Romantic Relationship Satisfaction and Ambulatory Blood Pressure During Social Interactions: Specificity or Spillover Effects?

Romantic Relationship Satisfaction and Ambulatory Blood Pressure During Social Interactions:... Abstract Background People in high-quality romantic relationships tend to have lower blood pressure (BP). People may experience lower BP specifically when interacting with romantic partners. Purpose This study parsed the effects of different types of social interactions on ambulatory BP (ABP) and tested whether romantic relationship satisfaction moderated these effects during interactions with partners in particular (specificity) or with others (spillover; e.g., friends, co-workers). Methods Partnered participants (N = 594) were drawn from a larger study on BP and cardiovascular health (age = 46.5 ± 9.3; 57.4% female). Participants reported on romantic relationship satisfaction and completed 24-hr ABP monitoring. At each reading, participants reported whether they had a social interaction and with whom. Multilevel models accounted for nesting of data over time. Results Romantic relationship satisfaction significantly modified the effects of some social interactions on systolic and diastolic BP (SBP, DBP). Participants with high (+1 SD) relationship satisfaction had significantly lower SBP (−0.77 mmHg, p = .02) during partner interactions compared with no social interaction; low-satisfaction (−1 SD) participants had a nonsignificant 0.59 mmHg increase (p = .14). A similar pattern emerged for DBP. Relationship satisfaction also modified SBP response during friend interactions (elevated SBP for low-satisfaction participants) and DBP response during “other” interactions (elevated DBP for high-satisfaction participants). Conclusion Participants with high levels of romantic relationship satisfaction experienced significantly lower BP during social interactions with their partner compared with situations without social interaction. Although there was some evidence for spillover to other types of relationships, effects were largely restricted to partner interactions. Relationship satisfaction, Ambulatory blood pressure, Blood pressure, Couples, Social interaction, Ecological momentary assessment Marriage is robustly associated with better cardiovascular outcomes in the long term [1–3] and predicts lower blood pressure (BP) on a daily basis [2, 4–6]. Yet all marriages are not created equal. The quality of a relationship matters, with higher quality relationships being uniquely associated with improved cardiovascular health (e.g., decreased left-ventricular mass, lower BP [7–9]). However, the mechanisms underlying the effects of marriage and marital quality on cardiovascular outcomes are not fully understood. It may be that there is a general, beneficial effect of high-quality relationships on cardiovascular outcomes, including lower average BP. On the other hand, people may experience lower BP specific to situations when interacting with a romantic partner and such effects may accumulate to predict better cardiovascular outcomes. These possibilities are suggested by research indicating that marital satisfaction and cohesion predict long-term outcomes (left-ventricular mass) and are associated with increased spousal contact [7]. Finally, it is possible that high-quality romantic relationships may have a broad impact on reactions to social situations, shaping the effects of social interactions on cardiovascular outcomes across relationship types. For instance, a satisfying romantic relationship might buffer effects of potentially stressful interactions with supervisors or co-workers, reflecting a positive “spillover” to nonpartner social interactions. This delineates three possibilities: the effect of high-quality relationships on cardiovascular health may (1) be specific to romantic partner interactions, (2) spill over to affect the nature of other social interactions, or (3) be general (i.e., function across all types of social interactions and also situations when one is not engaged in a social interaction). There is surprisingly little direct evidence about the influence of relationship satisfaction with one’s partner (i.e., spouse, significant other) on BP specific to interactions with one’s partner, and almost none concerning the influence of romantic relationship satisfaction on cardiovascular responses to interactions with specific other people (i.e., “spillover” effects). For example, although Gump et al. [4] examined BP during social interactions and found that romantic partner interactions were associated with significantly lower BP as compared with no interaction or interactions with others, “others” were not differentiated (e.g., co-worker vs. family interactions). This lack of differentiation could muddy the waters in understanding whether romantic relationship satisfaction moderates the effects of some types of interactions on BP, but not others (e.g., co-workers vs. strangers). Similarly, Joseph et al. [9] compared the positivity of a dichotomy, marital versus nonmarital interactions, and found that only positive marital interactions predicted lower carotid artery intima medial thickness. Taken together, these studies suggest that the BP-lowering effects of a high-quality romantic relationship may be specific to partner interactions as compared with other types of interactions (i.e., a “specificity” effect). Yet, the possibility of a “spillover” effect—that satisfaction with a primary romantic relationship may exert cardioprotective effects in other types of relationships (e.g., co-workers, friends)—has not been rigorously tested across relationship types. Indirect evidence from literature on social support suggests that spillover may be possible. For example, social support more broadly can serve as a buffer against the negative effects of momentary stress on BP [10], which might include potentially stressful interactions, although the buffering effect of social support on BP may be isolated to the lab [11]. Further support for romantic relationship-specific effects on cardiovascular risk markers exists at the other end of the satisfaction spectrum. Marital distress has been associated with higher BP at home, but not at work [12], indicating that marital distress may matter most in situations in which the romantic partner is most relevant (although the actual presence of a partner was not assessed). These results are suggestive, but they rely on a simple approach that assumes presence or absence of the romantic partner, and findings are mixed (e.g., the effect of marital ambivalence on BP did not differ by location and thus may not be specific to spousal interactions [13]). The possibilities of general, specific, and spillover effects of romantic relationship satisfaction on BP have not yet been tested in a single sample. Furthermore, because some of the specificity effects from prior research described above must be inferred, there is not yet direct evidence within a single, ecologically valid study of whether this effect is truly specific to partner interactions. A more fine-grained approach to understanding the influence of romantic relationship satisfaction on the cardiovascular response to both romantic partner and other social interactions is warranted. In this study, we estimated participants’ ambulatory BP (ABP) during social interactions with both romantic partners and others, compared with situations involving no social interaction, and tested whether satisfaction with the primary romantic relationship was related to differential BP levels during those interactions. We hypothesized that participants who were highly satisfied with their romantic relationship would have reduced BP in general and especially so during interactions with their partner (i.e., “specificity” effects). In addition, we explored the possibility that BP might be reduced during social interactions with nonpartners in people reporting high romantic relationship satisfaction (vs. no interaction or vs. low relationship satisfaction; i.e., “spillover” effects). Methods Participants Participants were drawn from a larger study examining the intersection of BP in the clinic setting, ABP, and cardiovascular health (Masked Hypertension Study [MHTN]; see [14] for further details). To be eligible for the primary study, participants were required to be (i) 21+ years old, (ii) working 17.5+ hr per week, (iii) able to speak and read English, (iv) not taking BP-lowering medications, and (v) have a pre-enrollment screening BP of less than 160/105 mmHg. Participants were excluded if they reported a history of cardiovascular disease (e.g., congestive heart failure, myocardial infarction), chronic renal disease, liver disease, thyroid disease, or adrenal disease, cancer that was not in remission for a minimum of 6 months, active substance abuse, or a serious mental health illness. Participants were also excluded if they had evidence of secondary hypertension, were taking any cardiovascular medication other than statins, or were pregnant. Study procedures were approved by the Columbia University Medical Center Institutional Review Board and the Stony Brook University Institutional Review Board, and participants completed informed consent before completing study procedures. For the present study, participants (N = 594) who completed both ABP monitoring and a measure of romantic relationship satisfaction were included; participants did not complete the measure of romantic relationship satisfaction if they said they were not married and not living with a partner. Individuals, and not couples, were recruited to participate in the study, although a small number of couples (<5) did participate. Procedure The MHTN [14] study comprised five visits, of which two are relevant to the current study. At the first visit, participants were given a packet of psychosocial questionnaires, including the measure of romantic relationship satisfaction, and asked to complete and return it within two weeks (i.e., by the time of their third visit). At the third visit, participants were fitted with an ABP monitor programmed to take readings every 28–30 min for 24 hr. Participants were also trained in the use of an electronic diary for ecological momentary assessments (EMAs) that asked about recent social interactions, and were asked to complete a diary entry immediately following each BP reading (i.e., fixed, interval-contingent reporting). Momentary covariates known to affect ABP readings (e.g., posture, location) [6, 15] were also assessed by the diary. ABP readings and EMAs spanned two workdays for all participants (no weekend days). Although nocturnal ABP data were available, sleep BP readings were excluded from the analysis because the focus of this analysis was on the effect of different types of social interactions on ABP, and EMAs (and, ostensibly, social interactions) were not completed during sleep. Measures Romantic relationship satisfaction Relationship satisfaction with a current romantic partner was measured using a 32-item modified version of the Dyadic Adjustment Scale (DAS [16]). Example items included rating how often the participant and his or her partner agreed on demonstrations of affection (0 = always disagree to 5 = always agree) and how often the participant thought things were going well with his or her partner (0 = never to 5 = all the time; α = .84). ABP ABP readings were taken every 28–30 min using an ABP monitor (SpaceLabs, Model 90207, Snoqualmie, WA, USA). The monitor was fitted on the nondominant arm using an appropriate-sized cuff and was worn for 24 consecutive hours. Participants returned the BP monitor and electronic diary at the end of the 24-hr monitoring period. Per the recommendations of the device manufacturer, systolic BP (SBP) readings were required to fall within the range of 70–250 mmHg and diastolic BP (DBP) readings within the range of 40–150 mmHg while awake (30–150 during sleep); the small number of readings outside of these ranges were treated as errors (i.e., missing values). Social interaction Following each ABP reading, participants completed the electronic diary EMA noting whether they had had a social interaction at any point since the last reading (yes or no) and with whom. Options for type of social interaction were spouse/significant other, family, friends, clients, public, doctor/nurse, supervisor, co-worker, and other, and more than one response could be checked at any given assessment (to retain all interaction information, indicator variables were constructed for the presence/absence of each type of interaction; thus, the analysis estimated separate effects for each type of interaction, controlling for the presence/absence of all other types). Separate indicator variables (0 = absent, 1 = present) were created for each of the nine types of interaction; reference category corresponds to no social interaction (0 on all nine indicators). Data Analysis Strategy The diary reports and BP readings entail multiple observations from a single individual, violating traditional assumptions of independence. Furthermore, observations from an individual that are closer together in time are likely to be more similar to each other than those assessed further apart (serial autocorrelation). To accommodate the nested structure of the data, multilevel models were estimated using the MIXED procedure in SAS. The specified error structure accounted for both autocorrelation (i.e., increased similarity of observations close together in time) and random error using the “local” subcommand [17], and the intercept and social interaction (yes or no) were treated as random effects to account for unexplained individual differences in mean level of BP and in the effects of social interaction on BP. Two models were estimated for each outcome (SBP, DBP). In the first model, the main effects of relationship satisfaction (person-level factor) and the dichotomous indicator variables identifying types of social interaction (momentary factors, reference situation corresponds to “no interaction” when all indicator variables equal zero) were included. In the second model, the multiplicative interaction terms of romantic relationship satisfaction with each of the interaction type indicators were entered to test the moderating effect of relationship satisfaction with one’s spouse/partner on the relationship of each type of social interaction with BP. We additionally estimated a planned contrast comparing the multiplicative interaction term of romantic relationship satisfaction with interacting with a partner to the multiplicative interaction term of romantic relationship satisfaction with all other types of social interactions; as such, a significant, negative contrast would provide evidence for a specificity effect (i.e., romantic relationship satisfaction affects BP response primarily [or more] during partner interactions), whereas a nonsignificant contrast would provide evidence for spillover (i.e., romantic relationship satisfaction affects BP response similarly, on average, during both partner and nonpartner interactions). All models controlled for age, race, ethnicity, and BMI (person level factors), and posture, exertion, temperature, location, and recent consumption of a meal, caffeine, alcohol, or nicotine (momentary factors) which have been shown to affect BP [6, 15]. Results The 594 participants included in the present study represented 66.5% of the 893 who completed the ABP procedures. The mean number of valid awake ABP readings, either at the initial attempt or a retry 2 min later was 32.3 (SD = 5.0; median = 33, quartiles: 30, 35). This represented 92.5% of all attempted awake readings. The mean number of valid awake readings which had a “concurrent” EMA report (i.e., completed within 8 min of the ABP reading) was 24.5 observations for each participant (SD = 7.3; median = 26, quartiles: 20, 30), corresponding to a mean rate of 75.5% (SD = 18.4%; median = 80.0%, quartiles: 66.7, 88.6%). Mean age was 46.5 (SD = 9.3), and mean BMI was 27.7 (SD = 5.2). Participants were predominantly female (57.4%), 87.7% were married, 6.1% were Black, and 10.3% were Hispanic. The romantic relationship satisfaction scale had a mean of 108.2 (SD = 20.8; range = 9, 147); to keep this on a similar scale to other variables and improve interpretation this was divided by 10. Of 14,545 awake BP readings with a concurrent EMA report, 8,854 involved a reported interaction (60.9%). Frequencies of each interaction type are reported in Table 1. Table 1 Frequency of different interaction types across the N = 8,854 reported interactions Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  5,691 observations included no interactions. View Large Table 1 Frequency of different interaction types across the N = 8,854 reported interactions Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  5,691 observations included no interactions. View Large SBP Covariates predicted SBP in the expected directions (e.g., older participants and males had higher SBP, alcohol and nicotine use predicted higher SBP, greater physical exertion predicted higher SBP). In the first model predicting SBP, counter to our hypothesis about the general BP effect of romantic relationship satisfaction, there was a positive main effect of romantic relationship satisfaction on SBP; as relationship satisfaction increased, (mean) SBP also increased, B = 0.41 [0.05, 0.78] mmHg/10-point increase in relationship satisfaction, p = .03. The main effect of social interaction with a partner (compared with no social interaction) on SBP, B = −0.20 [−0.67, 0.28] mmHg, p = .42, was not significant; however, when participants reported interacting with a friend, co-worker, or “other,” they had higher SBP, Bs = 1.40, 0.46, and 1.24 mmHg, respectively. In contrast, interacting with a doctor or nurse was associated with lower SBP, B = −1.21 [−2.27, −0.15] mmHg, p = .03. In the second model, examining potential effect modification, relationship satisfaction significantly modified the effect of some types of social interactions on SBP (see Table 2). Specifically, romantic relationship satisfaction moderated the effect of interacting with a partner, B = −0.33 [−0.59, −0.06], p = .01, revealing the hypothesized BP-lowering effect of a positive romantic partnership during interactions with one’s partner. Participants with high (+1 SD) relationship satisfaction had significantly decreased SBP during interactions with a partner, B = −0.77 [−1.44, −0.10] mmHg, p = .02, compared with when there was no social interaction. In contrast, participants with low satisfaction (−1 SD) had nonsignificantly higher SBP, B = .59 [−.20, 1.37] mmHg, p = .14 (see Fig. 1). Fig. 1. View largeDownload slide Systolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Fig. 1. View largeDownload slide Systolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Participants’ relationship satisfaction with their partner also modified the SBP response to interacting with a friend, B = −0.49 [−0.80, −0.19], p < .01. The tendency for SBP to be higher during interactions with friends was amplified in those with low romantic relationship satisfaction (−1 SD; B = 2.40 [1.51, 3.28] mmHg, p < .001), and attenuated in those with high relationship satisfaction (+1 SD; B = 0.34 [−0.57, 1.26] mmHg, p = .46). The relationships of other types of social interactions with SBP (i.e., family, clients, public, doctor/nurse, supervisor, co-worker, and other) were not significantly modified by romantic relationship satisfaction. Further analyses (details not shown) provided no evidence that the moderating effect of romantic relationship satisfaction on the association between different types of social interactions and SBP differed by age or gender. The planned contrast comparing the moderating effect of romantic relationship satisfaction with partner interactions versus all other types of interactions predicting SBP was negative and significant, B = −0.39 [−0.68, −0.11], p < .01, providing additional evidence for specificity. DBP Covariates predicted DBP in the expected directions (e.g., males and older participants had higher DBP, alcohol and nicotine use predicted higher DBP, and greater physical exertion predicted higher DBP). In the first model predicting DBP, there was no effect of romantic relationship satisfaction, B = 0.09 [−0.18, 0.37] mmHg, p = .50, or interacting with a partner, B = 0.18 [−0.21, 0.58] mmHg, p = .36, on DBP. Interactions with a friend, client, supervisor, co-worker, or “other” were associated with higher DBP, Bs = 1.10, 0.78, 0.71, 0.55, and 0.74 mmHg, respectively. The second model tests whether romantic relationship satisfaction moderates the effect of different types of social interactions on DBP (see Table 2). In parallel to the negative coefficient for the interaction term for romantic relationship satisfaction and interacting with a romantic partner when predicting SBP, this interaction term’s coefficient was also negative and statistically significant when predicting DBP, B = −0.34 [−0.56, −0.13], p < .01. Simple slopes analysis showed that interacting with a significant other predicted an increase in DBP when romantic relationship satisfaction was low (−1 SD), B = 1.01 [0.37, 1.65] mmHg, p < .01, but had a nonsignificant, negative effect when relationship satisfaction was high (+1 SD), B = −0.19 [−0.97, 0.13] mmHg, p = .14 (see Fig. 2). Table 2 Coefficients and 95% CIs for the association of romantic relationship satisfaction (grand mean centered) with different types of social interactions predicting systolic blood pressure (SBP) and diastolic blood pressure (DBP)   SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*    SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*  +p < .10, *p < .05, **p < .01. View Large Table 2 Coefficients and 95% CIs for the association of romantic relationship satisfaction (grand mean centered) with different types of social interactions predicting systolic blood pressure (SBP) and diastolic blood pressure (DBP)   SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*    SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*  +p < .10, *p < .05, **p < .01. View Large Fig. 2. View largeDownload slide Diastolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Fig. 2. View largeDownload slide Diastolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Romantic relationship satisfaction also modified the effect of interacting with someone in the “other” category, B = 0.39 [0.08, 0.69], p = .01. Unexpectedly, interacting with an “other” was associated with higher DBP when participants reported high (+1 SD) relationship satisfaction, B = 1.48 [0.57, 2.39] mmHg, p < .01, whereas the association was nonsignificant when participants reported low (−1 SD) satisfaction, B = −0.14 [−1.10, 0.83] mmHg, p = .78. The association between other types of social interactions and DBP (i.e., family, friends, clients, public, doctor/nurse, supervisor, and co-worker) were not significantly modified by romantic relationship satisfaction. There was no evidence that the moderating effect of romantic relationship satisfaction on the association between different types of social interactions and DBP differed by age or gender. The planned contrast comparing the moderating effect of romantic relationship satisfaction with partner interactions versus all other types of interactions predicting DBP was negative and significant, B = −0.37 [−0.60, −0.14], p < .01, providing additional evidence for specificity. Discussion This is one of the largest studies to examine the link between different types of social interactions—at different levels of romantic relationship satisfaction—and BP. We examined BP response to social interactions with both a romantic partner and other individuals over the course of a normal work day in a large sample of healthy adults. Furthermore, we examined whether participants’ relationship satisfaction with their primary romantic relationship moderated the association between different types of social interactions and BP response, and whether there was a general effect of romantic relationship satisfaction on BP. In support of study hypotheses, participants with high levels of romantic relationship satisfaction experienced significantly lower BP during social interactions with their partner, compared with those with low levels of relationship satisfaction. Counter to study hypotheses, there was no evidence for a general effect of high romantic relationship satisfaction lowering BP. In fact, for SBP but not DBP, higher relationship satisfaction was associated with increased BP as a main effect. The most consistent finding was the moderation effect for romantic relationship satisfaction and partner interactions (i.e., “specificity”). Participants with high levels of relationship satisfaction experienced lower SBP during partner interactions, relative to no interactions (those with low levels of relationship satisfaction did not). Participants with low levels of relationship satisfaction experienced increased DBP during partner interactions, relative to no interactions, whereas those with high levels of relationship satisfaction did not; further, tests suggested that this effect was most pronounced during partner interactions (vs. all other types of interactions). In sum, for SBP and DBP, romantic relationship satisfaction was associated with either a decrease or lack of increase in BP during partner interactions. Although there was some support for the specificity hypothesis (i.e., that romantic relationship satisfaction would primarily modify BP response during interactions with one’s partner), spillover effects could not be ruled out. For example, interacting with a friend was associated with higher SBP when relationship satisfaction was low, but was not associated with SBP when relationship satisfaction was high. The moderation effect for friend interactions mirrors the direction for the moderation effect for interactions with a significant other; in both cases, SBP decreases more (or increases less) during social interactions for those with high romantic relationship satisfaction. The reason for these parallel findings is not clear. There may be a general tendency for people who are satisfied in one type of close relationship (i.e., a romantic relationship) to also be satisfied in other types of close relationships (i.e., friendships) and to garner similar BP benefits from each. Indeed, people who have more secure attachment styles may have closer and more satisfying relationships generally [18–20] such that there are “satisfaction” effects across different types of social interactions. Similarly, other personality differences that can affect relationship or interaction quality across multiple types of relationships (e.g., depressive symptoms, hostility) may also play a role [21]. We did not assess attachment style or the quality of participants’ relationships with friends or family, however, so we cannot know from these data whether their satisfaction with nonromantic others played a role. A more speculative possibility is that people in lower quality relationships who report low levels of romantic relationship satisfaction may vent to their friends, and so friend interactions may be more fraught. This is also suggested by attachment research indicating that anxiously attached individuals may self-disclose more to their friends [19]. It would be interesting to follow up with a study including conversation content to specifically code for relationship-related talk and to investigate subsequent effects on physiological responses. Future research might also measure attachment styles to incorporate a more nuanced analysis of spillover effects. A second, unexpected “spillover” effect concerned the fact that people in more satisfying romantic relationships exhibited higher DBP when interacting with people in the “other” category. It is unclear why this might happen, and it is difficult to speculate given the lack of clarity as to who “others” were (particularly given the wide range of possible categories of interaction partners). Still, it is important to note that the nature of this relationship was in the opposite direction of what was predicted (i.e., we found higher BP with greater relationship satisfaction, rather than lower BP). Allowing open-ended responses for the “other” category in future research might shed light on this phenomenon. The effect may also have emerged due to chance, however, suggesting caution in pursuing this without a firm, theoretical basis. Placing this study in the context of the broader literature on close relationships and cardiovascular risk, another key finding was that there was no evidence that a high-quality romantic relationship related to lower BP in general. Rather, greater romantic relationship satisfaction was associated with higher average SBP. This is unexpected, given the previously published association between higher quality relationships and better cardiovascular outcomes [2, 7–9], and suggests the importance of examining effects of relationship satisfaction across varying timescales, including a consideration of both short- and long-term mechanisms driving these effects in potentially opposite directions (e.g., immediate effects, trait-level effects, accumulated experiences). Research finding main effects of relationship quality on lowered cardiovascular risk long-term may be due either to stronger effects of close relationships on other mechanistic pathways, or to accumulated effects of satisfying partner interactions on BP [7, 9]. One hypothesis consistent with the latter explanation is that the beneficial effects of daily interactions on BP may accumulate and be amplified if romantic relationship satisfaction leads to increased contact with the significant other [3, 4]. It would be informative to test a series of models, in which (1) relationship satisfaction is included as a moderator of BP response to interactions with a significant other and (2) frequency of contact over an extended period of time is examined as a mediator of a satisfying romantic relationship’s positive effect on long-term cardiovascular outcomes. Such evidence would also provide critical mechanistic insight for the creation of couples-focused prevention programs to improve cardiovascular health. Although this study has many strengths, including a large sample size and strong ecological validity, there are a number of limitations. The sample in this study consisted of people who were working 17.5 hr a week or more, which limits the generalizability of the results. The sample was also restricted to nonhypertensive individuals. An important next step would be to replicate these effects in a hypertensive sample. There were also a small number of couples (4, 1.3% of the sample). Because participants were enrolled and studied as individuals and the number of couples is so small, we decided to ignore this small degree of interdependence. Another limitation of the study is that participants were asked to complete only 24 hr of ABP monitoring, which limited the variety of social interactions each participant was likely to experience. Future studies should consider the specific effects of different types of social interactions across a longer time period. Other potential explanatory variables (e.g., attachment, individual personality differences that may affect social interactions, such as hostility, satisfaction with different types of social relationships) should also be explored. That said, the primary goal of this analysis was to test the specific hypothesis that the effect on BP of interacting with one’s partner would be associated, in a protective manner, with romantic relationship satisfaction—that is, a factor that is specific to one’s relationship with that partner. The present analyses provide support for this hypothesis, although the observational nature of the study does not allow us to rule out the possibility that this protective effect could be due to other factors that are correlated with romantic relationship satisfaction. In addition, to truly parse “spillover” effects, one would like to examine the quality of daily interactions with nonpartners. For example, identifying negative interactions (e.g., arguments) would allow for an explicit test of whether romantic relationship satisfaction buffered increases in BP. Our inability to examine differences in the quality of various interactions may have masked significant patterns of effects. Finally, different approaches to assessing relationship satisfaction should be considered. For example, nuances in relationship satisfaction may be more accurately captured along both positive and negative dimensions rather than as a unidimensional construct [22, 23], and different dimensions may affect the relationship between social interactions and BP in different ways. In addition, some have suggested that discordant relationships are qualitatively different from more satisfying relationships, in support of a dyadic, taxonomic approach [24, 25]. Conclusion Romantic relationship satisfaction was not associated with lower BP on average. Instead, study results support the hypothesis that interacting with a significant other confers an acute cardiovascular benefit, but only in high-quality relationships. Furthermore, the beneficial effects of a satisfying romantic relationship for BP are mostly limited to interactions with that significant other. Spillover effects of relationship satisfaction to interactions with other (nonromantic) individuals could not, however, be ruled out. Randomized controlled trials are needed to determine whether improving relationship satisfaction within couples could lower BP for one or both individuals. Future research might also explore the association between frequency of social interactions with a partner in satisfied and less-satisfied romantic relationships, BP, and long-term cardiovascular health. Acknowledgments We are indebted to the study participants and research staff of the Masked Hypertension Study, without whose cooperation and dedication this study would not have been possible. This work was supported by grants P01-HL47540 (PI: J. Schwartz) from the National Heart, Lung, and Blood Institute. The research was also supported in part by National Center for Advancing Translational Sciences (formerly the National Center for Research Resources), National Institutes of Health, through Grant MO1-RR10710 (Stony Brook University), and Grant UL1-TR000040 (formerly, Grant UL1-RR024156; Columbia University). The content is solely the responsibility of the authors and does not necessarily represent the official view of the NIH. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards Authors Talea Cornelius, Jeffrey L. Birk, Donald Edmondson, and Joseph E. Schwartz declare that they have no conflict of interest. Author contributions JES conceived the primary outcome study, obtained research funding, supervised the data collection, and contributed to data analysis and revisions. TC conceived the research question for this manuscript, drafted the manuscript, and conducted data analysis. JLB and DE contributed to manuscript drafting and revisions. Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Procedures were approved by institutional review boards at Stony Brook University and Columbia University Medical Center. Informed Consent Informed consent was obtained from all individual participants included in the study. References 1. Coyne JC, Rohrbaugh MJ, Shoham V, Sonnega JS, Nicklas JM, Cranford JA. Prognostic importance of marital quality for survival of congestive heart failure. Am J Cardiol . 2001; 88: 526– 529. Google Scholar CrossRef Search ADS PubMed  2. Robles TF, Slatcher RB, Trombello JM, McGinn MM. Marital quality and health: A meta-analytic review. Psychol Bull . 2014; 140: 140– 187. Google Scholar CrossRef Search ADS PubMed  3. Smith TW, Baron CE, Grove JL. Personality, emotional adjustment, and cardiovascular risk: Marriage as a mechanism. J Pers . 2014; 82: 502– 514. Google Scholar CrossRef Search ADS PubMed  4. Gump BB, Polk DE, Kamarck TW, Shiffman SM. Partner interactions are associated with reduced blood pressure in the natural environment: Ambulatory monitoring evidence from a healthy, multiethnic adult sample. Psychosom Med . 2001; 63: 423– 433. Google Scholar CrossRef Search ADS PubMed  5. Holt-Lunstad J, Birmingham W, Jones BQ. Is there something unique about marriage? The relative impact of marital status, relationship quality, and network social support on ambulatory blood pressure and mental health. Ann Behav Med . 2008; 35: 239– 244. Google Scholar CrossRef Search ADS PubMed  6. Holt-Lunstad J, Uchino BN, Smith TW, Olson-Cerny C, Nealey-Moore JB. Social relationships and ambulatory blood pressure: Structural and qualitative predictors of cardiovascular function during everyday social interactions. Health Psychol . 2003; 22: 388– 397. Google Scholar CrossRef Search ADS PubMed  7. Baker B, Szalai JP, Paquette M, Tobe S. Marital support, spousal contact and the course of mild hypertension. J Psychosom Res . 2003; 55: 229– 233. Google Scholar CrossRef Search ADS PubMed  8. Grewen KM, Girdler SS, Light KC. Relationship quality: Effects on ambulatory blood pressure and negative affect in a biracial sample of men and women. Blood Press Monit . 2005; 10: 117– 124. Google Scholar CrossRef Search ADS PubMed  9. Joseph NT, Kamarck TW, Muldoon MF, Manuck SB. Daily marital interaction quality and carotid artery intima-medial thickness in healthy middle-aged adults. Psychosom Med . 2014; 76: 347– 354. Google Scholar CrossRef Search ADS PubMed  10. Bowen KS, Uchino BN, Birmingham W, Carlisle M, Smith TW, Light KC. The stress-buffering effects of functional social support on ambulatory blood pressure. Health Psychol . 2014; 33: 1440– 1443. Google Scholar CrossRef Search ADS PubMed  11. Howard S, Creaven AM, Hughes BM, O’Leary ÉD, James JE. Perceived social support predicts lower cardiovascular reactivity to stress in older adults. Biol Psychol . 2017; 125: 70– 75. Google Scholar CrossRef Search ADS PubMed  12. Carels RA, Sherwood A, Szczepanski R, Blumenthal JA. Ambulatory blood pressure and marital distress in employed women. Behav Med . 2000; 26: 80– 85. Google Scholar CrossRef Search ADS PubMed  13. Birmingham WC, Uchino BN, Smith TW, Light KC, Butner J. It’s complicated: Marital ambivalence on ambulatory blood pressure and daily interpersonal functioning. Ann Behav Med . 2015; 49: 743– 753. Google Scholar CrossRef Search ADS PubMed  14. Schwartz JE, Burg MM, Shimbo D, et al.   Clinic blood pressure underestimates ambulatory blood pressure in an untreated employer-based US population: Results from the masked hypertension study. Circulation . 2016; 134: 1794– 1807. Google Scholar CrossRef Search ADS PubMed  15. Schwartz JE, Warren K, Pickering TG. Mood, location and physical position as predictors of ambulatory blood pressure and heart rate: Application of a multi-level random effects model. Ann Behav Med . 1994; 16( 3): 210– 222. 16. Spanier GB. Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. J Marriage Fam . 1976; 38( 1): 15– 28. Google Scholar CrossRef Search ADS   17. Schwartz JE, Stone AA. The analysis of real-time momentary data: A practical guide. In Stone AA, Shiffman S, Atienza AA, Nebeling L, eds. The Science of Real-Time Data Capture: Self-Report in Health Research . Oxford: Oxford University Press; 2007: 76– 113. 18. Bippus AM, Rollin E. Attachment style differences in relational maintenance and conflict behaviors: Friends’ perceptions. Commun Rep . 2003; 16( 2): 113– 123. Google Scholar CrossRef Search ADS   19. Grabill CM, Kerns KA. Attachment style and intimacy in friendship. Pers Relationship . 2000; 7( 4): 363– 378. Google Scholar CrossRef Search ADS   20. Simpson JA. Influence of attachment styles on romantic relationships. J Pers Soc Psychol . 1990; 59( 5): 971– 980. Google Scholar CrossRef Search ADS   21. Smith TW, Baucom BRW. Intimate relationships, individual adjustment, and coronary heart disease: Implications of overlapping associations in psychosocial risk. Am Psychol . 2017; 72: 578– 589. Google Scholar CrossRef Search ADS PubMed  22. Mattson RE, Paldino D, Johnson MD. The increased construct validity and clinical utility of assessing relationship quality using separate positive and negative dimensions. Psychol Assess . 2007; 19: 146– 151. Google Scholar CrossRef Search ADS PubMed  23. Rogge RD, Fincham FD, Crasta D, Maniaci MR. Positive and negative evaluation of relationships: Development and validation of the Positive-Negative Relationship Quality (PN-RQ) scale. Psychol Assess . 2017; 29: 1028– 1043. Google Scholar CrossRef Search ADS PubMed  24. Beach SR, Fincham FD, Amir N, Leonard KE. The taxometrics of marriage: Is marital discord categorical? J Fam Psychol . 2005; 19: 276– 285. Google Scholar CrossRef Search ADS PubMed  25. Whisman MA, Beach SR, Snyder DK. Is marital discord taxonic and can taxonic status be assessed reliably? Results from a national, representative sample of married couples. J Consult Clin Psychol . 2008; 76: 745– 755. Google Scholar CrossRef Search ADS PubMed  © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Behavioral Medicine Oxford University Press

Romantic Relationship Satisfaction and Ambulatory Blood Pressure During Social Interactions: Specificity or Spillover Effects?

Loading next page...
 
/lp/ou_press/romantic-relationship-satisfaction-and-ambulatory-blood-pressure-E1rgtKGPvn
Copyright
© Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0883-6612
eISSN
1532-4796
D.O.I.
10.1093/abm/kay030
Publisher site
See Article on Publisher Site

Abstract

Abstract Background People in high-quality romantic relationships tend to have lower blood pressure (BP). People may experience lower BP specifically when interacting with romantic partners. Purpose This study parsed the effects of different types of social interactions on ambulatory BP (ABP) and tested whether romantic relationship satisfaction moderated these effects during interactions with partners in particular (specificity) or with others (spillover; e.g., friends, co-workers). Methods Partnered participants (N = 594) were drawn from a larger study on BP and cardiovascular health (age = 46.5 ± 9.3; 57.4% female). Participants reported on romantic relationship satisfaction and completed 24-hr ABP monitoring. At each reading, participants reported whether they had a social interaction and with whom. Multilevel models accounted for nesting of data over time. Results Romantic relationship satisfaction significantly modified the effects of some social interactions on systolic and diastolic BP (SBP, DBP). Participants with high (+1 SD) relationship satisfaction had significantly lower SBP (−0.77 mmHg, p = .02) during partner interactions compared with no social interaction; low-satisfaction (−1 SD) participants had a nonsignificant 0.59 mmHg increase (p = .14). A similar pattern emerged for DBP. Relationship satisfaction also modified SBP response during friend interactions (elevated SBP for low-satisfaction participants) and DBP response during “other” interactions (elevated DBP for high-satisfaction participants). Conclusion Participants with high levels of romantic relationship satisfaction experienced significantly lower BP during social interactions with their partner compared with situations without social interaction. Although there was some evidence for spillover to other types of relationships, effects were largely restricted to partner interactions. Relationship satisfaction, Ambulatory blood pressure, Blood pressure, Couples, Social interaction, Ecological momentary assessment Marriage is robustly associated with better cardiovascular outcomes in the long term [1–3] and predicts lower blood pressure (BP) on a daily basis [2, 4–6]. Yet all marriages are not created equal. The quality of a relationship matters, with higher quality relationships being uniquely associated with improved cardiovascular health (e.g., decreased left-ventricular mass, lower BP [7–9]). However, the mechanisms underlying the effects of marriage and marital quality on cardiovascular outcomes are not fully understood. It may be that there is a general, beneficial effect of high-quality relationships on cardiovascular outcomes, including lower average BP. On the other hand, people may experience lower BP specific to situations when interacting with a romantic partner and such effects may accumulate to predict better cardiovascular outcomes. These possibilities are suggested by research indicating that marital satisfaction and cohesion predict long-term outcomes (left-ventricular mass) and are associated with increased spousal contact [7]. Finally, it is possible that high-quality romantic relationships may have a broad impact on reactions to social situations, shaping the effects of social interactions on cardiovascular outcomes across relationship types. For instance, a satisfying romantic relationship might buffer effects of potentially stressful interactions with supervisors or co-workers, reflecting a positive “spillover” to nonpartner social interactions. This delineates three possibilities: the effect of high-quality relationships on cardiovascular health may (1) be specific to romantic partner interactions, (2) spill over to affect the nature of other social interactions, or (3) be general (i.e., function across all types of social interactions and also situations when one is not engaged in a social interaction). There is surprisingly little direct evidence about the influence of relationship satisfaction with one’s partner (i.e., spouse, significant other) on BP specific to interactions with one’s partner, and almost none concerning the influence of romantic relationship satisfaction on cardiovascular responses to interactions with specific other people (i.e., “spillover” effects). For example, although Gump et al. [4] examined BP during social interactions and found that romantic partner interactions were associated with significantly lower BP as compared with no interaction or interactions with others, “others” were not differentiated (e.g., co-worker vs. family interactions). This lack of differentiation could muddy the waters in understanding whether romantic relationship satisfaction moderates the effects of some types of interactions on BP, but not others (e.g., co-workers vs. strangers). Similarly, Joseph et al. [9] compared the positivity of a dichotomy, marital versus nonmarital interactions, and found that only positive marital interactions predicted lower carotid artery intima medial thickness. Taken together, these studies suggest that the BP-lowering effects of a high-quality romantic relationship may be specific to partner interactions as compared with other types of interactions (i.e., a “specificity” effect). Yet, the possibility of a “spillover” effect—that satisfaction with a primary romantic relationship may exert cardioprotective effects in other types of relationships (e.g., co-workers, friends)—has not been rigorously tested across relationship types. Indirect evidence from literature on social support suggests that spillover may be possible. For example, social support more broadly can serve as a buffer against the negative effects of momentary stress on BP [10], which might include potentially stressful interactions, although the buffering effect of social support on BP may be isolated to the lab [11]. Further support for romantic relationship-specific effects on cardiovascular risk markers exists at the other end of the satisfaction spectrum. Marital distress has been associated with higher BP at home, but not at work [12], indicating that marital distress may matter most in situations in which the romantic partner is most relevant (although the actual presence of a partner was not assessed). These results are suggestive, but they rely on a simple approach that assumes presence or absence of the romantic partner, and findings are mixed (e.g., the effect of marital ambivalence on BP did not differ by location and thus may not be specific to spousal interactions [13]). The possibilities of general, specific, and spillover effects of romantic relationship satisfaction on BP have not yet been tested in a single sample. Furthermore, because some of the specificity effects from prior research described above must be inferred, there is not yet direct evidence within a single, ecologically valid study of whether this effect is truly specific to partner interactions. A more fine-grained approach to understanding the influence of romantic relationship satisfaction on the cardiovascular response to both romantic partner and other social interactions is warranted. In this study, we estimated participants’ ambulatory BP (ABP) during social interactions with both romantic partners and others, compared with situations involving no social interaction, and tested whether satisfaction with the primary romantic relationship was related to differential BP levels during those interactions. We hypothesized that participants who were highly satisfied with their romantic relationship would have reduced BP in general and especially so during interactions with their partner (i.e., “specificity” effects). In addition, we explored the possibility that BP might be reduced during social interactions with nonpartners in people reporting high romantic relationship satisfaction (vs. no interaction or vs. low relationship satisfaction; i.e., “spillover” effects). Methods Participants Participants were drawn from a larger study examining the intersection of BP in the clinic setting, ABP, and cardiovascular health (Masked Hypertension Study [MHTN]; see [14] for further details). To be eligible for the primary study, participants were required to be (i) 21+ years old, (ii) working 17.5+ hr per week, (iii) able to speak and read English, (iv) not taking BP-lowering medications, and (v) have a pre-enrollment screening BP of less than 160/105 mmHg. Participants were excluded if they reported a history of cardiovascular disease (e.g., congestive heart failure, myocardial infarction), chronic renal disease, liver disease, thyroid disease, or adrenal disease, cancer that was not in remission for a minimum of 6 months, active substance abuse, or a serious mental health illness. Participants were also excluded if they had evidence of secondary hypertension, were taking any cardiovascular medication other than statins, or were pregnant. Study procedures were approved by the Columbia University Medical Center Institutional Review Board and the Stony Brook University Institutional Review Board, and participants completed informed consent before completing study procedures. For the present study, participants (N = 594) who completed both ABP monitoring and a measure of romantic relationship satisfaction were included; participants did not complete the measure of romantic relationship satisfaction if they said they were not married and not living with a partner. Individuals, and not couples, were recruited to participate in the study, although a small number of couples (<5) did participate. Procedure The MHTN [14] study comprised five visits, of which two are relevant to the current study. At the first visit, participants were given a packet of psychosocial questionnaires, including the measure of romantic relationship satisfaction, and asked to complete and return it within two weeks (i.e., by the time of their third visit). At the third visit, participants were fitted with an ABP monitor programmed to take readings every 28–30 min for 24 hr. Participants were also trained in the use of an electronic diary for ecological momentary assessments (EMAs) that asked about recent social interactions, and were asked to complete a diary entry immediately following each BP reading (i.e., fixed, interval-contingent reporting). Momentary covariates known to affect ABP readings (e.g., posture, location) [6, 15] were also assessed by the diary. ABP readings and EMAs spanned two workdays for all participants (no weekend days). Although nocturnal ABP data were available, sleep BP readings were excluded from the analysis because the focus of this analysis was on the effect of different types of social interactions on ABP, and EMAs (and, ostensibly, social interactions) were not completed during sleep. Measures Romantic relationship satisfaction Relationship satisfaction with a current romantic partner was measured using a 32-item modified version of the Dyadic Adjustment Scale (DAS [16]). Example items included rating how often the participant and his or her partner agreed on demonstrations of affection (0 = always disagree to 5 = always agree) and how often the participant thought things were going well with his or her partner (0 = never to 5 = all the time; α = .84). ABP ABP readings were taken every 28–30 min using an ABP monitor (SpaceLabs, Model 90207, Snoqualmie, WA, USA). The monitor was fitted on the nondominant arm using an appropriate-sized cuff and was worn for 24 consecutive hours. Participants returned the BP monitor and electronic diary at the end of the 24-hr monitoring period. Per the recommendations of the device manufacturer, systolic BP (SBP) readings were required to fall within the range of 70–250 mmHg and diastolic BP (DBP) readings within the range of 40–150 mmHg while awake (30–150 during sleep); the small number of readings outside of these ranges were treated as errors (i.e., missing values). Social interaction Following each ABP reading, participants completed the electronic diary EMA noting whether they had had a social interaction at any point since the last reading (yes or no) and with whom. Options for type of social interaction were spouse/significant other, family, friends, clients, public, doctor/nurse, supervisor, co-worker, and other, and more than one response could be checked at any given assessment (to retain all interaction information, indicator variables were constructed for the presence/absence of each type of interaction; thus, the analysis estimated separate effects for each type of interaction, controlling for the presence/absence of all other types). Separate indicator variables (0 = absent, 1 = present) were created for each of the nine types of interaction; reference category corresponds to no social interaction (0 on all nine indicators). Data Analysis Strategy The diary reports and BP readings entail multiple observations from a single individual, violating traditional assumptions of independence. Furthermore, observations from an individual that are closer together in time are likely to be more similar to each other than those assessed further apart (serial autocorrelation). To accommodate the nested structure of the data, multilevel models were estimated using the MIXED procedure in SAS. The specified error structure accounted for both autocorrelation (i.e., increased similarity of observations close together in time) and random error using the “local” subcommand [17], and the intercept and social interaction (yes or no) were treated as random effects to account for unexplained individual differences in mean level of BP and in the effects of social interaction on BP. Two models were estimated for each outcome (SBP, DBP). In the first model, the main effects of relationship satisfaction (person-level factor) and the dichotomous indicator variables identifying types of social interaction (momentary factors, reference situation corresponds to “no interaction” when all indicator variables equal zero) were included. In the second model, the multiplicative interaction terms of romantic relationship satisfaction with each of the interaction type indicators were entered to test the moderating effect of relationship satisfaction with one’s spouse/partner on the relationship of each type of social interaction with BP. We additionally estimated a planned contrast comparing the multiplicative interaction term of romantic relationship satisfaction with interacting with a partner to the multiplicative interaction term of romantic relationship satisfaction with all other types of social interactions; as such, a significant, negative contrast would provide evidence for a specificity effect (i.e., romantic relationship satisfaction affects BP response primarily [or more] during partner interactions), whereas a nonsignificant contrast would provide evidence for spillover (i.e., romantic relationship satisfaction affects BP response similarly, on average, during both partner and nonpartner interactions). All models controlled for age, race, ethnicity, and BMI (person level factors), and posture, exertion, temperature, location, and recent consumption of a meal, caffeine, alcohol, or nicotine (momentary factors) which have been shown to affect BP [6, 15]. Results The 594 participants included in the present study represented 66.5% of the 893 who completed the ABP procedures. The mean number of valid awake ABP readings, either at the initial attempt or a retry 2 min later was 32.3 (SD = 5.0; median = 33, quartiles: 30, 35). This represented 92.5% of all attempted awake readings. The mean number of valid awake readings which had a “concurrent” EMA report (i.e., completed within 8 min of the ABP reading) was 24.5 observations for each participant (SD = 7.3; median = 26, quartiles: 20, 30), corresponding to a mean rate of 75.5% (SD = 18.4%; median = 80.0%, quartiles: 66.7, 88.6%). Mean age was 46.5 (SD = 9.3), and mean BMI was 27.7 (SD = 5.2). Participants were predominantly female (57.4%), 87.7% were married, 6.1% were Black, and 10.3% were Hispanic. The romantic relationship satisfaction scale had a mean of 108.2 (SD = 20.8; range = 9, 147); to keep this on a similar scale to other variables and improve interpretation this was divided by 10. Of 14,545 awake BP readings with a concurrent EMA report, 8,854 involved a reported interaction (60.9%). Frequencies of each interaction type are reported in Table 1. Table 1 Frequency of different interaction types across the N = 8,854 reported interactions Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  5,691 observations included no interactions. View Large Table 1 Frequency of different interaction types across the N = 8,854 reported interactions Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  Type  n (%)  Spouse/significant other  1,960 (22.1)  Family  1,875 (21.2)  Friends  796 (9.0)  Clients  440 (5.0)  Public  378 (4.3)  Doctor/nurse  270 (3.0)  Supervisor  616 (7.0)  Co-worker  4,341 (49.0)  Other  445 (5.0)  5,691 observations included no interactions. View Large SBP Covariates predicted SBP in the expected directions (e.g., older participants and males had higher SBP, alcohol and nicotine use predicted higher SBP, greater physical exertion predicted higher SBP). In the first model predicting SBP, counter to our hypothesis about the general BP effect of romantic relationship satisfaction, there was a positive main effect of romantic relationship satisfaction on SBP; as relationship satisfaction increased, (mean) SBP also increased, B = 0.41 [0.05, 0.78] mmHg/10-point increase in relationship satisfaction, p = .03. The main effect of social interaction with a partner (compared with no social interaction) on SBP, B = −0.20 [−0.67, 0.28] mmHg, p = .42, was not significant; however, when participants reported interacting with a friend, co-worker, or “other,” they had higher SBP, Bs = 1.40, 0.46, and 1.24 mmHg, respectively. In contrast, interacting with a doctor or nurse was associated with lower SBP, B = −1.21 [−2.27, −0.15] mmHg, p = .03. In the second model, examining potential effect modification, relationship satisfaction significantly modified the effect of some types of social interactions on SBP (see Table 2). Specifically, romantic relationship satisfaction moderated the effect of interacting with a partner, B = −0.33 [−0.59, −0.06], p = .01, revealing the hypothesized BP-lowering effect of a positive romantic partnership during interactions with one’s partner. Participants with high (+1 SD) relationship satisfaction had significantly decreased SBP during interactions with a partner, B = −0.77 [−1.44, −0.10] mmHg, p = .02, compared with when there was no social interaction. In contrast, participants with low satisfaction (−1 SD) had nonsignificantly higher SBP, B = .59 [−.20, 1.37] mmHg, p = .14 (see Fig. 1). Fig. 1. View largeDownload slide Systolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Fig. 1. View largeDownload slide Systolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Participants’ relationship satisfaction with their partner also modified the SBP response to interacting with a friend, B = −0.49 [−0.80, −0.19], p < .01. The tendency for SBP to be higher during interactions with friends was amplified in those with low romantic relationship satisfaction (−1 SD; B = 2.40 [1.51, 3.28] mmHg, p < .001), and attenuated in those with high relationship satisfaction (+1 SD; B = 0.34 [−0.57, 1.26] mmHg, p = .46). The relationships of other types of social interactions with SBP (i.e., family, clients, public, doctor/nurse, supervisor, co-worker, and other) were not significantly modified by romantic relationship satisfaction. Further analyses (details not shown) provided no evidence that the moderating effect of romantic relationship satisfaction on the association between different types of social interactions and SBP differed by age or gender. The planned contrast comparing the moderating effect of romantic relationship satisfaction with partner interactions versus all other types of interactions predicting SBP was negative and significant, B = −0.39 [−0.68, −0.11], p < .01, providing additional evidence for specificity. DBP Covariates predicted DBP in the expected directions (e.g., males and older participants had higher DBP, alcohol and nicotine use predicted higher DBP, and greater physical exertion predicted higher DBP). In the first model predicting DBP, there was no effect of romantic relationship satisfaction, B = 0.09 [−0.18, 0.37] mmHg, p = .50, or interacting with a partner, B = 0.18 [−0.21, 0.58] mmHg, p = .36, on DBP. Interactions with a friend, client, supervisor, co-worker, or “other” were associated with higher DBP, Bs = 1.10, 0.78, 0.71, 0.55, and 0.74 mmHg, respectively. The second model tests whether romantic relationship satisfaction moderates the effect of different types of social interactions on DBP (see Table 2). In parallel to the negative coefficient for the interaction term for romantic relationship satisfaction and interacting with a romantic partner when predicting SBP, this interaction term’s coefficient was also negative and statistically significant when predicting DBP, B = −0.34 [−0.56, −0.13], p < .01. Simple slopes analysis showed that interacting with a significant other predicted an increase in DBP when romantic relationship satisfaction was low (−1 SD), B = 1.01 [0.37, 1.65] mmHg, p < .01, but had a nonsignificant, negative effect when relationship satisfaction was high (+1 SD), B = −0.19 [−0.97, 0.13] mmHg, p = .14 (see Fig. 2). Table 2 Coefficients and 95% CIs for the association of romantic relationship satisfaction (grand mean centered) with different types of social interactions predicting systolic blood pressure (SBP) and diastolic blood pressure (DBP)   SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*    SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*  +p < .10, *p < .05, **p < .01. View Large Table 2 Coefficients and 95% CIs for the association of romantic relationship satisfaction (grand mean centered) with different types of social interactions predicting systolic blood pressure (SBP) and diastolic blood pressure (DBP)   SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*    SBP (mmHg)  DBP (mmHg)  B [95% CI]  B [95% CI]  Relationship satisfaction (RS)  0.41 [0.04, 0.78]*  0.11 [−0.17, 0.39]  Spouse/significant other  −0.09 [−0.58. 0.39]  0.30 [−0.10, 0.69]  Family  −0.41 [−0.91, 0.11]  0.01 [−0.41, 0.43]  Friends  1.37 [0.72, 2.02]**  1.09 [0.55, 1.62]**  Clients  −0.35 [−1.26, 0.55]  0.92 [0.07, 1.57]*  Public  0.22 [−0.67, 1.10]  0.11 [−0.63, 0.85]  Doctor/nurse  −1.05 [−2.11, 0.02]+  −0.10 [−0.99, 0.78]  Supervisor  0.43 [−0.01, 0.88]  0.71 [0.13, 1.30]*  Co-worker  0.46 [0.07, 0.85]*  0.55 [0.23, 0.87]**  Other  1.20 [0.37, 2.03]**  0.67 [−0.02, 1.36]+  Spouse/significant other × RS  −0.33 [−0.59, −0.06]*  −0.34 [−0.56, −0.13]**  Family × RS  −0.01 [−0.25, 0.22]  −0.09 [−0.29, 0.10]  Friends × RS  −0.49 [−0.80, −0.19]**  −0.16 [−0.41, 0.08]  Clients × RS  −0.09 [−0.50, 0.32]  0.08 [−0.26, 0.41]  Public × RS  0.37 [−0.06, 0.81]+  −0.06 [−0.42, 0.30]  Doctor/nurse × RS  0.43 [−0.01, 0.88]+  0.04 [−0.34, 0.41]  Supervisor × RS  0.01 [−0.33, 0.34]  0.00 [−0.28, 0.28]  Co-worker × RS  0.14 [−0.03, 0.31]  0.03 [−0.11, 0.17]  Other × RS  0.17 [−0.20, 0.54]  0.39 [0.08, 0.69]*  +p < .10, *p < .05, **p < .01. View Large Fig. 2. View largeDownload slide Diastolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Fig. 2. View largeDownload slide Diastolic blood pressure change (mmHg) during social interactions with a significant other for participants high (+1 SD), average, and low (−1 SD) in satisfaction. Romantic relationship satisfaction also modified the effect of interacting with someone in the “other” category, B = 0.39 [0.08, 0.69], p = .01. Unexpectedly, interacting with an “other” was associated with higher DBP when participants reported high (+1 SD) relationship satisfaction, B = 1.48 [0.57, 2.39] mmHg, p < .01, whereas the association was nonsignificant when participants reported low (−1 SD) satisfaction, B = −0.14 [−1.10, 0.83] mmHg, p = .78. The association between other types of social interactions and DBP (i.e., family, friends, clients, public, doctor/nurse, supervisor, and co-worker) were not significantly modified by romantic relationship satisfaction. There was no evidence that the moderating effect of romantic relationship satisfaction on the association between different types of social interactions and DBP differed by age or gender. The planned contrast comparing the moderating effect of romantic relationship satisfaction with partner interactions versus all other types of interactions predicting DBP was negative and significant, B = −0.37 [−0.60, −0.14], p < .01, providing additional evidence for specificity. Discussion This is one of the largest studies to examine the link between different types of social interactions—at different levels of romantic relationship satisfaction—and BP. We examined BP response to social interactions with both a romantic partner and other individuals over the course of a normal work day in a large sample of healthy adults. Furthermore, we examined whether participants’ relationship satisfaction with their primary romantic relationship moderated the association between different types of social interactions and BP response, and whether there was a general effect of romantic relationship satisfaction on BP. In support of study hypotheses, participants with high levels of romantic relationship satisfaction experienced significantly lower BP during social interactions with their partner, compared with those with low levels of relationship satisfaction. Counter to study hypotheses, there was no evidence for a general effect of high romantic relationship satisfaction lowering BP. In fact, for SBP but not DBP, higher relationship satisfaction was associated with increased BP as a main effect. The most consistent finding was the moderation effect for romantic relationship satisfaction and partner interactions (i.e., “specificity”). Participants with high levels of relationship satisfaction experienced lower SBP during partner interactions, relative to no interactions (those with low levels of relationship satisfaction did not). Participants with low levels of relationship satisfaction experienced increased DBP during partner interactions, relative to no interactions, whereas those with high levels of relationship satisfaction did not; further, tests suggested that this effect was most pronounced during partner interactions (vs. all other types of interactions). In sum, for SBP and DBP, romantic relationship satisfaction was associated with either a decrease or lack of increase in BP during partner interactions. Although there was some support for the specificity hypothesis (i.e., that romantic relationship satisfaction would primarily modify BP response during interactions with one’s partner), spillover effects could not be ruled out. For example, interacting with a friend was associated with higher SBP when relationship satisfaction was low, but was not associated with SBP when relationship satisfaction was high. The moderation effect for friend interactions mirrors the direction for the moderation effect for interactions with a significant other; in both cases, SBP decreases more (or increases less) during social interactions for those with high romantic relationship satisfaction. The reason for these parallel findings is not clear. There may be a general tendency for people who are satisfied in one type of close relationship (i.e., a romantic relationship) to also be satisfied in other types of close relationships (i.e., friendships) and to garner similar BP benefits from each. Indeed, people who have more secure attachment styles may have closer and more satisfying relationships generally [18–20] such that there are “satisfaction” effects across different types of social interactions. Similarly, other personality differences that can affect relationship or interaction quality across multiple types of relationships (e.g., depressive symptoms, hostility) may also play a role [21]. We did not assess attachment style or the quality of participants’ relationships with friends or family, however, so we cannot know from these data whether their satisfaction with nonromantic others played a role. A more speculative possibility is that people in lower quality relationships who report low levels of romantic relationship satisfaction may vent to their friends, and so friend interactions may be more fraught. This is also suggested by attachment research indicating that anxiously attached individuals may self-disclose more to their friends [19]. It would be interesting to follow up with a study including conversation content to specifically code for relationship-related talk and to investigate subsequent effects on physiological responses. Future research might also measure attachment styles to incorporate a more nuanced analysis of spillover effects. A second, unexpected “spillover” effect concerned the fact that people in more satisfying romantic relationships exhibited higher DBP when interacting with people in the “other” category. It is unclear why this might happen, and it is difficult to speculate given the lack of clarity as to who “others” were (particularly given the wide range of possible categories of interaction partners). Still, it is important to note that the nature of this relationship was in the opposite direction of what was predicted (i.e., we found higher BP with greater relationship satisfaction, rather than lower BP). Allowing open-ended responses for the “other” category in future research might shed light on this phenomenon. The effect may also have emerged due to chance, however, suggesting caution in pursuing this without a firm, theoretical basis. Placing this study in the context of the broader literature on close relationships and cardiovascular risk, another key finding was that there was no evidence that a high-quality romantic relationship related to lower BP in general. Rather, greater romantic relationship satisfaction was associated with higher average SBP. This is unexpected, given the previously published association between higher quality relationships and better cardiovascular outcomes [2, 7–9], and suggests the importance of examining effects of relationship satisfaction across varying timescales, including a consideration of both short- and long-term mechanisms driving these effects in potentially opposite directions (e.g., immediate effects, trait-level effects, accumulated experiences). Research finding main effects of relationship quality on lowered cardiovascular risk long-term may be due either to stronger effects of close relationships on other mechanistic pathways, or to accumulated effects of satisfying partner interactions on BP [7, 9]. One hypothesis consistent with the latter explanation is that the beneficial effects of daily interactions on BP may accumulate and be amplified if romantic relationship satisfaction leads to increased contact with the significant other [3, 4]. It would be informative to test a series of models, in which (1) relationship satisfaction is included as a moderator of BP response to interactions with a significant other and (2) frequency of contact over an extended period of time is examined as a mediator of a satisfying romantic relationship’s positive effect on long-term cardiovascular outcomes. Such evidence would also provide critical mechanistic insight for the creation of couples-focused prevention programs to improve cardiovascular health. Although this study has many strengths, including a large sample size and strong ecological validity, there are a number of limitations. The sample in this study consisted of people who were working 17.5 hr a week or more, which limits the generalizability of the results. The sample was also restricted to nonhypertensive individuals. An important next step would be to replicate these effects in a hypertensive sample. There were also a small number of couples (4, 1.3% of the sample). Because participants were enrolled and studied as individuals and the number of couples is so small, we decided to ignore this small degree of interdependence. Another limitation of the study is that participants were asked to complete only 24 hr of ABP monitoring, which limited the variety of social interactions each participant was likely to experience. Future studies should consider the specific effects of different types of social interactions across a longer time period. Other potential explanatory variables (e.g., attachment, individual personality differences that may affect social interactions, such as hostility, satisfaction with different types of social relationships) should also be explored. That said, the primary goal of this analysis was to test the specific hypothesis that the effect on BP of interacting with one’s partner would be associated, in a protective manner, with romantic relationship satisfaction—that is, a factor that is specific to one’s relationship with that partner. The present analyses provide support for this hypothesis, although the observational nature of the study does not allow us to rule out the possibility that this protective effect could be due to other factors that are correlated with romantic relationship satisfaction. In addition, to truly parse “spillover” effects, one would like to examine the quality of daily interactions with nonpartners. For example, identifying negative interactions (e.g., arguments) would allow for an explicit test of whether romantic relationship satisfaction buffered increases in BP. Our inability to examine differences in the quality of various interactions may have masked significant patterns of effects. Finally, different approaches to assessing relationship satisfaction should be considered. For example, nuances in relationship satisfaction may be more accurately captured along both positive and negative dimensions rather than as a unidimensional construct [22, 23], and different dimensions may affect the relationship between social interactions and BP in different ways. In addition, some have suggested that discordant relationships are qualitatively different from more satisfying relationships, in support of a dyadic, taxonomic approach [24, 25]. Conclusion Romantic relationship satisfaction was not associated with lower BP on average. Instead, study results support the hypothesis that interacting with a significant other confers an acute cardiovascular benefit, but only in high-quality relationships. Furthermore, the beneficial effects of a satisfying romantic relationship for BP are mostly limited to interactions with that significant other. Spillover effects of relationship satisfaction to interactions with other (nonromantic) individuals could not, however, be ruled out. Randomized controlled trials are needed to determine whether improving relationship satisfaction within couples could lower BP for one or both individuals. Future research might also explore the association between frequency of social interactions with a partner in satisfied and less-satisfied romantic relationships, BP, and long-term cardiovascular health. Acknowledgments We are indebted to the study participants and research staff of the Masked Hypertension Study, without whose cooperation and dedication this study would not have been possible. This work was supported by grants P01-HL47540 (PI: J. Schwartz) from the National Heart, Lung, and Blood Institute. The research was also supported in part by National Center for Advancing Translational Sciences (formerly the National Center for Research Resources), National Institutes of Health, through Grant MO1-RR10710 (Stony Brook University), and Grant UL1-TR000040 (formerly, Grant UL1-RR024156; Columbia University). The content is solely the responsibility of the authors and does not necessarily represent the official view of the NIH. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards Authors Talea Cornelius, Jeffrey L. Birk, Donald Edmondson, and Joseph E. Schwartz declare that they have no conflict of interest. Author contributions JES conceived the primary outcome study, obtained research funding, supervised the data collection, and contributed to data analysis and revisions. TC conceived the research question for this manuscript, drafted the manuscript, and conducted data analysis. JLB and DE contributed to manuscript drafting and revisions. Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Procedures were approved by institutional review boards at Stony Brook University and Columbia University Medical Center. Informed Consent Informed consent was obtained from all individual participants included in the study. References 1. Coyne JC, Rohrbaugh MJ, Shoham V, Sonnega JS, Nicklas JM, Cranford JA. Prognostic importance of marital quality for survival of congestive heart failure. Am J Cardiol . 2001; 88: 526– 529. Google Scholar CrossRef Search ADS PubMed  2. Robles TF, Slatcher RB, Trombello JM, McGinn MM. Marital quality and health: A meta-analytic review. Psychol Bull . 2014; 140: 140– 187. Google Scholar CrossRef Search ADS PubMed  3. Smith TW, Baron CE, Grove JL. Personality, emotional adjustment, and cardiovascular risk: Marriage as a mechanism. J Pers . 2014; 82: 502– 514. Google Scholar CrossRef Search ADS PubMed  4. Gump BB, Polk DE, Kamarck TW, Shiffman SM. Partner interactions are associated with reduced blood pressure in the natural environment: Ambulatory monitoring evidence from a healthy, multiethnic adult sample. Psychosom Med . 2001; 63: 423– 433. Google Scholar CrossRef Search ADS PubMed  5. Holt-Lunstad J, Birmingham W, Jones BQ. Is there something unique about marriage? The relative impact of marital status, relationship quality, and network social support on ambulatory blood pressure and mental health. Ann Behav Med . 2008; 35: 239– 244. Google Scholar CrossRef Search ADS PubMed  6. Holt-Lunstad J, Uchino BN, Smith TW, Olson-Cerny C, Nealey-Moore JB. Social relationships and ambulatory blood pressure: Structural and qualitative predictors of cardiovascular function during everyday social interactions. Health Psychol . 2003; 22: 388– 397. Google Scholar CrossRef Search ADS PubMed  7. Baker B, Szalai JP, Paquette M, Tobe S. Marital support, spousal contact and the course of mild hypertension. J Psychosom Res . 2003; 55: 229– 233. Google Scholar CrossRef Search ADS PubMed  8. Grewen KM, Girdler SS, Light KC. Relationship quality: Effects on ambulatory blood pressure and negative affect in a biracial sample of men and women. Blood Press Monit . 2005; 10: 117– 124. Google Scholar CrossRef Search ADS PubMed  9. Joseph NT, Kamarck TW, Muldoon MF, Manuck SB. Daily marital interaction quality and carotid artery intima-medial thickness in healthy middle-aged adults. Psychosom Med . 2014; 76: 347– 354. Google Scholar CrossRef Search ADS PubMed  10. Bowen KS, Uchino BN, Birmingham W, Carlisle M, Smith TW, Light KC. The stress-buffering effects of functional social support on ambulatory blood pressure. Health Psychol . 2014; 33: 1440– 1443. Google Scholar CrossRef Search ADS PubMed  11. Howard S, Creaven AM, Hughes BM, O’Leary ÉD, James JE. Perceived social support predicts lower cardiovascular reactivity to stress in older adults. Biol Psychol . 2017; 125: 70– 75. Google Scholar CrossRef Search ADS PubMed  12. Carels RA, Sherwood A, Szczepanski R, Blumenthal JA. Ambulatory blood pressure and marital distress in employed women. Behav Med . 2000; 26: 80– 85. Google Scholar CrossRef Search ADS PubMed  13. Birmingham WC, Uchino BN, Smith TW, Light KC, Butner J. It’s complicated: Marital ambivalence on ambulatory blood pressure and daily interpersonal functioning. Ann Behav Med . 2015; 49: 743– 753. Google Scholar CrossRef Search ADS PubMed  14. Schwartz JE, Burg MM, Shimbo D, et al.   Clinic blood pressure underestimates ambulatory blood pressure in an untreated employer-based US population: Results from the masked hypertension study. Circulation . 2016; 134: 1794– 1807. Google Scholar CrossRef Search ADS PubMed  15. Schwartz JE, Warren K, Pickering TG. Mood, location and physical position as predictors of ambulatory blood pressure and heart rate: Application of a multi-level random effects model. Ann Behav Med . 1994; 16( 3): 210– 222. 16. Spanier GB. Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. J Marriage Fam . 1976; 38( 1): 15– 28. Google Scholar CrossRef Search ADS   17. Schwartz JE, Stone AA. The analysis of real-time momentary data: A practical guide. In Stone AA, Shiffman S, Atienza AA, Nebeling L, eds. The Science of Real-Time Data Capture: Self-Report in Health Research . Oxford: Oxford University Press; 2007: 76– 113. 18. Bippus AM, Rollin E. Attachment style differences in relational maintenance and conflict behaviors: Friends’ perceptions. Commun Rep . 2003; 16( 2): 113– 123. Google Scholar CrossRef Search ADS   19. Grabill CM, Kerns KA. Attachment style and intimacy in friendship. Pers Relationship . 2000; 7( 4): 363– 378. Google Scholar CrossRef Search ADS   20. Simpson JA. Influence of attachment styles on romantic relationships. J Pers Soc Psychol . 1990; 59( 5): 971– 980. Google Scholar CrossRef Search ADS   21. Smith TW, Baucom BRW. Intimate relationships, individual adjustment, and coronary heart disease: Implications of overlapping associations in psychosocial risk. Am Psychol . 2017; 72: 578– 589. Google Scholar CrossRef Search ADS PubMed  22. Mattson RE, Paldino D, Johnson MD. The increased construct validity and clinical utility of assessing relationship quality using separate positive and negative dimensions. Psychol Assess . 2007; 19: 146– 151. Google Scholar CrossRef Search ADS PubMed  23. Rogge RD, Fincham FD, Crasta D, Maniaci MR. Positive and negative evaluation of relationships: Development and validation of the Positive-Negative Relationship Quality (PN-RQ) scale. Psychol Assess . 2017; 29: 1028– 1043. Google Scholar CrossRef Search ADS PubMed  24. Beach SR, Fincham FD, Amir N, Leonard KE. The taxometrics of marriage: Is marital discord categorical? J Fam Psychol . 2005; 19: 276– 285. Google Scholar CrossRef Search ADS PubMed  25. Whisman MA, Beach SR, Snyder DK. Is marital discord taxonic and can taxonic status be assessed reliably? Results from a national, representative sample of married couples. J Consult Clin Psychol . 2008; 76: 745– 755. Google Scholar CrossRef Search ADS PubMed  © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Annals of Behavioral MedicineOxford University Press

Published: May 8, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off