Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Health Commun. 2021 Feb 9;37(8):1013–1021. doi: 10.1080/10410236.2021.1880685

Evidence of the Associations Between Individual and Partner Autonomy Support and Physiological Stress in the Context of Conversations about Weight Among Couples Who are Overweight or Obese during a 6-Month Intervention

Amanda Denes 1, John P Crowley 2, Ambyre LP Ponivas 3, Talea Cornelius 4, Ryan J Allred 5, Katelyn M Gettens 6,7, Theodore A Powers 8, Amy A Gorin 9
PMCID: PMC8349938  NIHMSID: NIHMS1667284  PMID: 33563035

Abstract

The aim of the study was to examine the impact of supportive communication on acute physiological stress responses during weight-related conversations taking place throughout a couples’ weight loss program. Participants were 47 married or cohabitating couples where each partner had a BMI of 25–40kg/m2. Couples were randomized as a dyad into a traditional weight loss program or a program that also included training in providing support to one’s partner throughout the weight loss process. Structured conversations between partners about weight management were videotaped at baseline and 6 months. Participants provided saliva samples before and after the conversations, which were assayed for cortisol and salivary alpha-amylase (sAA) to determine physiological stress and anxiety responses to conversations about weight. The results indicated that receiving support from one’s partner when discussing weight-related issues was associated with greater physiological stress, as indicated by higher cortisol and sAA levels, whereas providing support to one’s partner was associated with lower cortisol levels and higher sAA levels. The findings suggest that receiving support is not a universally positive experience, especially for populations facing health issues. The mixed findings for support provision align with previous studies identifying a negative association between affectionate communication and cortisol levels, as well as a positive association between sAA and anxiety and emotional arousal. The findings and their implications for understanding the physiological correlates of couples’ conversations about weight are discussed.

The interpersonal nature of health behaviors is becoming increasingly recognized, even as treatment models for weight loss remain largely individually focused. Behavioral treatment of weight loss (BWL) is the preferred intervention for individuals with a body mass index (BMI) of ≥30kg/m2 (≥25kg/m2 when other weight-related comorbidities exist), but the long-term results of the treatment have proved unsatisfactory (Butryn et al., 2011; NHLBI, 1998). One reason for the lack of long-term results may be due to the focus on weight as an individual issue. Weight management is often viewed as a “personal disorder,” but research can benefit from taking an ecological approach and recognizing that weight loss takes place in a broader social/emotional environment (Crowley et al., 2020; Egger & Swinburn, 1997; Sallis et al., 1998).

Self-determination theory (SDT; Ryan & Deci, 2000a, 2000b) provides a valuable framework for understanding the role of support and the interpersonal milieu in the weight loss process. The theory emphasizes the role of autonomy as one of three psychological needs pivotal to health and well-being (Ryan & Deci, 2000a). Autonomous self-regulation (ASR) involves a sense that an individual’s decisions are freely made and a matter of personal choice, thereby promoting internalization, which entails a person assigning value to a behavior and accepting responsibility for its implementation (Deci et al., 1994). Autonomy supportive behaviors should foster an environment where individuals can make personally meaningful choices based on intrinsic motivations and/or external motivations that are internalized in adaptive ways (Ryan & Deci, 2000b).

In a weight loss context, autonomy support includes communication that values a partner’s unique perspective, offers an explanation for change, presents choices, supports a partner’s initiative, and withholds criticism, control, or pressure (Silva et al., 2010). Previous research has shown that autonomous motivation has positive direct and indirect effects on weight loss and maintenance (Silva et al., 2011; Williams et al., 1996). Prior research has also demonstrated the importance of receiving support from partners for dietary change (Gorin et al., 2013) and weight loss outcomes (Crowley et al., 2020; Gorin et al., 2014; Powers et al., 2008; Silva et al., 2011). Taken together, supportive environments help promote ASR, which in turn contributes to achieving weight loss goals.

Couples’ Communication About Weight and Physiological Stress

Involving spouses or partners in weight management garners positive effects for individuals with respect to weight loss (see Black et al., 1990 and McLean et al., 2003 for reviews), yet most interventions focus only on one individual’s weight loss and do not expand to address the social and relational aspects of weight management (Leroux et al., 2013) or the potential stress surrounding couples’ weight loss. For some, discussing weight or “weight talk” (Bove & Sobal, 2011) can be a stressful experience that may trigger the physiological stress response system and impede weight loss efforts. The psychobiology of stress has focused on two systems: the hypothalamic-pituitary adrenal (HPA) axis and the locus coereleus/autonomic sympathetic nervous system (SNS) (Gordis et al., 2016). The stress hormone cortisol is released by the HPA axis in response to an acute stressor and has been linked to obesogenic behaviors, such as increased calorie consumption and craving unhealthy foods (Epel et al., 2001). Chronic stress and increased cortisol concentrations are thought to both contribute to weight gain and be indicative of obesity (see Jackson & Steptoe, 2018 and Tomiyama, 2014, for reviews), and are positively associated with weight-related discrimination, stigma, and dieting (e.g., Epel et al., 2001; Jackson & Steptoe, 2018; Tomiyama, 2014; Tomiyama et al., 2010). In contrast, supportive and affectionate communication has stress-ameliorating effects (Cohen & Wills, 1985; Floyd & Riforgiate, 2008).

The enzyme salivary alpha-amylase (sAA) is a reliable marker of activity in the SNS (Granger et al., 2006; Nater & Rohleder, 2009). sAA is associated with both stress and anxiety (Nater & Rohleder, 2009) and has been framed by communication scholars as indicative of a fight-or-flight reaction in response to novel or challenging situations (Afifi et al., 2011). sAA also increases when experiencing both positive and negative emotional states (Adam et al., 2011). Despite the responsiveness of sAA to a range of moods, prior communication scholarship suggests that positive communication environments buffer sAA reactivity. Indeed, dampened sAA responses and a faster recovery rate have been associated with greater perceived communication competency, social support, and appropriate communication from an interaction partner (Afifi et al., 2015a; Afifi et al., 2015b). Therefore, couples who are more supportive when discussing weight issues with their partner should become less physiologically stressed in response to such conversations.

Affection exchange theory (AET; Floyd, 2006) also speaks to the benefits of warm and supportive environments in garnering health benefits for individuals and relationships, particularly in regard to stress (see Denes et al., 2017 and Floyd et al., 2020 for reviews). In particular, prior work has established a negative association between affectionate communication (i.e., expressing feelings of fondness and positive regard via verbal, nonverbal, and supportive messages; Floyd & Morman, 1998) and stress (e.g., Floyd, 2002; Floyd et al., 2007). Taken together, such work suggests that romantic partners’ communication of positive regard and support can benefit individuals’ physiological stress response.

Exploring physiological stress responses to weight talk will provide a better understanding of the bodily responses that may underlie supportive communication and offer an explanatory mechanism for future interventions that investigate the influence of supportive communication on long-term health and weight loss outcomes. As such, the aim of the present study was to examine the impact of supportive communication on acute physiological stress responses during weight-related conversations taking place throughout a weight loss program for couples where both partners had BMIs between 25–40kg/m2. Couples were randomized into either a 6-month standard weight loss condition or to an enhanced needs supportive condition using an SDT framework (see Gorin et al., 2020 and Gorin et al., 2017 for full overview and protocol). Couples in the standard weight loss condition participated in a 6-month program that involved 1-hour weekly group meetings led by interventionists with advanced degrees in nutrition, exercise physiology, or behavioral psychology and experience delivering behavioral weight loss treatment. Partners were encouraged to increase physical activity, placed on a standard caloric and fat restricted diet, and taught behavioral and cognitive skills for weight loss and maintenance (see Gorin et al., 2017, 2020, for full details). Couples in the enhanced needs supportive condition received the same program, but also received handouts and engaged in group discussions, role-play, and ongoing monitoring to promote autonomy supportive communication, in line with the SDT framework. More specifically, they were encouraged to: “(a) elicit and acknowledge the other’s perspectives; (b) minimize efforts to control; (c) use nonjudgmental, noncritical language; (d) support each other’s initiatives for change; and (e) develop empathic responding” as a substitute for evaluative praise or condemnation (Gorin et al., 2020, p. 139).

The larger study revealed that participants generally lost between 7.1–10.4% of their weight, but that training in autonomy support provision did not influence weight loss or ASR (full results can be found in Gorin et al., 2020). Given these findings, the present analysis focuses on whether observations of supportive behaviors during couples’ conversations about weight, rather than the conditions to which they were assigned as part of the intervention, predict physiological stress responses to weight-related discussions. Potential confounding effects of the intervention were also considered prior to conducting the primary study analyses (see preliminary analyses below).

Given the research on ASR detailed above, it was hypothesized that receiving more support from a partner would buffer physiological stress responses when examining cortisol (H1) and sAA (H2). However, providing support can be both depleting and rewarding (Brown et al., 2003; Gosnell & Gable, 2017), and sAA has been positively associated with both anxiety and positive mood states (such as feeling excited or strong; Adam et al., 2011; Nater & Rohleder, 2009). Thus, providing support may lower post-conversation stress responses, or alternatively, may make one feel empowered or excited, thereby increasing arousal. Therefore, we queried whether providing more support to a partner would be associated with more or less physiological stress when examining cortisol (RQ1) and salivary alpha-amylase (RQ2). These patterns of association were tested across two weight-related conversations: one that occurred at baseline prior to the intervention, and one that occurred at 6 months (i.e., the conclusion of the intervention).

Materials and Method

Participants

All study procedures were approved by the Institutional Review Board at the University of Connecticut (registered under clinicaltrials.gov NCT02570009). Participants were 47 married or cohabitating couples where each partner had a BMI of 25–40 kg/m2 at baseline (at first observation for the present analysis, Mean = 33.87 kg/m2, SD = 5.36), and who provided saliva samples before and after a weight-related conversation as part of a larger study about couples’ weight loss. Participants between 18–70 years of age (Mean = 55.45 years, SD = 8.14) were recruited from the community. The majority were white (93.48%), in a heterosexual relationship (97.87%), and had a combined income of $75,000+ (83.15%). There was an approximately even split between male (48.91%) and female (51.09%) participants. Couples were excluded if either partner reported conditions that would interfere with their ability to safely or accurately participate in the weight loss intervention (e.g., chronic illnesses, heart conditions; see Gorin et al., 2017, for full list of eligibility requirements and recruitment procedures). If eligible, couples were invited to participate and given an honorarium of $25 at 6 months and $40 at 12 months for completing assessments.

Procedure

All couples received 6 months of weekly weight loss group meetings and the same core information about diet and physical activity. Those who were randomized (via a simple, variable-block length randomization) into an enhanced support condition also received training in how to provide autonomy support for weight loss. Full details regarding the treatment components for each condition can be found in Gorin et al. (2017, 2020).

Couples were assessed at baseline, 3, 6, and 12 months by a research assistant blinded to group assignment. Structured conversations between partners about weight management were videotaped at baseline (after orientation but prior to the start of the intervention), 6 months, and 12 months. Participants completed a questionnaire prior to the conversation assessing their current health and medication use. After completing the questionnaire, the first saliva sample was collected using the oral swab method. Then participants were escorted into a room with two chairs facing a camera for a 10-minute conversation. The first 5 minutes involved a task focused on acclimating partners to the presence of the camera. After 5 minutes, a research assistant knocked on the door and informed the participants to engage in the second part of the conversation task, where they were asked to discuss weight loss issues occurring in the context of their relationship. At the conclusion of the conversation, participants were brought into separate rooms to complete post-conversation surveys and provide additional saliva samples (collected at 5, 20, and 40 minutes post-conversation). Saliva samples were provided at 0 and 6 months to investigate physiological responses to conversations about weight management. Thus, the present analysis only focuses on these two time points.

Measures

Participants completed a range of measures related to the larger project. The measures relevant to the present investigation are detailed below.

Demographic Information

Participants responded to several items assessing key demographic variables, such as sex, age, marital status, race/ethnicity, and income.

Salivary Cortisol and Alpha Amylase

Saliva samples were collected using SalivaBio’s Oral Swab Device (Salimetrics, State College, PA). Participants were instructed to place the swab under their tongue for 3 minutes, until it was fully saturated. Swabs were placed into a storage tube and frozen at −20 prior to being shipped on dry ice to Salimetrics’ SalivaLab (Carlsbad, CA). The pre-conversation and 20 and 40 minute post-conversation samples were assayed for cortisol, which has been found to peak at 20 minutes following a stressor for obese individuals (Therrien et al., 2010). The pre-conversation and 5 and 20 minute post-conversation samples were assayed for sAA, given its more immediate and short-lived response (Nater & Rohleder, 2009).

Samples were assayed in duplicate using the Salimetrics High Sensitivity Cortisol Assay Kit, without modifications to the manufacturer’s protocol. The average intra-assay coefficient of variation was 4.60% and the average inter-assay coefficient of variation was 6.00%, which adhere to the criteria for accuracy and repeatability in salivary bioscience. Sample test volume was 25 μL of saliva per determination. The assay has a lower limit of sensitivity of 0.007012 μg/dL, and a standard curve range from 0.012 μg/dL to 3.00 μg/dL.

Video Coding

Two research assistants were trained to code conversations for autonomy support provision during the latter 5 minutes of the discussion task, which focused on the couples’ weight loss-related issues. First, coders were provided with definitions and examples of autonomy support. Second, coders practiced coding, together and independently, on the first 5 minutes of the videos. Coders used this activity to refine the codebook and address discrepancies. Finally, participants independently coded 10% of the videos. Responses from this final portion were compared using ReCal OIR (http://dfreelon.org/utils/recalfront/recal-oir/) to determine the level of intercoder reliability (ICR). After completing 10% of the video coding with an acceptable ICR of .8 or better for each variable, the remaining 90% of the videos were divided between coders and then coded separately.

The research assistants were instructed to code the 5-minute videos at the halfway point and conclusion of the conversation. Autonomy support was measured using 11 items (Gorin et al., 2014; Powers et al., 2008; Williams, et al., 2006) on a 7-point Likert type scale (1 = “Not at all” and 7 = “Completely”). Coders rated each partner individually. Example items include “He/she seemed to value his/her partner’s opinions and choices” (individual language) and “He/she used controlling language such as using ‘should’ or telling his/her partner what to do” (reverse-coded). Each participant’s scores were averaged across items coded at 2:30 and 5 minutes and demonstrated acceptable reliability (α = .90 at baseline, α = .91 at six months).

Results

All analyses were conducted using SAS v. 9.4. Cortisol and sAA data were screened for implausible readings, resulting in the removal of data from one participant who had cortisol levels above 6.00. To determine the appropriate transformation, the log of the interquartile range for each individual’s cortisol and sAA values (separately) were regressed on the log of the individual’s median value for cortisol and sAA, respectively. Subtracting 1 from the regression coefficient results in the appropriate power transformation, with a regression coefficient close to 1 indicating that a log transformation is appropriate (Tukey, 1977). Multilevel repeated measures models were conducted to account for interdependence between couple members in the data (i.e., individuals are nested within dyads). Because all couples were not distinguishable by sex (i.e., some couples were same-sex), random variance and covariances were constrained to be the same for each member of the couple (Kenny et al., 2006). Due to flexibility handling missing data in multilevel models, all participants with complete data for at least one assessment were included in the analysis.

To test the hypotheses, models were specified estimating the impact of (1) individuals’ own support provision and (2) their partner’s support provision (i.e., receiving support from their partner) on individuals’ (1) cortisol and (2) sAA, resulting in 4 different combinations of primary predictor and outcome (2×2). As depicted in Figures 1 and 2, four models were specified estimating the impact of individuals’ own support provision (i.e., actor effects of one’s support on their own outcomes, covarying partner support and partner physiological stress; see Figure 1) and their partner’s support provision (i.e., partner effects of one’s own support on the partner’s outcomes, covarying partner support and individuals’ own physiological stress; see Figure 2) on individuals’ cortisol or sAA. Note that these analyses were conducted in separate models due to nonconvergence when testing covariances of random effects in preliminary models (significant results were not substantively different in sensitivity analyses including actor and partner effects simultaneously).

Figure 1.

Figure 1.

Open in a new tab

Actor effects. Model of one’s own support provision predicting own physiological stress.

Figure 2.

Figure 2.

Open in a new tab

Partner effects. Model of one’s own support provision predicting partner’s physiological stress.

To determine necessary covariates, preliminary analyses examined potential effects of the larger study intervention (i.e., changes from baseline to 6 months) on provision of support and on cortisol and sAA, including potential effects of the intervention on changes in the pattern of cortisol/sAA readings from pre- to post-conversation. Sensitivity analyses were conducted controlling for baseline BMI and sex, both of which may influence participants’ propensity for discussing weight issues and engaging in supportive communication processes (e.g., Afifi et al., 2016; Gettens et al., 2018). Potential interactions with baseline BMI and sex were also explored.

Preliminary Analyses

Mean individual support provision was 4.93 (SD = 0.69) at baseline and 4.90 (SD = 0.74) at six months. Values for cortisol and sAA can be seen in Table 1. Preliminary analyses were conducted to examine the larger impact of the intervention, which focused on promoting autonomous self-regulation using the SDT framework, on the study variables. There was no evidence that the larger study intervention impacted changes in support provision over time (ps > .612). The pattern of cortisol (p = .314) and sAA (p = .573) before and after the conversation did not differ between baseline and 6 months. Neither was the three-way interaction between pattern of cortisol (p = .991) or sAA (p = .780), timepoint (i.e., baseline vs. 6 months), and intervention significant. The interaction between assessment timepoints (i.e., baseline vs. 6 months) and intervention was not significant in predicting sAA (p = .841), but it was significant in predicting cortisol (p = .045). The natural log of cortisol was predicted to decrease between baseline and 6 months (across all 3 readings) for control participants (B −0.20, 95% CI [−0.34, −0.07], p = .003), but there was no change for intervention participants (p = .954). In sum, the preliminary analyses found no evidence that the intervention influenced support provision, cortisol, or sAA. However, for consistency, all models included time, intervention, and the intervention by time (baseline vs. 6 months) as covariates. An indicator for the order of cortisol assessments (i.e., pre-conversation, 20 minutes post-conversation, and 40 minutes post-conversation) at each lab visit (i.e., baseline and 6 months) was also included in all models.

Table 1.

Cortisol and Salivary Alpha Amylase (Raw Score) Means at Baseline and 6 Months

N Cortisol
(μg/dL)		 Salivary Alpha-Amylase
(μg/dL)
Time Order M (SD) M (SD)
Baseline Pre-Conversation 79 0.12 (0.22) 113.36 (100.06)
Post-Conversation 1 80 0.10 (0.19) 152.17 (151.52)
Post-Conversation 2 81 0.09 (0.11) 126.90 (136.41)
6 Months Pre-Conversation 66 0.10 (0.08) 132.06 (117.73)
Post-Conversation 1 63 0.11 (0.17) 171.03 (175.58)
Post-Conversation 2 65 0.13 (0.25) 155.59 (127.27)
Open in a new tab

Note: For cortisol, post-conversation 1 was assessed at 20 minutes after the conversation and post-conversation 2 was assessed at 40 minutes after the conversation. For salivary alpha-amylase, post-conversation 1 was assessed at 5 minutes after the conversation and post-conversation 2 was assessed at 20 minutes after the conversation

Primary Analyses

Testing of H1 revealed that the main effect of partners’ support provision (i.e., receiving support from one’s partner) on individuals’ cortisol levels across all assessments was not significant (p = .239), nor did the interaction of partners’ support provision and the timing of the cortisol collection during the conversation predict cortisol levels (p = .447). However, the effect of partners’ support provision differed between baseline and 6 months (B = 0.17, 95% CI [0.02, 0.33], p = .027). At baseline, partners’ support was not associated with one’s own cortisol (B = −0.01, 95% CI [−0.14, 0.12], p = .878), but at 6 months, if one’s partner provided higher levels of support, they tended to have higher cortisol, B = 0.16, 95% CI [0.02, 0.31], p = .025. The three-way interaction between order, time, and partners’ support was not significant, p = .656. In sum, H1 was not supported.

The results of RQ1 revealed a significant main effect of individuals’ own support provision on their cortisol levels across all assessments (B = −0.15, 95% CI [−0.26, −0.03], p = .012), such that the more support individuals provided, the lower their cortisol levels. These effects did not depend on the order of the cortisol collection (p = .796), whether the collection was at baseline or 6 months (i.e., baseline vs. 6 months; p = .127), nor their three-way interaction with individual support provision (p = .707).

Testing of H2 revealed that the main effect of partners’ support provision on individuals’ sAA levels across all assessments was not significant (p = .351), nor did the interaction of partners’ support and the timing of the sAA collection during the conversation predict sAA (p = .414). However, the effect of partners’ support differed between baseline and 6 months (B = −0.22, 95% CI [−0.39, −0.06], p = .006). At baseline, partners’ support provision was positively associated with one’s own sAA levels (B = 0.16, 95% CI [0.01, 0.30], p = .031), but the effect was not significant at 6 months (B = −0.07, 95% CI [−0.22, 0.09], p = .401). The three-way interaction between order, time, and partners’ support provision was not significant (p = .674). In sum, H2 was not supported.

The results of RQ2 revealed that there was a significant main effect of individuals’ own support provision on their sAA levels across all assessments (B = 0.13, 95% CI 0.00, 0.25], p = .047), such that the more support individuals provided, the higher their sAA levels. These effects did not depend on the order of the sAA collection (p = .811), the timing of the collection (p = .108), nor their three-way interaction with individual autonomy support provision (p = .964).

Sensitivity Analyses

No significant interactions with baseline BMI emerged for either cortisol or sAA. Only one significant interaction emerged between sex and individual support provision in models predicting sAA (B = −0.30, 95% CI [−0.54, −0.07], p = .011). The effect of individual support provision on sAA was positive and significant for men (B = 0.24, 95% CI [0.09, 0.40], p = .002), but not for women (B = −0.06, 95% CI [−0.25, 0.13], p = .536).

Discussion

The study results, in contrast to the predicted associations, indicated that receiving support from one’s partner when discussing weight-related issues is associated with greater physiological stress, though the results varied between baseline and 6-months. At baseline, partners’ autonomy support provision was not associated with an individual’s own cortisol levels, but at 6 months, the more support provided by one’s partner, the higher the individual’s cortisol levels. Whether examining cortisol levels before or after the conversation, at the 6-month data collection, participants who received more support from their partner were more physiologically stressed throughout the conversation. Receiving support is not a universally positive experience, especially for populations facing health issues (Hays et al., 1997; Warner et al., 2011). For individuals who are confident in their own capabilities, receiving support may threaten autonomy (Warner et al., 2011). Thus, receiving high levels of support from a partner may have the unintended consequence of signaling a lack of trust in one’s own competency and ability to achieve weight loss goals, which may be associated with increased physiological stress. Given that this finding only emerged at 6-months and not baseline, it may suggest a pattern of learned responsiveness, such that individuals become attuned to their partner’s stress over time, and therefore learn to provide more support in response. In other words, one partner’s stress may elicit the other partner’s support.

It is also possible that the support individuals received did not match the support they desired. Researchers have used the concept of standards for support to emphasize the need to not only consider the quality of supportive messages, but also whether such messages are in line with the amount and type of support the recipient desires (Joseph et al., 2015). Indeed, receiving an under- or over-provision of support is associated with declining marital quality, and receiving more support than desired is even more detrimental in some cases (Brock & Lawrence, 2009). This line of research reveals that gaps or discrepancies between the type and amount of support a person desires versus what they receive can be deleterious for individuals and their relationships. Though speculative, it is possible that the positive association between receiving support and cortisol in the present study may represent an over-provision of support. In other words, individuals may have received more support than desired, which may have resulted in elevated physiological stress.

Despite its positive association with cortisol, support may nonetheless be beneficial. Although greater autonomy support from a partner during the weight-related conversation was associated with higher cortisol levels, the benefits of such support may transcend single conversations, and in the long term, assist individuals in meeting their weight-related goals. Indeed, prior research reveals that autonomy support aids in weight loss outcomes (Gorin et al., 2014; Powers et al., 2008; Silva et al., 2011). As such, support may be salubrious for both weight management and relational well-being long-term, and potentially exert downstream benefits for general stress levels.

A positive association also emerged between partner support provision and individuals’ sAA, but only at baseline, which echoes sAA’s link to arousal more broadly (including both approach and avoidance behaviors; Adam et al., 2011; Fortunato et al., 2008), and not only stress. Higher sAA levels during the initial weight loss conversation may reflect individuals’ positive feelings and excitement in response to receiving autonomy support from their partner. It would make sense that receiving autonomy support, and the resulting positive affect associated with that support, is heightened at baseline for individuals given that the experience is novel and that the effect is allayed 6 months later.

Providing Support is Linked to Lower Stress but Greater Arousal

Given that a negative association emerged between support provision and cortisol regardless of when during the conversation the sample was collected, it is unclear whether providing support to a partner influences one’s cortisol levels, or whether cortisol levels drive supportive behavior. Recent theorizing that draws upon work on ego depletion and self-control suggests that when an individual is depleted of resources within a stressful interaction, it can be difficult to engage in positive partner communication (Afifi et al., 2016). Given evidence for increased cortisol levels as a marker of allostatic load (McEwen, 2000), and the association between allostatic load and reduced cognitive ability (Beckie, 2012), individuals with lower cortisol may have more cognitive resources to engage in autonomy supportive behaviors as they communicate about weight loss with their partner.

The negative association between providing support and cortisol may also be reflective of the benefits of affectionate communication. Indeed, prior work reveals that individuals with a greater propensity for communicating affection report lower stress levels and overall better mental health (Floyd, 2002). Expressing affection is also associated with higher self-esteem, greater relationship satisfaction, and less fear of intimacy, even after controlling for the effects of receiving affection (Floyd et al., 2005). Providing support is considered one form of affectionate communication (Floyd & Morman, 1998), and as such, the present findings serve to reinforce the advantages of expressing affection for physiological stress and the underlying tenet of AET that affectionate communication offers benefits for physical health (see Floyd et al., 2020, for a review).

Conversely, individuals who provided more support to their partner had higher levels of sAA (regardless of when the sAA was collected during the conversation). These findings make sense when considering that sAA increases in response to challenging tasks (McKay, 2010) and is positively associated with anxiety and emotional arousal, including positive mood states (Adam et al., 2011; see Nater & Rohleder, 2009, for a review). Thus, the contradictory cortisol and sAA findings may reflect the fact that providing support can be rewarding and stress-ameliorating (i.e., lower cortisol), but can also be challenging and exciting (i.e., higher sAA).

The sensitivity analyses revealed that for male participants, their own support was positively associated with their sAA levels, but the same association was not significant for female participants. Providing support may be uncomfortable for men given that its qualities are uncharacteristic of prescriptive gender norms for masculine behavior (Afifi et al., 2016), and thus the link between support provision and sAA may be stronger for men. However, given that only one significant interaction with sex was revealed across all the models, the findings indicate that generally the association between support provision and physiological stress was similar for women and men.

Strengths, Limitations, and Future Directions

Investigating couples’ physiological stress responses to weight talk offers an innovative means of understanding how everyday communication may impede or improve the success of weight loss interventions. However, a potential limitation of the study is the failure to assess other aspects of the autonomy support training, such as perceived competency and frequency of support provision prior to the intervention; therefore, future research should consider measuring individuals’ efficacy to engage in supportive communication. It is also important to note that several relevant conditions were not assessed that may have influenced the findings, such as hypo/hypercortisolism, smoking, menstrual cycle status, and hormonal contraceptive use. Furthermore, the findings are based on observational coding of supportive interactions. Further research should also examine individuals’ perceptions of their partner’s support to determine if meaningful differences emerge with respect to the associations between support and physiological stress when considering the actual viewpoints of conversation partners as opposed to relying exclusively on trained coders. Finally, it worth noting that due to the limitations of our available funding and the size of the parent study in which the present analyses were embedded, our study only focused on 47 couples, and therefore, we may have been underpowered to detect significant effects. However, the lower measurement error associated with physiological data permits a smaller sample size, as reflective in other studies exploring physiological stress responses to couples’ difficult conversations (e.g., Aloia & Solomon, 2015; Denes et al., 2020). Despite these limitations, the present study contributes to the literature by using dyadic data to investigate the actor and partner effects of support provision on physiological stress responses to couples’ conversations about weight loss, revealing that cortisol and sAA each have unique associations with providing and receiving autonomy support in close relationships.

The unexpected positive association between receiving support and cortisol points to the need for future research to unpack the complex experience of discussing weight with a romantic partner, particularly when both partners are involved in weight loss efforts, to determine the conditions under which support can aid in managing the stress of weight management. For example, future studies should ask participants to identify baseline levels of support received prior to the intervention as well as preferred ways of receiving support, and use such information to design interventions specific to relational partners’ needs. Rather than assume that all couples, or even both partners within a couple, profit from the same type of support, individuals may benefit more if support is matched to their unique preferences and needs. It would also be important to reassess support preferences throughout the spectrum of the BWL program, as the support desired when embarking on a weight loss program may shift as partners face obstacles, reach goals, and transition to management strategies. Such an approach would also allow researchers to determine the effect of support training (as part of the intervention) on support gaps. For example, in the present study, it is possible that being trained to provide autonomy support established certain expectations surrounding the quality of support that one expected to receive. If partners were then unable to provide such support or did so in a way that was perceived to be inadequate by their partner, then the support training may have inadvertently increased the discrepancy between the type or amount of support the recipients desired and what they actually received. Taken together, tracking partners’ preferences, expectations, and received support, as well as relational well-being, health outcomes, and stress over the course of the intervention would help disentangle the longitudinal effects of supportive conversations and physiological responses to such discussions on physical and relational health.

In sum, the present study makes a unique contribution to the current literature by exploring both cortisol and sAA levels in response to conversations about weight loss (as part of a 6-month BWL intervention) among couples where both partners were overweight or obese. The findings indicate that receiving support from a romantic partner when engaging in “weight talk” is associated with greater physiological stress and/or arousal, but that providing support may be stress-ameliorating. Such results point to the importance of exploring both actor and partner effects by using dyadic data to better understand the interdependent nature of not only couples’ communication, but also their physiological response patterns.

Author Note and Acknowledgments:

We would like to thank the TEAMS Research Group: Maja Barnouw, Kate Boudreau, Christina Cardwell, Jaime Foster, Lynn Guibbory, Nataliya Korostensky, Elizabeth Lamonte, Nana Marfo, Tabea Mueller, Ambyre Ponivas, Emma-Bee Schwarz, Arielle Sherman-Golembeski, Erin Stolz, Julia Werth, Olivia Wilson, Sanna Wortel, and Emily Wyckoff.

Funding: This work was supported by the National Institutes of Health/National Heart, Lung, and Blood Institute under Grant NIH/NHLBI HL125157. This study is registered under clinicaltrials.gov NCT02570009. This work was also supported by the Office of the Vice President for Research at the University of Connecticut through their Research Excellence Program.

Footnotes

Disclosure Statement: The authors declare that they have no conflict of interest.

Data Availability Statement: The study data is not available, but coding materials can be obtained from the first author.

Ethical Standards and Consent: All study procedures were approved by the Institutional Review Board at the University of Connecticut. Informed consent was obtained from all individual participants included in the study.

Contributor Information

Amanda Denes, Department of Communication, University of Connecticut.

John P. Crowley, Department of Communication, University of Delaware

Ambyre L.P. Ponivas, Departments of Communication Studies and Psychology, Young Harris College

Talea Cornelius, Center for Behavioral Cardiovascular Health, Columbia University Irving Medical Center.

Ryan J. Allred, Department of Communication Studies, University of Wisconsin Oshkosh

Katelyn M. Gettens, Department of Psychiatry, Massachusetts General Hospital Department of Neurology, Brigham and Women’s Hospital.

Theodore A. Powers, Psychology Department, University of Massachusetts Dartmouth

Amy A. Gorin, Institute for Collaboration on Health, Intervention, and Policy (InCHIP) and Psychological Sciences, University of Connecticut

References

  1. Adam EK, Hoyt LT, & Granger DA (2011). Diurnal alpha amylase patterns in adolescents: Associations with puberty and momentary mood states. Biological Psychology, 88, 170–173. 10.1016/j.biopsycho.2011.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Afifi TD, Granger DA, Denes A, Joseph A, & Aldeis D (2011). Parents’ communication skills and adolescents’ salivary α-amylase and cortisol response patterns. Communication Monographs, 78(3), 273–295. 10.1080/03637751.2011.589460 [DOI] [Google Scholar]
  3. Afifi TD, Merrill AF, & Davis S (2016). The theory of resilience and relational load. Personal Relationships, 23, 663–683. 10.1111/pere.12159 [DOI] [Google Scholar]
  4. Afifi TD, Davis S, Merrill A, Coveleski S, Denes A, & Afifi WA (2015). In the wake of the Great Recession: Economic uncertainty, communication, and physiological stress responses in families. Human Communication Research. 41, 268–302. 10.1111/hcre.12048 [DOI] [Google Scholar]
  5. Afifi TD, Granger D, Joseph A, Denes A, & Aldeis D (2015). The influence of divorce and parents’ communication skills on adolescents’ and young adults’ stress reactivity and recovery. Communication Research, 42, 1009–1042. 10.1177/0093650213509665 [DOI] [Google Scholar]
  6. Aloia LS, & Solomon DH (2015). Conflict intensity, family history, and physiological stress reactions to conflict within romantic relationships. Human Communication Research, 41, 367–389. 10.1111/hcre.12049 [DOI] [Google Scholar]
  7. Beckie TM (2012). A systematic review of allostatic load, health, and health disparities. Biological Research for Nursing, 14, 311–346. 10.1177/1099800412455688 [DOI] [PubMed] [Google Scholar]
  8. Black DR, Gleser LJ, & Kooyers KJ (1990). A meta-analytic evaluation of couples weight-loss programs. Health Psychology, 9, 330. 10.1037//0278-6133.9.3.330 [DOI] [PubMed] [Google Scholar]
  9. Bove CF, & Sobal J (2011). Body weight relationships in early marriage. Weight relevance, weight comparisons, and weight talk. Appetite, 57, 729–742. 10.1016/j.appet.2011.08.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brock RL, & Lawrence E (2009). Too much of a good thing: Underprovision versus overprovision of partner support. Journal of Family Psychology, 23, 181–192. 10.1037/a0015402 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Brown SL, Nesse RM, Vinokur AD, & Smith DM (2003). Providing social support may be more beneficial than receiving it: Results from a prospective study of mortality. Psychological Science, 14, 320–327. 10.1111/1467-9280.14461 [DOI] [PubMed] [Google Scholar]
  12. Butryn ML, Webb V, & Wadden TA (2011). Behavioral treatment of obesity. Psychiatric Clinics of North America, 34, 841–59. 10.1016/j.psc.2011.08.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310–357. 10.1037/0033-2909.98.2.310 [DOI] [PubMed] [Google Scholar]
  14. Crowley JP, Burke TJ, Denes A, Allred RJ, & Carberry E (2020). Navigating relational turbulence in the wake of weight loss transitions: A support marshaling analysis. Personal Relationships, 27, 630–651. 10.1111/pere.12335 [DOI] [Google Scholar]
  15. Deci EL, Eghrari H, Patrick BC, & Leone DR (1994). Facilitating internalization: The self‐determination theory perspective. Journal of Personality, 62, 119–142. 10.1111/j.1467-6494.1994.tb00797.x [DOI] [PubMed] [Google Scholar]
  16. Denes A, Bennett M, & Winkler KL (2017). Exploring the benefits of affectionate communication: Implications for interpersonal acceptance-rejection theory. Journal of Family Theory & Review, 9, 491–506. 10.1111/jftr.12218 [DOI] [Google Scholar]
  17. Denes A, Crowley JP, Winkler KL, Ponivas ALP, Dhillon A, & Bennett M (2020). Exploring the effect of pillow talk on relationship satisfaction and physiological stress responses to couples’ difficult conversations. Communication Monographs, 87, 267–290. 10.1080/03637751.2020.1726424 [DOI] [Google Scholar]
  18. Egger G, & Swinburn B (1997). An “ecological” approach to the obesity pandemic. BMJ, 315, 477–480. 10.1136/bmj.315.7106.477 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Epel E, Lapidus R, McEwen B, & Brownell K (2001). Stress may add bite to appetite in women: a laboratory study of stress-induced cortisol and eating behavior. Psychoneuroendocrinology, 26, 37–49. 10.1016/S0306-4530(00)00035-4 [DOI] [PubMed] [Google Scholar]
  20. Floyd K (2002). Human affection exchange V: Attributes of the highly affectionate. Communication Quarterly, 50, 135–152. 10.1080/01463370209385653 [DOI] [Google Scholar]
  21. Floyd K (2006). Communicating affection: Interpersonal behavior and social context. Cambridge University Press. [Google Scholar]
  22. Floyd K, Hess JA, Miczo LA, Halone KK, Mikkelson AC, & Tusing KJ (2005). Human affection exchange: VIII. Further evidence of the benefits of expressed affection. Communication Quarterly, 53, 285–303. 10.1080/01463370500101071 [DOI] [Google Scholar]
  23. Floyd K, Mikkelson AC, Tafoya MA, Farinelli L, La Valley AG, Judd J, Haynes MT, Davis KL, & Wilson J (2007). Human affection exchange: XIII. Affectionate communication accelerates neuroendocrine stress recovery. Health Communication, 22, 123–132. 10.1080/10410230701454015 [DOI] [PubMed] [Google Scholar]
  24. Floyd K, & Morman MT (1998). The measurement of affectionate communication. Communication Quarterly, 46, 144–162. 10.1080/01463379809370092 [DOI] [Google Scholar]
  25. Floyd K, Pavlich CA, Dinsmore DR, & Ray CD (2020). The physiology of affectionate communication. In Aloia LS, Denes A, & Crowley JP (Eds.), The Oxford handbook of the physiology of interpersonal communication (pp. 31–47). Oxford University Press. [Google Scholar]
  26. Floyd K, & Riforgiate S (2008). Affectionate communication received from spouses predicts stress hormone levels in healthy adults. Communication Monographs, 75, 351–368. 10.1080/03637750802512371 [DOI] [Google Scholar]
  27. Fortunato CK, Dribin AE, Granger DA, & Buss KA (2008). Salivary alpha‐amylase and cortisol in toddlers: Differential relations to affective behavior. Developmental Psychobiology: The Journal of the International Society for Developmental Psychobiology, 50, 807–818. 10.1002/dev.20326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Gettens KM, Carbonneau N, Koestner R, Powers TA, & Gorin AA (2018). The role of partner autonomy support in motivation, well-being, and weight loss among women with higher baseline BMI. Families, Systems, & Health, 36, 347–356. 10.1037/fsh0000362 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gordis EB, Granger DA, Susman EJ, & Trickett PK (2006). Asymmetry between salivary cortisol and α-amylase reactivity to stress: Relation to aggressive behavior in adolescents. Psychoneuroendocrinology, 31, 976–987. 10.1016/j.psyneuen.2006.05.010 [DOI] [PubMed] [Google Scholar]
  30. Gorin AA, Powers TA, Gettens K, Cornelius T, Koestner R, Mobley AR, Pescatello LS, & Huedo-Medina TB (2020). A randomized controlled trial of a theory-based weight-loss program for couples. Health Psychology, 39(2), 137–146. 10.1037/hea0000808 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gorin AA, Powers TA, Gettens K, Cornelius T, Koestner R, Mobley AR, Pescatello LS, & Medina TH (2017). Project TEAMS (Talking about Eating, Activity, and Mutual Support): a randomized controlled trial of a theory-based weight loss program for couples. BMC Public Health, 17(1), 749. 10.1186/s12889-017-4732-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Gorin AA, Powers TA, Koestner R, Wing RR, & Raynor HA (2014). Autonomy support, self-regulation, and weight loss. Health Psychology, 33, 332–339. 10.1037/a0032586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Gorin AA, Raynor HA, Fava J, Maguire K, Robichaud E, Trautvetter J, Crane M, & Wing RR (2013). Randomized controlled trial of a comprehensive home environment-focused weight-loss program for adults. Health Psychology, 32, 128–137. 10.1037/a0026959 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Gosnell CL, & Gable SL (2017). You deplete me: Impacts of providing positive and negative event support on self-control. Personal Relationships, 24, 598–622. 10.1111/pere.12200 [DOI] [Google Scholar]
  35. Granger DA, Kivlighan KT, Blair C, El-Sheikh M, Mize J, Lisonbee JA, Buckhalt JA, Stroud LR, Handwerger K, & Schwartz EB (2006). Integrating the measurement of salivary α-amylase into studies of child health, development, and social relationships. Journal of Social and Personal Relationships, 23(2), 267–290. 10.1177/0265407506062479 [DOI] [Google Scholar]
  36. Hays JC, Saunders WB, Flint EP, Kaplan BH, & Blazer DG (1997). Social support and depression as risk factors for loss of physical function in late life. Aging & Mental Health, 1, 209–220. 10.1080/13607869757092 [DOI] [Google Scholar]
  37. Jackson SE, & Steptoe A (2018). Obesity, perceived weight discrimination, and hair cortisol: A population-based study. Psychoneuroendocrinology, 98, 67–73. 10.1016/j.psyneuen.2018.08.018 [DOI] [PubMed] [Google Scholar]
  38. Joseph A, Afifi TD, & Denes A (2016). (Unmet) Standards for emotional support and their short-and medium-term consequences. Communication Monographs, 83, 163–193. 10.1080/03637751.2015.1068432 [DOI] [Google Scholar]
  39. Kenny DA, Kashy DA, & Cook WL (2006). Dyadic data analysis. New York: Guilford Press. [Google Scholar]
  40. Leroux JS, Moore S, & Dubé L (2013). Beyond the “I” in the obesity epidemic: a review of social relational and network interventions on obesity. Journal of Obesity, 2013, 1–10. 10.1155/2013/348249 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. McEwen BS (2000). The neurobiology of stress: From serendipity to clinical relevance. Brain Research, 886, 172–189. 10.1016/S0006-8993(00)02950-4 [DOI] [PubMed] [Google Scholar]
  42. McKay KAC, Buen JE, Bohan KJ, & Maye JP (2010). Determining the relationship of acute stress, anxiety, and salivary α-amylase level with performance of student nurse anesthetists during human-based anesthesia simulator training. AANA Journal, 78, 301–309. [PubMed] [Google Scholar]
  43. McLean N, Griffin S, Toney K, & Hardeman W (2003). Family involvement in weight control, weight maintenance and weight-loss interventions: a systematic review of randomised trials. International Journal of Obesity, 27, 987–1005. [DOI] [PubMed] [Google Scholar]
  44. Nater UM, & Rohleder N (2009). Salivary alpha-amylase as a non-invasive biomarker for the sympathetic nervous system: current state of research. Psychoneuroendocrinology, 34, 486–496. 10.1016/j.psyneuen.2009.01.014 [DOI] [PubMed] [Google Scholar]
  45. NHLBI (1998). Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: Executive summary. expert panel on the identification, evaluation, and treatment of overweight in adults. The American Journal of Clinical Nutrition, 68, 899–917. 10.1093/ajcn/68.4.899 [DOI] [PubMed] [Google Scholar]
  46. Powers TA, Koestner R, & Gorin AA (2008). Autonomy support from family and friends and weight loss in college women. Families, Systems, & Health, 26, 404–416. 10.1037/1091-7527.26.4.404 [DOI] [Google Scholar]
  47. Ryan RM, & Deci EL (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68–78. 10.1037/110003-066X.55.1.68 [DOI] [PubMed] [Google Scholar]
  48. Ryan RM, & Deci EL (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54–67. 10.1006/ceps.1999.1020 [DOI] [PubMed] [Google Scholar]
  49. Sallis J, Bauman A, & Pratt M (1998). Environmental and policy interventions to promote physical activity. American Journal of Preventive Medicine, 15, 379–397. 10.1016/S0749-3797(98)00076-2 [DOI] [PubMed] [Google Scholar]
  50. Silva MN, Markland D, Carraça EV, Vieira PN, Coutinho SR, Minderico CS, Matas MG, Sardinha LB, & Teixeira PJ (2011). Exercise autonomous motivation predicts 3-yr weight loss in women. Medicine & Science in Sports & Exercise, 43, 728–737. 10.1249/MSS.0b013e3181f3818f [DOI] [PubMed] [Google Scholar]
  51. Silva MN, Vieira PN, Coutinho SR, Minderico CS, Matos MG, Sardinha LB, & Teixeira PJ (2010). Using self-determination theory to promote physical activity and weight control: a randomized controlled trial in women. Journal of Behavioral Medicine, 33, 110–122. 10.1007/s10865-009-9239-y [DOI] [PubMed] [Google Scholar]
  52. Therrien F, Drapeau V, Lalonde J, Lupien SJ, Beaulieu S, Doré J, Tremblay A, & Richard D (2010). Cortisol response to the Trier Social Stress Test in obese and reduced obese individuals. Biological Psychology, 84, 325–329. 10.1016/j.biopsycho.2010.03.013 [DOI] [PubMed] [Google Scholar]
  53. Tomiyama AJ (2014). Weight stigma is stressful. A review of evidence for the Cyclic Obesity/Weight-Based Stigma model. Appetite, 82, 8–15. 10.1016/j.appet.2014.06.108 [DOI] [PubMed] [Google Scholar]
  54. Tomiyama AJ, Mann T, Vinas D, Hunger JM, DeJager J, & Taylor SE (2010). Low calorie dieting increases cortisol. Psychosomatic Medicine, 72, 357–364. 10.1097/PSY.0b013e3181d9523c [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Tukey JW (1977). Exploratory data analysis. London: Pearson. [Google Scholar]
  56. Warner LM, Ziegelmann JP, Schüz B, Wurm S, Tesch-Römer C, & Schwarzer R (2011). Maintaining autonomy despite multimorbidity: Self-efficacy and the two faces of social support. European Journal of Ageing, 8, 3–12. 10.1007/s10433-011-0176-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Williams GC, Grow VM, Freedman ZR, Ryan RM, & Deci EL (1996). Motivational predictors of weight loss and weight-loss maintenance. Journal of Personality and Social Psychology, 70, 115–126. 10.1037/0022-3514.70.1.115 [DOI] [PubMed] [Google Scholar]
  58. Williams GC, Lynch MF, McGregor HA, Ryan RM, Sharp D, & Deci EL (2006). Validation of the” important other” climate questionnaire: Assessing autonomy support for health-related change. Families, Systems, & Health, 24, 179–194. 10.1037/1091-7527.24.2.179 [DOI] [Google Scholar]

RESOURCES