Emotion regulation strategies relate to ambulatory cardiovascular activity in an American Indian community

Abstract

Objective:

Poor emotion regulation is associated with risk for cardiovascular disease (CVD). However, much of this research is conducted in primarily White samples, thus limiting our understanding of this relationship in other racial/ethnic groups. American Indians (AIs) are uniquely and disproportionately at risk for CVD. As such, the present study aimed to examine the relationships between emotion regulation strategies and ambulatory cardiovascular activity in an entirely AI sample.

Methods:

The sample consisted of 100 AI adults living on a tribal reservation. Emotion regulation strategies (expressive suppression, cognitive reappraisal) were assessed via the Emotion Regulation Questionnaire (ERQ). Using Ecological Momentary Assessment (EMA), daily measurements of psychological stress and ambulatory cardiovascular activity were taken during a seven-day monitoring period. Statistical analyses included bivariate correlations, hierarchical linear regression models, and mediation models.

Results:

Expressive suppression was associated with higher systolic and diastolic blood pressure, as well as higher pulse rate. In contrast, cognitive reappraisal was associated with lower systolic and diastolic blood pressure, lower pulse rate, as well as lower average daily psychological stress. These results remained statistically significant after adjusting for age, sex, body mass index, anxiety, depression, and early life trauma. In addition, psychological stress mediated the associations between blood pressure and cognitive reappraisal, but not expressive suppression.

Conclusions:

These results provide preliminary evidence for divergent associations of two emotion regulation strategies with cardiovascular activity and psychological stress in an AI community. Modifying health interventions to include training in effective emotion regulation may be beneficial.

INTRODUCTION

How we regulate our day-to-day emotions can influence our long-term mental and physical well-being (1). Emotion regulation is the modification of which emotion is being experienced, when it is experienced, and how it is expressed (2). The inability to effectively manage one’s emotions is associated with risk for psychopathology (e.g., depression; 3) as well as physical illness (e.g., cardiovascular disease [CVD]; 4). As such, identifying which emotion regulation strategies are most predictive of poor health outcomes is crucial for reducing disease risk and developing effective interventions.

Two of the most frequently researched emotion regulation strategies are cognitive reappraisal and expressive suppression (5). Cognitive reappraisal involves reinterpretation of a situation to modify its emotional impact and is often associated with decreased negative emotions and increased positive emotions (5, 6). Alternatively, expressive suppression (i.e., the inhibition of emotional expression) is associated with decreased positive but not negative emotions, as well as increased sympathetic nervous system arousal and risk for CVD (1, 4, 5, 7). For example, a one standard deviation increase in suppression is associated with a 10% increase in estimated likelihood of developing CVD in 10 years (8). However, most of this past research has been conducted in limited samples (e.g., primarily White), thus restricting our ability to understand how emotion regulation differentially impacts health across various cultures (9, 10).

Evidence suggests that selection, as well as perceived consequences of emotion regulation strategies vary depending on context and cultural values (10). Expressive suppression is more maladaptive in individualist cultures compared to collectivist cultures (11, 12). In U.S. samples, the use of expressive suppression is higher amongst ethnoracial minorities, suggesting that ethnoracial minorities may be more motivated to down-regulate emotional expression (for review, see 13). A recent review found that African Americans frequently use expressive suppression as a coping mechanism in response to racial discrimination (14), and research suggests that suppression may in fact be health protective in this context, specifically for African American women (15). Moreover, while cognitive reappraisal is found to be beneficial in predominately non-Hispanic White samples (1, 5), it is less helpful in the context of oppression, particularly for Latinos (16).

That said, there is one population that has received little to no attention in the emotion regulation literature, namely, American Indians (AIs). This is a noteworthy gap in the literature, as the AI community is known to experience disproportionate levels of mental and physical health disparities, including but not limited to mood disorders, diabetes, and CVD (17–19). One recent prospective study found that greater use of expressive suppression and lower use of cognitive reappraisal predicted post-traumatic stress symptoms in response to the COVID-19 pandemic in a nationally representative sample of AI adults (20). To our knowledge, no study to date has attempted to identify how emotion regulation strategies relate to physical health outcomes in an AI sample.

Research has identified CVD to be a leading cause of death in the AI community (for review, see 21). These findings highlight a need for research that examines how psychosocial factors, such as emotion regulation, contribute to the prevalence of CVD in this community. As such, the purpose of this study was to examine the relationships between habitual use of two emotion regulation strategies (expressive suppression, cognitive reappraisal) and variables known to be associated with risk for CVD (psychological stress, ambulatory cardiovascular activity; 22, 23) in a sample of AI adults. Ecological Momentary Assessment (EMA) was used to assess these relationships during the everyday lives of AI adults, thus allowing us to capture experiences as they naturally occur in real life settings (24). This approach affords strong ecological validity (24) and has previously been utilized in other groups to obtain daily measures of health risk factors (25, 26).

METHODS

Participants and Procedures

This study was approved by a tribal college Institutional Review Board and the Montana State University Institutional Review Board and utilized community based participatory research (CBPR) methods. CBPR emphasizes a partnership approach to research, such that researchers and community members are equitably involved in every stage of the research process (27, 28). Therefore, a Community Advisory Board (CAB) consisting of four AI adult community members reviewed and approved the research plan, questionnaires, and measurements, as well as provided feedback on the interpretation of results and dissemination decisions. In particular, the CAB played a vital role in determining the timing and duration of all measurements, such as to not place too much burden on the study participants. This study was not preregistered. Data collection began in July of 2018 and finished in October of 2018.

Participants (N = 100; Mean (SD) [range] age = 42.20 (15.05) [20–78] years; 54.6% female) were recruited through advertisements placed on AI social media sites. To be eligible, participants needed to be over the age of 18, self-identify as AI, and currently reside on an AI reservation. Participants were excluded if they were clinically diagnosed with a sleep disorder or chronic health condition, or if they were taking regular prescription medicine. We did not conduct formal a priori power analyses for this study. However, post hoc sensitivity analyses reveal that our sample size was large enough to detect a minimum small to medium effect size of the main outcomes (f² = 0.10), with power set at 0.80 and α = .05. The initial study visit took place on a Tribal Community College campus in Montana, during which participants provided written informed consent and completed questionnaires. The project coordinator obtained anthropometric measurements and installed the Illumivu EMA application (www.lifedatacorp.com) onto participants’ mobile devices. Participants were prompted through the mobile application to respond to survey questions at the beginning and end of each day for a seven-day monitoring period. Participants were also given an ambulatory blood pressure monitor, which they wore for three weekdays of the monitoring period.

Measures

Emotion Regulation Strategies

Habitual use of expressive suppression and cognitive reappraisal was assessed using the Emotion Regulation Questionnaire (ERQ; 5). The ERQ includes 10 items, which can be divided into two subscales: expressive suppression (four items, e.g., “I control my emotions by not expressing them”) and cognitive reappraisal (six items, e.g., “When I’m faced with a stressful situation, I make myself think about it in a way that helps me stay calm”). Participants were asked to indicate how much they agree with each statement using a seven-point Likert-type scale (1 = “strongly disagree” to 7 = “strongly agree”). A higher subscale score indicates greater use of that respective strategy. In the present study, Cronbach’s α = 0.87 for reappraisal subscale and 0.83 for suppression subscale.

Psychological Stress

Psychological stress was assessed using a subscale pulled from the Depression, Anxiety, and Stress Scale-21 items (DASS-21; 29). This seven-item subscale is designed to measure the degree of chronic non-specific arousal with regards to the past week. In the present study, the instructions were adapted to ask participants to indicate on a four-point scale how much each statement applied to them over the course of each day (0 = “Did not apply to me at all” to 3 = “Applied to me very much or most of the time”). Participants completed the questionnaire at the end of each day during the week-long monitoring period. Example items include, “I found it difficult to relax,” and “I found myself getting agitated.” A higher score indicates greater psychological stress. In the present study, across the week-long monitoring period, the Cronbach’s α = 0.78 for the DASS-21 stress subscale.

Ambulatory Cardiovascular Activity

Ambulatory blood pressure (i.e., systolic and diastolic blood pressure; SBP/DBP) and pulse rate (PR) was obtained using the QardioArm blood pressure monitor, which has been previously validated in adults (30). For three weekdays, participants were prompted by the Illumivu EMA application on their mobile device to self-activate the blood pressure cuff at 10 AM, 1 PM, 4 PM, and 7 PM, resulting in a total of 12 ambulatory cardiovascular activity measurements. Prior to obtaining each recording, participants were reminded by the application to sit down with their legs uncrossed. Participants were asked to remain seated for one minute before initiating the reading.

Covariates

Demographic variables, such as age and sex (1 = male, 2 = female) were assessed via self-report and included as covariates. Self-reported anxiety, depression, and early life trauma were also included as covariates, given these variables are known to be associated with emotion regulation, as well as our outcome variables (3, 31, 33). Symptoms of anxiety and depression were assessed using the 14-item Hospital Anxiety and Depression Scale (HADS; 32; Cronbach’s α for anxiety subscale = 0.85, depression subscale = 0.88). Early life trauma was assessed via the Risky Families Questionnaire (RFQ; 33; Cronbach’s α = 0.70), which examines exposure to physical and emotional abuse or neglect between the ages of 5–15. Lastly, there is an established relationship between body mass index (BMI) and blood pressure (34). As such, BMI was calculated and controlled for using height and weight measurements obtained during the initial visit (BMI = weight (lbs) / [height (in)]² × 703).

Statistical Analyses

Pearson’s product moment correlations were employed to determine associations between the main variables of interest and covariates. Next, four separate hierarchical linear regressions were performed to assess whether emotion regulation strategies were associated with ambulatory SBP, ambulatory DBP, ambulatory PR, and psychological stress. The analyses were first run without adjustment, then with adjustment for demographic covariates (i.e., age, sex, BMI), and then with additional adjustment for emotional distress covariates (i.e., anxiety, depression, early life trauma). In all models, a priori established covariates were entered in Step 1, while suppression and reappraisal were simultaneously entered in Step 2. Follow up mediation models were conducted via PROCESS for SPSS (Version 4.0; 35), through which indirect effects were investigated separately for each model to determine whether psychological stress mediated the associations between emotion regulation and cardiovascular activity.

Data were screened for missing values. Missing values were observed for cognitive reappraisal (n = 8), expressive suppression (n = 3), or both (n = 6). Eleven participants were also missing 1–2 physiology readings across the three days of monitoring. Little’s MCAR test (36) confirmed this data was missing completely at random χ²(754) = 769.52, p = 0.34. Participants were excluded from any analysis in which they had missing emotion regulation data but were not removed if missing 1–2 physiology readings (i.e., final averages were simply calculated from available data). It should be noted that we also did not exclude any physiological data that fell above or below certain thresholds. Statistical analyses were conducted in SPSS version 28 (IBM Corp, USA). Results were reported as statistically significant if p values were ≤ .05.

RESULTS

Bivariate correlations for the main variables of interest and covariates are in Table 1. No statistically significant association was observed between ERQ suppression and ERQ reappraisal (r = .19, p = .081), suggesting independence between the subscales. Emotion regulation strategies were negatively associated with early life trauma (r = −.42, p < .001 for reappraisal, r = −.24, p = .024 for suppression), but were not significantly related to any of the other covariates (ps ≥ .12). Sample demographics and descriptive statistics are in Supplementary Digital Content, Table S1. All cardiovascular activity values were normally distributed.

Table 1.

Bivariate Correlations for Main Variables and Covariates

Variable	1	2	3	4	5	6	7	8	9	10	11
1. Age	__
2. Sex	.14	__
3. BMI	.09	.02	__
4. Anxiety	.11	.06	.03	__
5. Depression	.004	.13	.03	.61**	__
6. Early Life Trauma	.19	.18	.23*	.27**	.31**	__
7. Suppression	.16	.03	.02	−.10	.004	−.24*	__
8. Reappraisal	−.01	−.08	−.05	−.13	−.17	−.42**	.19	__
9. Psychological Stress	.14	.14	.18	.12	.12	.14	−.03	−.42**	__
10. Ambulatory SBP	.04	.12	.08	.04	.14	.26*	.23*	−.47**	.43**	__
11. Ambulatory DBP	.06	.07	.08	.09	.11	.12	.29**	−.36**	.47**	.70**	__
12. Ambulatory PR	.08	.09	.20*	.04	.12	.24*	.20	−.36**	.34**	.77**	.58**

Note.

^*Denotes p < .05

^**denotes p < .001.

Emotion Regulation and Ambulatory Cardiovascular Activity

In unadjusted models, suppression was statistically significantly and positively associated with ambulatory SBP, B = 1.97, p = .001, 95 % CI [0.81, 3.13], DBP, B = 1.60, p = .001, 95 % CI [0.65, 2.55], and PR, B = 1.23, p = .013, 95 % CI [0.27, 2.20]. Reappraisal was statistically significantly and negatively associated with ambulatory SBP, B = −3.12, p < .001, 95 % CI [−4.22, −2.02], DBP, B = −1.91, p < .001, 95 % CI [−2.81, −1.02], and PR, B = −1.83, p < .001, 95 % CI [−2.74, −0.92]. Together, suppression and reappraisal significantly explained 30.2% of the variance in ambulatory SBP, 21.8% of the variance in ambulatory DBP, and 17.4% of the variance in ambulatory PR.

After adjusting for demographic covariates in Step 1, suppression and reappraisal were still statistically significant predictors when entered in Step 2, together explaining 29.8% of the variance in ambulatory SBP, F(5, 77) = 7.97, p < .001, 19.6% of the variance in ambulatory DBP, F(5, 77) = 5.00, p < .001, and 18.7% of the variance in ambulatory PR, F(5, 77) = 4.78, p < .001.

Finally, additional adjustment for emotional distress covariates, entered in Step 1, did not significantly explain any variance in average ambulatory SBP, DBP, or PR (ps ≥ .18). However, the inclusion of suppression and reappraisal in Step 2 significantly explained 28% of the variance in ambulatory SBP, F(8, 74) = 4.99, p < .001, 17.1% of the variance in ambulatory DBP, F(8, 74) = 3.12, p = .004, and 17.1% of the variance in ambulatory PR, F(8, 74) = 3.11, p = .004. Higher use of suppression predicted higher ambulatory cardiovascular activity, whereas higher use of reappraisal predicted lower ambulatory cardiovascular activity (see Table 2).

Table 2.

Regression Models for Suppression and Reappraisal Predicting Ambulatory Cardiovascular Activity

	Ambulatory SBP			Ambulatory DBP			Ambulatory PR
Model	B	p	95% CI	B	p	95% CI	B	p	95% CI
Step 1
Age	0.03	.69	−0.12, 0.17	0.03	.59	−0.08, 0.15	0.02	.69	−0.09, 0.13
Sex	2.60	.22	−1.59, 6.78	1.64	.33	−1.70, 4.99	1.24	.44	−1.94, 4.42
BMI	−0.06	.74	−0.42, 0.30	0.01	.92	−0.27, 0.30	0.17	.21	−0.10, 0.45
Anxiety	−0.36	.33	−1.10, 0.38	0.07	.82	−0.52, 0.65	−0.07	.79	−0.63, 0.48
Depression	0.20	.52	−0.41, 0.80	0.05	.85	−0.44, 0.53	0.05	.84	−0.41, 0.51
Early Life Trauma	0.20	.074	−0.02, 0.43	0.03	.76	−0.15, 0.21	0.16	.067	−0.01, 0.33
Step 2
Suppression	1.92*	.003	0.67, 3.18	1.55*	.004	0.51, 2.58	1.26*	.016	0.24, 2.29
Reappraisal	−2.85*	< .001	−4.08, −1.63	−1.96*	< .001	−2.97, −0.95	−1.55*	.003	−2.55, −0.55

Note. N = 82.

^*Denotes statistically significant

Emotion Regulation and Psychological Stress

In the unadjusted model, suppression and reappraisal significantly explained 14.3% of the variance in psychological stress, F(2, 80) = 7.84, p < .001. That said, suppression was not a statistically significant predictor of psychological stress, B = 0.04, p = .51, 95 % CI [−0.08, 0.16]. In contrast, reappraisal was a statistically significant predictor, B = −0.22, p < .001, 95 % CI [−0.33, −0.11]. When adjusting for demographic covariates, suppression was still non-significant (p = .68), while reappraisal remained a statistically significant predictor, B = −0.21, p < .001 95 % CI [−0.33, −0.10], explaining 14.6% of the variance in psychological stress, F(5, 77) = 3.81, p = .004.

Lastly, the additional inclusion of emotional distress covariates in Step 1 did not significantly explain any variance in psychological stress (p = .37). The inclusion of suppression and reappraisal in Step 2 significantly accounted for 15.2% of variance in psychological stress, F(8, 74) = 2.84, p = .008. This was primarily driven by reappraisal, such that greater use of reappraisal was associated with lower reported psychological stress, B = −0.24, p < .001, 95 % CI [−0.37, −0.12]. Suppression was still not a significant predictor of reported psychological stress (see Supplementary Digital Content, Table S2).

Mediation Analyses

Psychological stress significantly mediated the relationship between reappraisal and SBP, B = −0.65, 95 % CI [−1.37, −0.09], as well as reappraisal and DBP, B = −0.75, 95 % CI [−1.36, −0.23], but not reappraisal and PR, B = −0.38, 95 % CI [−1.03, 0.08]. In other words, individuals who are high in reappraisal are more likely to demonstrate lower blood pressure activity through lower perceptions of psychological stress. Psychological stress was not found to mediate the relationships between suppression and SBP, DBP, or PR (see Supplemental Digital Content, Figures S1 and S2).

DISCUSSION

The present study examined whether emotion regulation strategies were associated with ambulatory cardiovascular activity and psychological stress in an AI sample. Higher use of expressive suppression was associated with higher ambulatory cardiovascular activity. In contrast, higher use of cognitive reappraisal was associated with lower ambulatory cardiovascular activity and lower psychological stress. These associations were independent of age, sex, BMI, anxiety, depression, and early life trauma. Psychological stress was also found to mediate the negative association between reappraisal and blood pressure, such that higher reappraisal was associated with lower psychological stress, which in turn, was associated with lower ambulatory blood pressure.

These results align with previous emotion regulation research, such that expressive suppression is identified as detrimental for health, whereas cognitive reappraisal is protective (1–6). These results also align with a recent study in AIs, which found similar relationships between emotion regulation and post-traumatic stress symptoms (20). The constant effort required to inhibit expression in the face of an intense emotional experience is physiologically taxing and may be a pathway through which suppression impacts later cardiovascular health (7, 8). Indeed, experimental studies have identified associations between suppression and increased physiological responses to laboratory stress or emotion-induction tasks (1, 5, 37–40). These findings are particularly important within the context of AIs, given high rates of chronic conditions that implicate high blood pressure, such as CVD, obesity, and diabetes (17–19). Interestingly, suppression was not associated with psychological stress in the present study. Prior research has shown that while suppressors feel more negative emotions than nonsuppressors, they express less of it, resulting in no difference between suppressors and nonsuppressors in the reporting of negative emotions (5). Alternatively, while reappraisal is often found to decrease negative emotions and sympathetic nervous system responses (1, 6, 39), some research suggests that it may be less helpful in oppressed populations (16). This was not supported in the present study. In fact, the present effects for reappraisal were slightly larger than that of suppression, suggesting habitual reappraisal may be health protective in the current sample through its influence on decreased perceptions of psychological stress.

The current study is not without limitations. First, the analyses were correlational, and the results could have been impacted by a third variable (41). However, our analyses adjusted for several confounding variables. Second, this study was not designed to look at daily or situational fluctuations in emotion regulation and cardiovascular activity, but rather to get a comprehensive measure of resting cardiovascular activity and to eliminate the “white coat effect” (42). Third, the present findings are unique to the AI community. Researchers should not attempt to generalize to other racial/ethnic groups. While anxiety and depression scores in the present sample are higher compared to normative scores (43), this is in line with previous work showing higher levels of anxiety and depression in this population (18, 44, 45). Fourth, physiological measurements only consisted of ambulatory SBP, DBP, and PR across a three-day period. While these measurements provided a reliable assessment of average cardiovascular activity in the everyday lives of AI adults, it would be useful in the future to have a more comprehensive profile of their physiological risk for CVD. Fifth, while the present study involved a community advisory board and community-based authorship, it is our hope that this line of work will lead to future funded projects that include training on data collection, data analyses, and dissemination of findings to further opportunities for students and other community members to become independent investigators. Lastly, this study only examined habitual use of emotion regulation strategies. Future research should consider also including state measures of emotion regulation to see if daily fluctuations in emotion regulation differentially associate with the outcome variables.

In conclusion, this is the first study to examine whether emotion regulation strategies related to ambulatory cardiovascular activity and psychological stress in an AI community with high incidence of CVD. Habitual use of expressive suppression (but not cognitive reappraisal) was associated with higher resting cardiovascular activity. This investigation highlights a need for targeted health interventions in the AI community that aim to shift community members away from using suppression as a coping strategy.

List of Acronyms:

CVD

cardiovascular disease

SBP

systolic blood pressure

DBP

diastolic blood pressure

PR

pulse rate

CBPR

community based participatory research

CAB

community advisory board

EMA

ecological momentary assessment

ERQ

Emotion Regulation Questionnaire

DASS-21

The Depression, Anxiety, and Stress Scale-21

HADS

Hospital Anxiety and Depression Scale

RFQ

Risky Families Questionnaire

BMI

body mass index