599 - Heart rate variability and subjective self-report as measures of stress in simulated neonatal resuscitation

Poster Number: 599
Publication Number: 599.118

Helen Hu, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Tina A. Leone, Columbia University, New York, NY, United States; Kathleen G. Brennan, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Diana Vargas, Columbia University, New York, NY, United States; Rakesh Sahni, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States

Background: Stress can impair performance and learning in simulation studies. Heart rate variability (HRV) refers to the beat-to-beat difference between consecutive heart beats and is a physiologic marker of stress. Increasingly stressful simulations correlate with changes in HRV and with decreases in performance. There is limited literature on how physiologic measures of stress such as HRV relate to self-reported stress in neonatal resuscitation. 

Objective: The objective of this study was to describe the stress response during simulated neonatal resuscitation via HRV and self-reported levels of stress.

Design/Methods: Participants were prospectively enrolled neonatology fellows and advanced practice providers who completed a complex simulated neonatal resuscitation while wearing a biometric device that continuously measured HRV data. Participants rated their perception of stress during the simulation along a 5-point Likert-type scale. Respondents were categorized as High Perceivers (HP; “Agree” or “Strongly Agree”) and Low Perceivers (LP; “Neutral”, “Disagree”, or “Strongly Disagree”) for each question. HRV analysis was completed to yield time-domain and frequency domain indices including HR, rate-to-rate interval (RR), root mean square of successive RR interval differences (RMSSD; reflecting sympathetic activity), and low to high frequency power HRV ratio (LF/HF ratio; reflecting sympathetic activity). Data were analyzed in 1-minute intervals, and time intervals with the lowest mean HR (Trough Minute; Tmin) and highest mean HR (Peak Minute; Pmin) were identified as measures of extremes of physiologic response. Average HRV indices were compared between HP and LP responses.

Results: This study includes preliminary data on 18 participants from an ongoing randomized controlled trial. Most participants (63%) responded “Agree” or “Strongly agree” when asked if they felt nervous or stressed during the simulation but most also agreed to questions on confidence (Figure 1). Minute analysis of HRV indices (Figure 2) showed a significant difference in mean HR, mean RR, and mean RMSSD between Pmin and Tmin demonstrating increased stress response during Pmin (Table 1). In response to statements of stress, HP’s and LP’s had differences in HR and RR during Pmin (Table 1).Conclusion(s): Stress response during simulated neonatal resuscitation was demonstrated by both self-report and physiologic measures. Participants who agreed more strongly to feeling stressed during the simulation had higher HR and decreased HRV. Higher perception of confidence was also related to a greater stress response.
Figure 1: Self-reported survey responses and proportion of participants per response along 5-point Likert-type scaleOn self-reported measures of stress, a majority of participants (63%) responded “Agree” or “Strongly agree” when asked if they felt nervous or stressed during the simulation, however a majority of participants disagreed with subsequent statements assessing perceived stress. The majority of participants agreed with statements regarding feeling confident.
Figure 2: Mean heart rate (HR) and mean rate-to-rate (RR) interval per minute of simulation for all participants