(PS12-7) Inflammatory Implications of Anger Expression: Lower IL-6 and CRP Associated with Anger Suppression in the MIDUS Data

Background:   Chronic inflammation is implicated in the etiology of heart disease and other serious medical conditions. Inflammatory markers, such as interleukin-6 (IL-6) and C-reactive protein (CRP) are recognized as both predictors of chronic illnesses as well as “scar markers” of previous trauma, and are frequently included in measures of allostatic load. The link between inflammatory markers IL-6 and CRP is well established in the literature concerning the relationship between mental illnesses, such as depression, and allostatic load (Valkanova et al., 2013; Honkalampi et al., 2021); however, less research has focused on the link between anger (and anger expression) and pro-inflammatory markers. The emotion of anger is associated with physiological arousal (Williams et al., 2001), which suggests that a disposition towards anger may be associated with allostatic load and cardiovascular disease, but the evidence for this is mixed (Schulman & Stromberg, 2007; Chida & Steptoe, 2009). Few studies have examined how styles of coping with anger might influence inflammatory processes, but those that have found evidence that different styles are associated with risk of future cardiovascular disease and with higher levels of inflammation (Davidson & Mostofsky, 2010; Boylan & Ryff, 2013; Gross et al., 2014). In this study we examined the relation between blood serum levels of inflammatory markers IL-6 and CRP and styles of anger expression, specifically anger suppression (anger-in) versus expression (anger-out).


Methods: Using data from the MIDUS 2 biomarkers project, this study included 1,255 adult participants who completed measures of anger expression by self-report and provided blood samples for IL-6 and CRP analyses (Ryff et al., 2009). The State-Trait Anger Expression Inventory (Spielberger, 1999) was used to assess anger expression, with subscales measuring Anger-In (AI) and Anger-Out (AO) tendencies. We used the MIDUS Refresher project as a second sample to replicate our findings (n = 863).
Results &

Conclusions: Multivariate regression analyses revealed a significant negative relation in the MIDUS 2 data between AI and IL-6 (β = -.06, p = .04) and AI and CRP (β = -.07, p = .03). We also found a significant positive relation between AO and IL-6 (β = .06, p = .04), as well as AO and CRP (β = .06, p = .03). The relations between AI and IL-6 as well as AI and CRP were replicated in MIDUS Refresher (β = -.12, p = .001; β = -.08, p = .04), while the associations between AO and inflammatory markers did not replicate. We also found significant positive interactions in MIDUS 2 between trait anger and both AI and AO in predicting IL-6 (β = .34, p = .04; β = .60, p < .001), but not in MIDUS Refresher. Interestingly, the interaction suggests that AI has a negative relationship with IL-6 when trait anger is low, but when trait anger is high, AI has a positive relationship with IL-6. This suggests that suppressing anger might be healthy when anger is infrequent, but unhealthy when anger is frequent or intense.