Dynamic bidirectional relationships between perceived stress and emotion regulation in emergency medical service clinicians

Abstract

Emergency medical services (EMS) clinicians are first responders who experience recurrent occupational stressors. Cross-sectional research suggests that higher self-regulation of emotions may be related to lower stress, especially in individuals with regular substance use. However, temporal dynamics are unclear. Our objective was to identify real-time dynamics between perceived stress and emotion regulation in EMS clinicians who regularly use substances. Participants were full-time EMS clinicians reporting alcohol and/or cannabis use ≥2x/week. Participants completed five daily ecological momentary assessments (EMAs) at semi-random times for 28 days. We used a continuous-time structural equation model with Bayesian estimation to identify dynamics between perceived stress and emotion regulation (both within-person centered and standardized). The 110 participants completed 12,234 EMAs (81.3% adherence). Higher perceived stress predicted lower future emotion regulation (standardized estimate = -0.68 [-1.05, -0.31]). Inversely, higher emotion regulation predicted lower future perceived stress (standardized estimate = -2.25 [-3.38, -1.15]). We identified bidirectional relationships between perceived stress and emotion regulation in the daily lives of EMS clinicians with regular substance use. While results may not be generalizable to EMS clinicians who do not regularly use substances, we identified emotion regulation as a future interventional target to reduce real-time stress in this highest-risk group.

Fig. 1. Participant recruitment and enrollment flow diagram.

Study participants were screened for eligibility based on eight different inclusion and exclusion criteria. Participants then had to complete more than 1 day of data to be included in the final analytical sample.

Fig. 2. Trajectories of perceived stress and emotion regulation within-person over time.

Observed data points of perceived stress and emotion regulation for a random sample of five individual subjects over 28 days. Emotion regulation levels were high and more stable, while perceived stress levels were moderate with more variation over the entire duration of the study.

Fig. 3. Continuous time auto-regressive and cross-regressive standardized effects of perceived stress and emotion regulation in emergency medical services clinicians.

Standardized auto-regressive effects of perceived stress and emotion regulation (on themselves) and cross-regressive effects of perceived stress and emotion regulation (on each other) from a Bayesian continuous time model. Data are plotted as a posterior median (solid line) with a 95% credibility interval (dotted lines). Time intervals are in hours; auto-regression of emotion regulation (red); auto-regression of perceived stress (purple); cross-regression of emotion regulation on perceived stress (green); cross-regression of perceived stress on emotion regulation (blue).

Fig. 4. Post hoc moderation of the dynamic bidirectional relationships between perceived stress and emotion regulation over time by baseline alcohol and cannabis use severity scores.

The alcohol use sample (blue) included 102 participants who endorsed drinking alcohol and the cannabis use sample (red) included 64 participants who endorsed cannabis use. The null relationships between perceived stress on emotion regulation (A) and emotion regulation on perceived stress (B) are visualized in the alcohol use sample. The null relationships between perceived stress on emotion regulation (C) and emotion regulation on perceived stress (D) are visualized in the cannabis use sample. The solid lines represent the estimated regression lines and the gray shaded regions represent the 95% CI.