Park and Lee(2002), we modeled the covariance matrices for dyad and measure-ment occasion using the “type=un@ar(1)” option. Importantly, thismodel allowedus to examine predictors of ABPwhile controlling for thedependency within dyads and measurement occasions. The output ofthese random regression models were parameter estimates (b) withthe appropriate within-subjects covariance structures considered.As recommended by Campbell and Kashy (2002), we used theSatterthwaite approximation to determine the appropriate degrees offreedom.

3. Results

3.1. Preliminary analyses

We first examined potential covariates such as posture or activitylevel that might need to be statistically controlled in the analysis ofABP (Kamarck et al., 1998; Marler et al., 1988). Consistent with priorresearch, results of this initial model revealed that age, gender,household income, body mass, posture, temperature, activity level,prior alcohol, and prior exercise were independent predictors ofhigher ambulatory SBP (p'sb .05). In addition, age, gender, householdincome, body mass, posture, activity level, and a prior mealindependently predicted ambulatory DBP (p'sb .05). Consistent withprior work, these factors along with time (i.e., first reading, secondreading) were thus statistically controlled in all analyses involvingABP (Kamarck et al., 1998).

Table 4Main effects of functional social support on ABP in the gender interaction model.

Variable SBP DBP

b S.E. t df p b S.E. t df p

Informational 2.80 0.96 0.92 594 .36 0.42 0.61 0.23 615 .82Belonging −1.82 0.86 −0.67 590 .50 −1.75 0.54 −1.06 716 .29Tangible 1.61 0.92 0.56 594 .58 1.27 0.58 0.72 720 .47Emotional 6.53 1.07 1.90 599 .06 5.00 0.67 2.40 728 .02⁎

⁎ pb .05.

320 K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

3.2. Do functional components of social support predict ABP?

A primary aim of this study was to test for differences in ABP aspredicted by functional components of social support. However, wefirst examined whether global perceived support predicted ABPlevels. Consistent with the importance of examining specific supportcomponents, global social support did not significantly predict eitherambulatory SBP (b=−1.98, SE=1.30, p=.13) or DBP (b=−0.69,SE=0.82, p=.40). We next examined ABP in analyses wherefunctional components of social support (as assessed by the ISELsubscales of informational, belonging, emotional, and tangible support)were included as separate predictor variables. As shown in Table 3,higher emotional support significantly predicted lower ambulatory SBP(b=−2.82, SE=1.06 p=.008) and DBP (b=−1.52, SE=0.67,p=.025). Higher informational support also predicted lower ambula-tory SBP (b=−1.8, SE=0.96, p=.05), although it did not significantlypredict DBP (b=−.85, SE=0.61, p=.16). Other functional compo-nents of social support (belonging, tangible) were not separatepredictors of either ambulatory SBP or DBP.

Emotional and informational support were both predictors ofambulatory SBP. As a result we also examined their independent linksto ABP. Entering both of these support components into the model atthe same time revealed that emotional support (b=−2.63, SE=1.06,p=.01) but not informational support (b=−1.58, SE=0.96, p=.10)predicted lower ambulatory SBP. These data suggest that emotionalsupport was a stronger predictor of ABP during daily life.

3.3. Does gender moderate the links between functional components ofsocial support and ABP?

A second aim of this study was to test for gender-relateddifferences in ABP as predicted by functional components of socialsupport. We thus examined ABP in analyses where functionalcomponents of social support (informational, belonging, emotional,and tangible support), gender, and the cross-product term (based onthe centered main effects of functional support and gender) werepredictor variables (Table 4). Again, we first examined if global socialsupport interacted with gender in predicting ABP. The interactionbetween global social support and gender did not significantly predictambulatory SBP (b=−3.15, SE=2.51, p=.21) or DBP (b=−1.40,SE=1.58, p=.38).

As shown in Table 5, consistent with our predictions, theinteraction of emotional support and gender predicted ambulatorySBP (b=−5.90, SE=2.07, p=.005) and DBP (b=−4.28, SE=1.30,p=.001). There were no significant interactions between gender andinformational, belonging, or tangible support. As depicted in Fig. 1, wegraphed the form of the significant interaction by plotting predictedambulatory SBP (top panel) and DBP (bottom panel) values forwomen and men one standard deviation above and below the meanfor emotional support. Follow-up analyses blocking on genderrevealed that high levels of emotional support predicted lower ABPfor women (p'sb .001), but not for men (p's> .35).

Table 3Main effects of functional social support on ABP.

Variable SBP

b S.E. t df

Informational −1.84 0.96 −1.93 613Belonging 0.58 0.86 0.67 610Tangible −1.00 0.92 −1.09 619Emotional −2.82 1.06 −2.67 615

⁎ pb .05.

3.4. Do functional components of social support and gender predict dailyinterpersonal processes?

We also tested for differences in interpersonal transactions thatoccur as a function of support components. As part of this study,couples spent an evening at home with each other while measures ofpartner responsiveness, perceived positivity of interactions, perceivednegativity of interactions, and self-disclosure were taken. Controllingfor time, age, and gender, individuals high in global support reportedlower perceived negative interactions (b=−0.21, SE=0.10, p=.04)and greater self-disclosure (b=0.42, SE=0.14, p=.004). Consistentwith our predictions, analyses of specific support componentsrevealed that emotional support most consistently predicted dailylife interpersonal processes in couples (Table 6, 7). Higher emotionalsupport was related to greater perceived partner responsiveness(b=0.28, SE=0.10, p=.006) and self-disclosure (b=0.37, SE=0.12,p=.002), and lower perceived home negativity (b=−0.18,SE=0.08, p=.04). The only other components of support to predictthese daily life processes were belonging support and tangiblesupport. Belonging support predicted greater home positivity(b=0.17, SE=0.08, p=.03) and self-disclosure (b=0.27, SE=0.09,p=.003), whereas tangible support also predicted greater self-disclosure (b=0.21, SE=0.10, p=.04). Given that emotional sup-port, tangible support, and belonging support were all predictors ofself-disclosure, follow-up analyses examined their independents linksby entering all of these support components into the model. Resultsshowed that only emotional support remained a significant predictorof greater self-disclosure (b=0.28, SE=0.13, p=.04).

We next examined whether gender moderated the links betweensupport components and these daily life processes in couples.Contrary to our predictions, no gender X emotional support in-teractions were significant in the prediction of daily interpersonalprocesses. However, gender did moderate the link between globalsupport and partner responsiveness (b=−0.48, SE=0.23, p=.04) andperceived interaction positivity (b=−0.54, SE=0.23, p=02). Thesestatistical interactions with global support primarily reflected theinfluence of belonging support on responsiveness (b=−0.31,SE=0.15, p=.04) and the influence of tangible support on homepositivity (b=−0.36, SE=0.16, p=.03). Follow-up analyses revealedthat these statistical interactions reflected stronger links between social

DBP

p b S.E. t df p

.05 −0.85 0.61 −1.40 769 .16

.50 0.39 0.54 0.72 766 .47

.28 −0.13 0.58 −0.23 768 .82

.01 −1.52 0.67 −2.25 767 .02⁎

Table 5Interaction effects of gender and functional social support on ABP.

Variable SBP DBP

b S.E. t df p b S.E. t df p

Informational −3.02 1.89 −1.60 615 .11 −0.86 1.19 −0.72 748 .47Belonging 1.56 1.68 0.93 612 .36 1.45 1.79 1.37 749 .17Tangible −1.71 1.79 −0.95 623 .34 −0.96 1.13 −0.85 764 .40Emotional −5.90 2.07 −2.85 628 .005⁎⁎ −4.28 1.30 −3.30 775 .001⁎⁎⁎

⁎⁎ pb .01.⁎⁎⁎ pb .001.

321K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

support and greater partner responsiveness/interaction positivity inhusbands (p'sb .05) and not wives (p's>.28).

4. Discussion

The primary aims of this study were to (a) examine the linksbetween distinct functional components of social support and ABP,(b) explore gender differences in these associations. An ancillary aimwas to also examine these questions using measures of dailyinterpersonal processes. Most of our major predictions were sup-ported. The results indicate that higher emotional and informationalsupport predicted lower ABP. Emotional support, in particular,predicted lower ABP for women, but not for men. High emotionalsupport also predicted greater perceived partner responsiveness, self-disclosure, and lower perceived interaction negativity when spendingan evening at home with a spouse. Inconsistent with our hypotheses,the links between emotional support and interpersonal processeswere not moderated by gender. We did find, however, that husbandshigh in belonging support were more likely to perceive greaterpartner responsiveness, whereas husbands high in tangible supportwere more likely to perceive greater interaction positivity.

In accordance with our first hypothesis, higher emotional andinformational support predicted lower ABP. Overall global support andother functional components of social support (tangible, belonging)

b= 0.37, p=.83

b= -5.26, p=.00

115

120

125

130

135

140

Low High

Am

bula

tory

SB

P

Emotional Support

Male

Female

b=0.93, p=.35

b=-3.65, p=.00

73

74

75

76

77

78

79

Low High

Am

bula

tory

DB

P

Emotional Support

Male

Female

Fig. 1. Predicted ambulatory SBP (top panel) and DBP (bottom panel) scores for menand women one SD above and below the mean for emotional support.

were not significantly associatedwith ABP.Moreover, themain effect ofemotional support on ABP was primarily seen in women. These resultsare particularly important to consider given the variable findings inprevious social support andABP studies,most ofwhich examined globalperceptions of support or gender differences (Holt-Lunstad et al., 2008;Linden et al., 1993; Steptoe, 2000; Vella et al., 2008). In addition, studiesthat found a significant link between global perceptions of support andABP tended to have a stronger emotional or informational supportcomponent inmeasurement (Brownley et al., 1996; Linden et al., 1993).Importantly, when both emotional support and informational supportwere examined as predictors at the same time, only emotional supportwas related to ABP. These data suggest that of these two supportcomponents, emotional support may be especially important toconsider in future work.

Consistent with our predictions, women benefited most fromemotional support. These findings are consistent with prior literatureon the preference, utilization, and receipt of emotional support inwomen (Flaherty and Richman, 1989). Whether due to socialization,biological differences, or some other unknown factor, women tend tohave a more interdependent view of the self and devote morepsychological and social resources to establishing and maintainingsocial ties compared to men (Burleson, 2003; Cross and Madson,1997). These data suggest that women are more sensitive to the“affective tone” of relationships. Although women may benefit fromthe positive (especially emotional) aspects, this also implies they aremore adversely influenced by negativity in relationships, which isalso consistent with the prior literature (Kiecolt-Glaser and Newton,2001). However, emotional support is a more nurturing and lesscontrolling form of social support, and thus may activate less of thesenegative aspects, while also promoting relational interdependencethrough empathy and affirmation (Trobst, 2000).

Ancillary analyses also showed that social support predictedinterpersonal processes between husbands and wives during anevening at home. Emotional support was most predictive of theseprocesses, with individuals high in emotional support reportinggreater perceived partner responsiveness, self-disclosure and lowernegativity. However, unlike results for ABP, gender did not moderateany of these associations. These results are consistent with theuncoupling that can occur between self-report and physiologicalprocesses (e.g., Uchino et al., 2012) and highlights the importance of

Table 6Main effects of functional social support on perceived partner responsiveness andself-disclosure in couple interactions at home.

Variable Perceived partner responsiveness Self-disclosure

b S.E. t df p b S.E. t df p

Informational −0.07 0.10 −0.71 293 .48 0.08 0.12 0.68 307 .50Belonging 0.12 0.08 1.56 290 .12 0.27 0.09 2.95 303 .003⁎⁎

Tangible 0.02 0.08 0.22 275 .93 0.21 0.10 2.09 289 .04⁎

Emotional 0.28 0.10 2.76 283 .01⁎⁎ 0.37 0.12 3.09 297 .002⁎⁎

⁎ pb .05.⁎⁎ pb .01.

Table 7Main effects of functional social support on perceived positivity and negativity ofcouple interactions at home.

Variable Positivity Negativity

b S.E. t df p b S.E. t df p

Informational −0.17 0.10 −1.75 292 .08 −0.07 0.08 −0.88 339 .38Belonging 0.17 0.08 2.20 288 .03⁎ −0.10 0.07 −1.53 338 .13Tangible 0.11 0.09 1.28 272 .20 −0.12 0.07 −1.68 319 .09Emotional 0.18 0.10 1.75 282 .08 −0.18 0.09 −2.10 326 .04⁎

⁎ pb .05.

322 K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

data from different levels of analysis (Cacioppo and Petty, 1986). Ofcourse, we were primarily interested in more direct health-relevantpathways as indexed by ABP but future research might consider linksbetween emotional support, gender, and more specific emotionalprocesses (e.g., social emotions, Dickerson and Kemeny, 2004) whichhave been linked to both social situations and health-relevantbiological processes.

We did find that gender moderated some of the associationsbetween perceived social support and interpersonal processes. Morespecifically, husbands high in belonging support were more likely toperceive greater partner responsiveness, whereas husbands high intangible support were more likely to perceive greater interactionpositivity. Although one might expect these links to be stronger inwomen, Neff and Karney (2005) found that wives provided supportin a more responsive manner compared to husbands. As a result,these interpersonal links may be stronger in men because they arepicking-up (accurately) on the responsiveness or positive behaviorsof their wives (Neff and Karney, 2005).

It is important to consider the conceptual delineation betweenperceived and received social support when interpreting thesefindings. The ISEL is a measure of perceived social support — or theinternal perception of one's available support to mobilize if neededwithin his or her network (Cohen et al., 1985). Importantly, perceivedsocial support differs from received social support, which refers toactual supportive interpersonal transactions or exchanges (Sarasonet al., 1990; Kaul and Lakey, 2003; Lakey and Lutz, 1996). These twoconceptualizations are not highly correlated, which suggests they areseparable constructs (Barrera, 2000; Uchino, 2009). This is importantbecause perceived support has more consistent beneficial links tohealth as compared to received support (Uchino, 2009). This may bedue to received support being perceived by the recipient as a threat topersonal independence if unsolicited (Bolger and Amarel, 2007) orcreating social inequity or relational debt (Shumaker and Brownell,1984).

It is also important to note that perceived support was onlymeasured at baseline and we did not have corresponding measures ofreceived support during daily life. So what is captured by ourperceived support measure that explains its links to ABP? Perceivedsocial support is a fairly stable construct and hence might be expectedto be stable during the course of our study (Lakey and Cassady, 1990;Uchino, 2009). As perceived social support is most consistently tied topositive health outcomes, it may be that perceiving social support asavailable, but not necessarily using it, leads to cognitive appraisals ofgreater resources (Wills and Shinar, 2000). Even if such resources arenot utilized, knowing it is there gives the person a greater sense ofcontrol, independence, or self-efficacy, without the downsides ofactually receiving support as mentioned above (Bolger and Amarel,2007; Thoits, 2011). However, perceived support and received supportare correlated so it is possible that individuals during the study werereceiving support. Future research is necessary to examine thispossibility by using more specific assessments of daily support.

There are several limitations of the present study. Because studiescollecting ABP in daily life capture data in the context of daily

processes in participants' natural environments, ABP protocols nicelybridge together epidemiology methods – which give data on realworld outcomes – and laboratory studies — which allow for greatercontrol and manipulation of contextual factors. However, whilelending greater ecological validity to the study, ABP studies alsosacrifice experimental control and the ability to detect cause-and-effect relationships. At most, we are able to examine patterns amongsupport and cardiovascular function which could inform relevantexperimental or intervention studies.

In spite of the aforementioned limitations of the current study,there are a number of strengths. First, dimensions of support wereoften better predictors of ABP compared to global support. Thishighlights the utility of specific functional support approaches infuture work. Second, SBP has been identified as a strong linearpredictor of cardiovascular disease (CVD) for both men and women(Mason et al., 2004; Psaty et al., 2001). Similarly, a 2.5 mm Hgdifference in DBP translates to a 12% reduction in coronary heartdisease risk, regardless of whether an individual is normo- orhypertensive (MacMahon et al., 1990). In the current study, therewas a predicted difference of more than 3 mm Hg between high andlow emotional support for women for both ambulatory SBP and DBP.Thus, if a sustained increase in blood pressure increases risk ofcardiovascular disease, then the current findings suggest clinicallysignificant differences in ABP as a function of social support.

Third, a recent meta-analysis demonstrated that ABP monitoringis more accurate and effective at diagnosing hypertension anddetermining the appropriate intervention than looking only at bloodpressure in the clinic (Hodgkinson et al., 2011). As such, for the firsttime in several decades, the United Kingdom's National Institute onClinical Excellence revised its diagnosis standards (Mayor, 2011).Similarly, a recent computer modeling study suggests ABP monitoringmay be more cost-effective at proper diagnosis of ABP in the long runas compared to traditional diagnosis based on blood pressurereadings in the clinic (Lovibond et al., 2011). Thus, studies of socialsupport consequences on ABP and the underlying pathways for thislink have important health implications.

References

Barrera, M., 2000. Social support research in community psychology. In: Rappaport, J.,Seidman, E. (Eds.), Handbook of Community Psychology. Kluwer Academic/Plenum,New York, NY, pp. 215–245.

Barth, J., Schneider, S., von Känel, R., 2010. Lack of social support in the etiology and theprognosis of coronary heart disease: a systematic review and meta-analysis.Psychosomatic Medicine 72, 229–238.

Berkman, L.F., Leo-Summers, L., Horwitz, R.I., 1992. Emotional support and survivalafter myocardial infarction. Annals of Internal Medicine 117 (12), 1003–1009.

Birmingham, W., Uchino, B.N., Smith, T.W., Light, K.C., & Carlisle, M., in preparation.Spousal ambivalence is linked to elevated ambulatory blood pressure during daily life.

Bjorklund, K., Lind, L., Zethelius, B., Berglund, L., Lithell, H., 2004. Prognosticsignificance of 24-h ambulatory blood pressure characteristics for cardiovascularmorbidity in a population of elderly men. Journal of Hypertension 22 (9),1691–1697.

Bolger, N., Amarel, D., 2007. Effects of social support visibility on adjustment to stress:experimental evidence. Journal of Personality and Social Psychology 92 (3),458–475.

Brownley, K.A., Light, K.C., Anderson, N.B., 1996. Social support and hostility interact toinfluence clinic, work, and home blood pressure in black and white women andmen. Psychophysiol 33, 434–445.

Burleson, B.R., 2003. The experience and effects of emotional support: what the studyof cultural and gender differences can tell us about close relationships, emotion,and interpersonal communication. Pers Relatsh 10, 1–23.

Cacioppo, J.T., Petty, R.E., 1986. The elaboration likelihood model of persuasion. In:Berkowitz, L. (Ed.), Advances in Experimental Social Psychology, vol. 19. AcademicPress, Orlando, FL, pp. 123–205.

Cacioppo, J.T., Malarkey, W.B., Kiecolt-Glaser, J.K., Uchino, B.N., Sgoutas-Emch, S.A.,Sheridan, J.F., Berntson, G.G., Glaser, R., 1995. Heterogeneity in neuroendocrine andimmune responses to brief psychological stressors as a function of autonomiccardiac activation. Psychosomatic Medicine 57, 154–164.

Campbell, L., Kashy, D.A., 2002. Estimating actor, partner, and interaction effects fordyadic data using PROC MIXED and HLM: a user-friendly guide. Pers Relatsh 9,327–342.

Cassel, J., 1976. The contribution of the social environment to host resistance. AmericanJournal of Epidemiology 104, 107–123.

323K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

Cobb, S., 1976. Social support as moderator of life stress. Psychosomatic Medicine 38,300–314.

Cohen, S., 1988. Psychosocial models of the role of social support in the etiology ofphysical disease. Health Psychology 7, 269–297.

Cohen, S., McKay, G., 1984. Social support, stress, and the buffering hypothesis:a theoretical analysis. In: Baum, A., Taylor, S.E., Singer, J.E. (Eds.), Handbook ofPsychology and Health. Lawrence Erlbaum, Hillsdale, NJ, pp. 253–267.

Cohen, S., Wills, T.A., 1985. Stress, social support, and the buffering hypothesis.Psychological Bulletin 98, 310–357.

Cohen, S., Mermelstein, R.J., Kamarck, T., Hoberman, H.M., 1985. Measuring thefunctional components of social support. In: Sarason, I.G., Sarason, B.R. (Eds.),Social Support: Theory, Research, and Applications. Martinus, Nijhoff, The Hague,The Netherlands, pp. 73–94.

Cross, S.E., Madson, L., 1997. Models of the self: self-construals and gender.Psychological Bulletin 122 (1), 5–37.

Cutrona, C.E., Russell, D.W., 1990. Type of social support and specific stress: toward atheory of optimal matching. In: Sarason, B.R., Sarason, I.G., Pierce, G.R. (Eds.), SocialSupport: An Interactional View, Wiley Series on Personality Processes. John Wiley& Sons, Oxford, England, pp. 319–366.

Dickerson, S.S., Kemeny, M.E., 2004. Acute stressors and cortisol responses: atheoretical integration and synthesis of laboratory research. Psychological Bulletin130 (3), 355–391.

Flaherty, J., Richman, J., 1989. Gender differences in the perception and utilization ofsocial support: theoretical perspective and an empirical test. Social Science &Medicine 28 (12), 1221–1228.

Galecki, A.J., 1994. General class of covariance structures for two or more repeatedfactors in longitudinal data analysis. Communications in Statistics — Theory andMethods 23, 3105–3119.

Gilligan, C., 1982. In a Different Voice. Harvard University Press, Cambridge: MA.Goodwin, J., Bilous, M., Winship, S., Finn, P., Jones, S.C., 2007. Validation of the Oscar 2

oscillometric 24-h ambulatory blood pressure monitor according to the BritishHypertension Society Protocol. Blood Pressure Monitoring 12, 113–117.

Heller, K., Swindle, R.W., Dusenbury, L., 1986. Component social support processes:comments and integration. Journal of Consulting and Clinical Psychology 54 (4),466–470.

Hodgkinson, J., Mant, J., Martin, U., Guo, B., Hobbs, F.D.R., Deeks, J.J., Heneghan, C.,Roberts, N., McManus, R.J., 2011. Relative effectiveness of clinic and home bloodpressure monitoring compared with ambulatory blood pressure monitoring indiagnosis of hypertension: systematic review. BMJ 342, d362.

Holt-Lunstad, J., Birmingham, W., Jones, B., 2008. Is there something unique aboutmarriage? The relative impact of marital status, relationship quality, and networksocial support on ambulatory blood pressure and mental health. Annals ofBehavioral Medicine 35 (2), 239–244.

Holt-Lunstad, J., Smith, T.B., Layton, J.B., 2010. Social relationships andmortality risk: a meta-analytic review. PLoS Medicine 7 (7), e1000316. doi:10.1371/journal.pmed.1000316.

House, J.S., Landis, K.R., Umberson, D., 1988. Social relationships and health. Science241, 540–545.

Kamarck, T.W., Shiffman, S.M., Smithline, L., Goodie, J.L., Paty, J.A., Gnys, M., Yi-Kuan, J.J.,1998. Effects of task strain, social conflict, and emotional activation on ambulatorycardiovascular activity: daily life consequences of recurring stress in a multiethnicadult sample. Health Psychology 17, 17–29.

Kaul, M., Lakey, B., 2003. Where is the support in perceived support? The role ofgeneric relationship satisfaction and enacted support in perceived support'srelation to low distress. Journal of Social and Clinical Psychology 22 (1),59–78.

Kawachi, I., Colditz, G.A., Ascherio, A., Brimm, E.B., Giovannucci, E., Stampfer, M.J.,Willett, W.C., 1996. A prospective study of social networks in relation to totalmortality and cardiovascular disease in men in the USA. Journal of Epidemiologyand Community Health 50, 245–251.

Kiecolt-Glaser, J.K., Newton, T.L., 2001. Marriage and health: his and hers. PsychologicalBulletin 127 (4), 472–503.

King, K.B., Reis, H.T., Porter, L.A., Norsen, L.H., 1993. Social support and long-termrecovery from coronary artery surgery. Health Psychology 12 (1), 56–63.

Lakey, B., Cassady, P.B., 1990. Cognitive processes in perceived social support. Journal ofPersonality and Social Psychology 59 (2), 337–343.

Lakey, B., Lutz, C.J., 1996. Social support and preventive and therapeutic interventions.In: Pierce, G.R., Sarason, B.R., Sarason, I.G. (Eds.), Handbook of Social Support andthe Family. Plenum Press, New York, pp. 435–465.

Laurenceau, J.P., Feldman Barrett, L., Pietromonaco, P.R., 1998. Intimacy as aninterpersonal process: the importance of self-disclosure, partner disclosure, andperceived partner responsiveness in interpersonal exchanges. Journal of Person-ality and Social Psychology 74 (5), 1238–1251.

Laurenceau, J.P., Feldman Barrett, L., Rovine, M.J., 2005. The interpersonal processmodel of intimacy in marriage: a daily-diary and multilevel modeling approach.Journal of Family Psychology 19, 314–323.

Linden, W., Chambers, L., Maurice, J., Lenz, J.W., 1993. Sex differences in social support,self-deception, hostility, and ambulatory cardiovascular activity. Health Psychol-ogy 12 (5), 376–380.

Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1996. SAS System for MixedModels. SAS institute Inc., Cary, NC.

Lovibond, K., Jowett, S., Barton, P., Caulfield, M., Heneghan, C., Hobbs, R., Hodgkinson, J.,Mant, J., Martin, U., Williams, B., Wonderling, D., McManus, R., 2011. Cost-effectiveness of options for the diagnosis of high blood pressure in primary care: amodeling study. Lancet 378, 1219–1230.

MacMahon, S., Peto, R., Cutler, J., Collins, R., Sorlie, P., Neaton, J., Abbott, R., Godwin, J.,Dyer, A., Stamler, J., 1990. Blood pressure, stroke, and coronary heart disease: part

1, prolonged differences in blood pressure: prospective observational studiescorrected for the regression dilution bias. Lancet 335, 765–774.

Marler, M.R., Jacob, R.G., Lehoczky, J.P., Shapiro, A.P., 1988. The statistical analysis oftreatment effects in 24-hour ambulatory blood pressure recordings. Statistics inMedicine 7, 697–716.

Mason, P.J., Manson, J.E., Sesso, H.D., Albert, C.M., Chown, M.J., Cook, N.R., Greenland, P.,Ridker, P.M., Glynn, R.J., 2004. Blood pressure and risk of secondary cardiovascularevents in women: the Women's Antioxidant Cardiovascular Study (WACS).Circulation 109 (13), 1623–1629.

Mayor, S., 2011. Hypertension diagnosis should be based on ambulatory blood pressuremonitoring, NICE recommends. BMJ 343, d5421.

Neff, L.A., Karney, B.R., 2005. Gender differences in social support: a question of skill orresponsiveness? Journal of Personality and Social Psychology 88 (1), 79–80.

Nezlek, J.B., 2008. An introduction to multilevel modeling for social and personalitypsychology. Social and Personality Psychology Compass 2, 842–860.

Orth-Gomer, K., Rosengren, A., Wilhelmsen, L., 1993. Lack of social support andincidence of coronary heart disease in middle-aged Swedish men. PsychosomaticMedicine 55 (1), 37–43.

Park, T., Lee, Y.J., 2002. Covariance models for nested repeated measures data: analysisof ovarian steroid secretion data. Statistics in Medicine 21, 143–164.

Penninx, B.W.J.H., van Tilburg, T., Kriegsman, D.M.W., Deeg, D.J.H., Boeke, A.J.P., vanEijk, J.T.M., 1997. Effects of social support and personal control resources onmortality in older age: the Longitudinal Aging Study Amsterdam. American Journalof Epidemiology 146, 510–519.

Perloff, D., Sokolow, M., Cowan, R., 1983. The prognostic value of ambulatory bloodpressure. JAMA : The Journal of the American Medical Association 249, 2793–2798.

Pickering, T., Harshfield, G., Devereux, R., Laragh, J., 1985. What is the role ofambulatory blood pressure monitoring in the management of hypertensivepatients? Hypertension 7, 171–177.

Psaty, B.M., Furberg, C.D., Kuller, L.H., Cushman, M., Savage, P.J., Levine, D., O'Leary, D.H.,Bryan, R.N., Anderson, M., Lumley, T., 2001. Association between blood pressurelevel and the risk of myocardial infarction, stroke, and total mortality. Archives ofInternal Medicine 161, 1183–1192.

Reis, H.T., 2007. Steps toward the ripening of relationship science. Pers Relatsh 14,1–23.

Reis, H.T., Wheeler, L., 1991. Studying social interaction with the Rochester InteractionRecord. In: Zanna, M.P. (Ed.), Advances in Experimental Social Psychology, vol. 24.Academic Press, San Diego, CA, pp. 270–318.

Sanbonmatsu, D.M., Uchino, B.N., Birmingham, W., 2011. On the importance ofknowing your partner's views: attitude familiarity is associated with betterinterpersonal functioning and lower ambulatory blood pressure in daily life.Annals of Behavioral Medicine 41, 131–137.

Sarason, I.G., Sarason, B.R., Pierce, G.R. (Eds.), 1990. Social Support: An InteractionalView. John Wiley & Sons, Inc., New York.

Schwartz, J.E., Stone, A.A., 1998. Strategies for analyzing ecological momentaryassessment data. Health Psychology 17, 6–16.

Seeman, T.E., 1996. Social ties and health: the benefits of social integration. Annals ofEpidemiology 6 (5), 442–451.

Shumaker, S.A., Brownell, A., 1984. Toward a theory of social support: closingconceptual gaps. Journal of Social Issues 40 (4), 11–36.

Shumaker, S.A., Hill, D.R., 1991. Gender differences in social support and physicalhealth. Health Psychology 10 (2), 102–111.

Singer, J.D., 1998. Using SAS PROC MIXED to fit multilevel models, hierarchical models,and individual growth models. Journal of Educational and Behavioral Statistics 23,323–355.

Slavin, L.A., Rainer, K.L., 1990. Gender differences in emotional support and depressivesymptoms among adolescents: a prospective analysis. American Journal ofCommunity Psychology 18 (3), 407–420.

Steptoe, A., 2000. Stress, social support and cardiovascular activity over the workingday. International Journal of Psychophysiology 37 (3), 299–308.

Steptoe, A., Lundwall, K., Cropley, M., 2000. Gender, family structure and cardiovascularactivity during the working day and evening. Social Science & Medicine 50 (4),531–539.

Thoits, P.A., 2011. Mechanisms linking social ties and support to physical and mentalhealth. Journal of Health and Social Behavior 52 (2), 145–161.

Trobst, K.K., 2000. An interpersonal conceptualization and quantification of socialsupport transactions. Personality and Social Psychology Bulletin 26 (8), 971–986.

Uchino, B.N., 2004. Social Support and Physical Health: Understanding the HealthConsequences of Our Relationships. Yale University Press, New Haven, CT.

Uchino, B.N., 2009. Understanding the links between social support and health: a life-span perspective with emphasis on the separability of perceived and receivedsupport. Perspectives on Psychological Science 4, 236–255.

Uchino, B.N., Bowen, K., Carlisle, M., Birmingham, W., 2012. Psychological pathwayslinking social support to health outcomes: A visit with the "ghosts" of researchpast, present, and future. Social Science & Medicine 74 (7), 949–957.

Vella, E.J., Kamarck, T.W., Shiffman, S., 2008. Hostility moderates the effects of socialsupport and intimacy on blood pressure in daily social interactions. HealthPsychology 27 (2), S155–S162.

Weiss, H.M., Beal, D.J., Lucy, S.L., MacDermid, S.M., 2004. Constructing EMA Studieswith PMAT: The Purdue Momentary Assessment Tool User's Manual. PurdueUniversity Military Family Research Institute, West Lafayette, IN.

Wills, T.A., 1985. Supportive functions of interpersonal relationships. In: Cohen, S., Syme,L.S. (Eds.), Social Support and Health. Academic Press, San Diego, CA, pp. 61–82.

Wills, T.A., 1991. Social support and interpersonal relationships. In: Clark, M.S. (Ed.),Prosocial Behavior: Review of Personality and Social Psychology, vol. 12. SagePublications, Inc., Thousand Oaks, CA, pp. 265–289.

324 K.S. Bowen et al. / International Journal of Psychophysiology 88 (2013) 317–324

Wills, T.A., Shinar, O., 2000. Measuring perceived and received social support. In:Cohen, S., Gordon, L., Gottlieb, B. (Eds.), Social Support Measurement andIntervention: A Guide for Health and Social Scientists. Oxford University Press,New York, NY, pp. 86–135.

Wilson, D.K., Ampey-Thornhill, G., 2001. The role of gender and family support ondietary compliance in an African American adolescent hypertension preventionstudy. Annals of Behavioral Medicine 23 (1), 59–67.

Wilson, D.K., Kliewer, W., Bayer, L., Jones, D., Welleford, A., Heiney, M., Sica, D.A., 1999.The influence of gender and emotional versus instrumental support on cardiovas-cular reactivity in African-American adolescents. Annals of Behavioral Medicine 21(3), 235–243.

Xu, J., Kochanek, K.D., Murphy, S.L., Tejada-Vera, B., 2010. Deaths: final data for 2007.National Vital Statistics Reports 58 (19).