Slow-tempo piano music listening and short-term anxiety and burnout trajectories in a predominantly nursing sample of ICU healthcare workers: A randomized controlled trial.

Abstract

Background: Healthcare workers in intensive care units (ICUs) are chronically exposed to high-intensity, high-risk clinical environments, where anxiety and burnout are highly prevalent. Music listening is a low-burden, non-pharmacological supportive strategy with potential stress-reducing effects; however, randomized trials evaluating standardized slow-tempo piano music interventions among ICU healthcare workers remain limited.

Objective: To compare short-term trajectories in self-reported generalized anxiety symptoms and weighted MBI-GS burnout index values between ICU healthcare workers assigned to a slow-tempo piano music intervention and those assigned to a passive control condition.

Methods: This single-center, open-label, two-group parallel randomized controlled trial was conducted in the ICUs of a tertiary hospital in China. A total of 126 ICU healthcare workers were enrolled and randomly assigned in a 1:1 ratio to the piano music group or the passive control group, with 63 participants in each group. The piano music group received a 2-week standardized slow-tempo piano music listening intervention consisting of four 30-minute sessions per week, for a total of eight sessions. The listening sessions were conducted outside ICU working hours, after the end of the participant's shift and before bedtime. The passive control group maintained their usual routines and received neither the study-specified music intervention nor a study-arranged 30-minute rest period. Anxiety and burnout were assessed after the 1st, 4th, and 8th sessions using the Generalized Anxiety Disorder-7 scale and the Maslach Burnout Inventory-General Survey. No outcome assessment was conducted before the first intervention session; therefore, the first assessment (T1) was a post-session assessment rather than a pre-intervention baseline. The study was retrospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500097216).

Results: All 126 randomized participants completed the three assessments. In the full sample, statistically significant group × time interactions were identified for GAD-7 scores (Wald χ²(2) = 46.29, p < .001) and the weighted MBI-GS burnout index (Wald χ²(2) = 6.78, p = .034). In the T1-adjusted analyses, the passive control group had higher GAD-7 scores than the piano music group at T2 (adjusted mean difference = 1.47, 95% CI [0.69, 2.25], p < .001), whereas the adjusted difference at T3 was not statistically significant (adjusted mean difference = 0.85, 95% CI [-0.03, 1.72], p = .057). The passive control group also had higher adjusted weighted MBI-GS burnout index values at T2 (adjusted mean difference = 0.52, 95% CI [0.24, 0.80], p < .001) and T3 (adjusted mean difference = 0.97, 95% CI [0.71, 1.24], p < .001). In the nurse-only sensitivity analysis, the group × time interaction remained statistically significant for GAD-7 scores but not for the weighted MBI-GS burnout index.

Conclusion: The piano music and passive control groups showed different short-term trajectories in self-reported anxiety symptoms and the weighted burnout index. However, because the study lacked a genuine pre-intervention assessment and used a passive control condition, the findings should be regarded as preliminary associations in short-term outcome trajectories, not as evidence of intervention efficacy or a specific effect of piano music. Further trials incorporating a pre-intervention baseline, active comparison conditions, and longer follow-up are warranted.