Impact of a stress coping strategy on perceived stress levels and performance during a simulated cardiopulmonary resuscitation: a randomized controlled trial

Early start of effective cardiopulmonary resuscitation (CPR) significantly improves survival of patients following cardiopulmonary arrest in and outside the hospital [1, 2]. While most studies in the past focused on technical aspects of CPR, recently the importance of non-technical factors such as teamwork, communication and leadership have been recognized [3‐8]. Several studies have demonstrated that CPR causes significant mental stress in rescuers [9‐15], and health care workers often feel unprepared to manage stress and conflicts in a cardiac arrest situation [16‐18]. This is important because the stress experienced in an emergency situation may impair performance. In line with that, we recently found that feeling stressed and overwhelmed while delivering CPR was associated with worse CPR performance [14]. Hence, stress reducing measures may improve performance in critical situations.

To reduce stress, the focus on attentional processes may be a promising venue. Stress can have two opposite effects on attention. First, stress narrows attention [19]. For tasks that are relatively easy, narrowing attention can lead to improved performance by supporting a focus on the task [20]. However, narrowing of attention entails the danger of not noticing potentially important information, a phenomenon known as “tunnel vision” [21]. Furthermore, it is also related to premature closure, which is characterized by making decisions based on insufficient consideration of information available [22]. The second mechanism refers to an impaired ability to suppress irrelevant information, increased distractibility, ultimately leading to misjudgements of priorities [23‐25]. A related mechanism refers to non-systematic scanning of informational cues [22]. However, information that distracts from task priorities may not only stem from external events but may also be generated internally, for instance by worrying about one’s own performance (intrusive thoughts) [26]. Such intrusive thoughts may be increased by noticing that one’s performance is not optimal [27], thus possibly leading to a vicious circle.

The response to acute stress is highly dependent on the individual’s perception of demands and resources [10, 28], and on stress reactivity [29]. Therefore, stress management training may reduce stress [30]. Indeed, stress management training has had positive effects not only on stress indicators but also on performance [31, 32]. However, although such procedures have been implemented in medical settings, they typically have not been evaluated in terms of medical performance [33].

Based on positive results achieved in other performance settings [15, 32, 34], and based on our recent finding that a brief leadership instruction improves CPR performance [5, 35] we hypothesized that a brief task-focusing strategy may reduce stress and improve CPR performance. The aim of this study was therefore to (1) describe the stress patterns experienced during a CPR situation; (2) investigate whether the perceived stress was associated with CPR performance in terms of hands-on time and time to start CPR; (3) to investigate whether this task focusing strategy reduces perceived stress levels, and (4) whether this translates into better CPR performance. Based on findings that clear, directive leadership can enhance performance in cardiac resuscitation [5, 8], we further (5) investigated if stress was associated with fewer leadership statements.

This study investigated the influence of a short task-focusing strategy on perceived stress levels and performance of rescuers in a simulated CPR scenario. We found an increase in stress/overload levels during the resuscitation period and an association of stress/overload with CPR performance. Students in the intervention group reported less perceived stress/overload during the resuscitation period, and they had about 10% more hands-on time; performance differences did not, however, reach statistical significance.

Previous research has demonstrated that mental stress impairs performance of rescuers in emergency situations [14], which was also validated within this analysis. This may be due to several causal pathways. First, mental stress has been shown to impair the attentional resources because the cognitive system is in danger of becoming overloaded. During stressful situations, participants may selectively focus their attention to selected tasks only, thereby neglecting other potentially relevant information. As stress increases, the ability to filter out irrelevant information may decrease, leading to increased distractibility [23‐25]. Second, studies found that stress impairs retrieval from memory; for example stress due to public speaking has been associated with impairments on tasks that required remembering previously learned information [40]; in our case, retrieval of existing knowledge about the treatment algorithm may have been impaired. Third, stress has also been shown to impair rational decision-making [10, 25]. Finally, stress has also been implicated in loss of team perspective and decreased team performance [31, 41]. Importantly, decreased performance due to stress may in turn further increase mental stress of rescuers leading to a vicious cycle.

Only few studies have evaluated the effectiveness of interventions to reduce chronic stress in medical practice. Effects of such interventions have included a reduction in perceived stress-levels for treatment groups [42, 43], increased assertiveness scores [44, 45], and increases in job satisfaction [46]. One study of behavioral training in general practitioners demonstrated a benefit in developing skills at coping with stress [30]. This training improved the general practitioner’s quality of work life and reduced their work-related psychological distress; yet, these were chronic stress situations and CPR related stress is an acute stress reaction. Similar stress coping strategies for acute emergency situations, such as CPR, are largely lacking. A recent German study investigated the effects of crew resource management (CRM) training including psychological teaching on the performance of intensive care professionals in a randomized-controlled trial [11]. The training did reduce stress, but no significant difference in the stress response or medical performance was noted in comparison to a group receiving traditional training; note that the CRM training was not specifically focusing on stress. Our intervention aimed at bringing the attention of rescuers to the important elements of the task and to task priorities by posing two task-focusing questions in case they felt overwhelmed by stress in a CPR situation. This intervention had a small but significant benefit in terms of perceived levels of feeling stress/overload; yet, no statistically significant improvement in performance was observed, although the intervention group did have 10% more hands-on time.

Although there is no definite explanation for the lack of association of our intervention with better performance, several explanations may be considered. First, this may be due to the relatively small sample size and lack of power. Although the intervention demonstrated a 10% relative increase in performance, this difference did not reach statistical significance; this was true overall and in different subgroups (Figure 3). Second, the intervention was very short, including only two questions the participants were expected to ask themselves. It is possible that the intensity of this intervention was not high enough to affect performance. This explanation is supported by the finding that stress levels, although decreasing overall during the resuscitation, did not significantly decrease during the most vulnerable and most stressful period, that is, when CPR was started; this was true in the intention-to-treat and the per protocol analysis. Thus, the intervention may not have been intense enough to influence stress levels to such a degree that stress-induced impairments of performance were successfully countered. Still, it has to be noted that the effect of the intervention on hands-on time was close to statistical significance (P = .059) in quartile of students that was most highly stressed. Furthermore, if the difference of 5.5 seconds in hands-on time between experimental and control group (and of 13.1 seconds in the most highly stressed quartile) can be confirmed in future studies, this would indicate a notable improvement in performance considering the low intensity of the intervention.

Interestingly, within this study we found that more leadership statements (such as commands, decisions what and how to do, task distribution among others) were associated with earlier start and longer duration of uninterrupted CPR performance. This validates previous observational research [8] and a randomized controlled trial that demonstrated a benefit from a brief leadership debriefing in terms of CPR performance [5, 35]. Within the present trial, the task-focusing strategy did not increase the number of leadership statements, which may partly explain the lack of improvement in CPR performance. Perhaps a combination of stress-related and leadership-related instructions would yield stronger results.

This study has a number of limitations. The small number of participants included in this study limited the power of our analyses and increased the risk for type II errors. Although previous studies showed that participants rated the simulated resuscitation in a high fidelity simulator as highly realistic [36, 37] and also perceived substantial stress [39], participants might still have perceived the simulated resuscitation as less stressful than a real life resuscitation. Participants were medical students and the results may not unconditionally be applied to more experienced intensive care physicians. Students were performing alone. Finally, only participants in the intervention group were made aware about the importance of stress during CPR and thus may have responded differently to the stress questionnaires (Hawthorne effect). Thus, measurement of performance is the preferred outcome measure and should be used in the future for similar research. As an alternative design, both groups could be made aware of stress but only one could receive a stress reduction intervention.

A brief stress-coping strategy moderately decreased perceived stress without, however, significantly affecting performance of rescuers in a simulated CPR scenario strongly enough to yield a statistically significant difference. Further studies into the effect of stress and stress reducing strategies are warranted; they should consider an intervention that is still short yet somewhat stronger, for instance, by including not only questions but also self-guiding statements [47] and possibly a combination with instructions regarding leadership [5, 35].