2022 Volume 21 Issue 2 Pages 52-63
This study aimed to investigate the impact of the coronavirus disease 2019 (COVID-19) pandemic on bystander reaction to out-of-hospital cardiac arrest (OHCA) and present a strategy to improve the quality of bystander cardiopulmonary resuscitation (CPR) in the COVID-19 era. This systematic review and meta-analysis of observational studies were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A meta-analysis using the Hunter–Schmidt method in 13 previous studies identified using our criteria revealed that the pandemic is likely to decrease the rate of bystander CPR (log odds ratio with 95% confidence interval: −0.11; −0.22, −0.01) with large heterogeneity (I2 = 75.8%). Meta-regression analysis including global area and publication year revealed that heterogeneity is significantly affected by the global region (P = 0.048) and is the largest in Europe. A meta-analysis of 10 studies reporting the incidence of public-access defibrillation (PAD, automated external defibrillator use by bystanders) revealed that the PAD incidence is likely to be lowered by the pandemic (−0.30 [−0.47, −0.13]). The medical control system in each prefecture in Japan should play a central role as a community-based life-saving system to maintain and improve the quality of bystander CPR. The medical control strategy in the COVID-19 era should include continuous quality improvements of dispatcher-assisted CPR instruction and the system to recruit well-trained and qualified citizens to the scene of OHCA.
High-quality prehospital resuscitation efforts, beginning with the early bystander recognition of cardiac arrest, are key determinants of better outcomes of out-of-hospital cardiac arrest (OHCA) [1,2]. Cardiopulmonary resuscitation (CPR) initiated by bystanders who detect cardiac arrest before emergency medical service (EMS) arrival increases the chance of OHCA survival. Improving the incidence and quality of bystander CPR is a key issue addressed by regional medical control systems for prehospital emergency care, which is achieved by continuous quality improvement for CPR education to citizens [3] and dispatcher-assisted CPR (DA-CPR) or telephone CPR instruction [4].
Early investigations published shortly after the pandemic caused by coronavirus disease 2019 (COVID-19), particularly in Europe, reported the potential impact of the pandemic on prehospital resuscitation efforts and EMS and bystander response to OHCA [5,6]. However, later investigations that were mostly published by the end of 2020 reported contradictory results regarding the effect of the pandemic bystander response to OHCA [7–12]. A recent web-based questionnaire survey on citizens' willingness to perform CPR showed a marked reduction [13]. Traditional CPR training courses offered to the public are currently subject to restrictions in many countries, including Japan [14].
In collaboration with three undergraduate students in our department, the first author (HI), with a 15-year career as a medical director of the EMS in Ishikawa Medical Control Council, conducted a systematic review and meta-analysis of observational studies on the impact of the COVID-19 pandemic on bystander reactions to OHCA. HI then presented strategies to augment the incidence and quality of bystander CPR in the COVID-19 era in collaboration with the last author (TW), who wis a paramedic in the Niigata City Fire Department.
We conducted this systematic review and meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [15]. The primary review question developed using the PICO (population, intervention or exposure, comparison, outcome) framework was as follows: “In patients with non-EMS-witnessed OHCA (OHCA witnessed before EMS arrival) (P), does the COVID-19 pandemic (E), compared to a non-pandemic period (C), influence the incidence of bystander CPR (O)?”.
Article search and selection
We systematically searched the PubMed, Google Scholar, BioMed Central, and Cochrane databases for observational studies. We also reviewed and selected relevant references from the articles identified using this search strategy. Both prospective and retrospective observational studies in peer-reviewed journals were screened for eligibility to compare the bystander CPR incidences between the COVID-19 pandemic and a non-pandemic period. We excluded 1) systematic reviews, editorials, theses, and non-peer-reviewed articles; 2) study protocols; and 3) articles that did not report data on bystander response to OHCA. We reviewed the final set of articles included in this review according to the risk of bias in non-randomized studies of interventions (ROBINS-I) [16]. The inclusion criteria were non-overlapping population (or database), standard EMS system with advanced life support (ALS) procedures, and even lengths of pre-pandemic and pandemic periods.
Re-calculation of the number of patients not receiving bystander CPR in non-EMS-witnessed cases
Some studies included in our meta-analysis excluded EMS-witnessed OHCA (cases in which cardiac arrest was witnessed by EMS only after their contact with patients) in the participant selection process. Other studies have reported the number of EMS-witnessed cases without excluding them. The incidence of bystander CPR is extremely low in EMS-witnessed cases because patients do not experience cardiac arrest before EMS arrival. Therefore, we re-calculated the number of patients not receiving bystander CPR in non-EMS-witnessed cases in studies reporting the number of EMS-witnessed cases.
Statistical analyses
The Hunter¬–Schmidt random-effect model [17] in StataMP-16 (StataCorp, Texas, USA) was used for the meta-analysis. Heterogeneity was quantified using τ2, I2, and H2 statistics. The I2 statistic estimates the proportion of variation between the effect sizes due to heterogeneity relative to the pure sampling variation. An I2 value of > 50 indicated substantial heterogeneity, while an H2 value of 1 indicated perfect homogeneity among the studies.
Study outcomes
The primary outcome was the incidence of bystander CPR, and the secondary outcome was the incidence of public access defibrillation (PAD).
Characteristics of studies
Our database search identified 232 non-duplicate articles. A total of 26 studies were included in this review after excluding irrelevant articles based on the title, abstract, and screening by text review. After further assessment by the ROBINS-I and application of our inclusion criteria, 13 articles [18–30] were finally included in the meta-analysis (Fig. 1). The main characteristics of these studies are summarized in Table 1. We assumed that all articles included in the meta-analysis had an overall serious risk of bias.
 |
Pandemic effects on bystander response to OHCA
Although the 13 studies showed large heterogeneity, the Hunter-Schmidt method detected a significant reduction in the bystander CPR rate during the pandemic period (log odds ratio with 95% confidence interval: −0.11 [−0.22, −0.01], Fig. 2) and a large heterogeneity (I2 =75.8%). Further analysis using meta-regression methods, including publication year and global region (Europe, North America, and Asia/Oceania), revealed that the global region is likely to be a major factor associated with heterogeneity (Table 2). The heterogeneity in Asia/Oceania (I2 = 47.2%) was smaller than that in Europe (I2 = 88.4%) and North America (I2 = 74.6%) (Supplemental Fig. 1).
 |
Similarly, the Hunter–Schmidt method applied to 10 studies (Fig. 3) reporting the number of patients with PAD (defibrillation by citizens with an automated external defibrillator) revealed a significantly reduced incidence of PAD compared to that of bystander CPR during the pandemic period (−0.30 [−0.47, −0.13]), with smaller heterogeneity (I2 = 73.6%).
Impact of the pandemic on other OHCA characteristics and outcomes
Previous studies have shown that the pandemic altered other OHCA characteristics and outcomes (Table 3). The total number of OHCA cases activating EMS [26, 31] and the proportion of unwitnessed [19,31] and at-home OHCA cases [19,20] increased, whereas the proportion of resuscitation attempts by EMS decreased during the pandemic [20,31]. The decreased proportion of shockable initial electrocardiogram rhythm [23,31] and prolonged EMS response time interval (delayed EMS arrival) [19,31], in addition to suppressed bystander response to OHCA shown by our meta-analysis, might cause worse outcomes of OHCA during the pandemic. These changes have been confirmed in several systematic reviews [5–7].
 |
Undetermined impact of the pandemic on factors associated with high-quality bystander CPR and better outcomes of OHCA
High-quality CPR performed earlier after proper recognition of cardiac arrest is a crucial component of effective basic life support (BLS) actions, leading to better OHCA outcomes [32] Currently, the effect of the pandemic on the quality of CPR remains unknown. Little is known about the effects of the pandemic on factors associated with the quality of CPR (Table 4).
 |
In many countries, BLS or CPR training has switched from conventional (instructor-led and classroom-based) to hybrid or online-only training, as the former carries a risk of airborne virus transmission [14]. In Japan, the number of conventional training courses held by fire departments has markedly decreased after the pandemic [33]. A close relationship has been reported between recent CPR training experience and the quality of CPR [34] or willingness to perform CPR [35]. Therefore, the proportion of citizens with recent training experiences has likely decreased after the pandemic. The pandemic may alter bystander age [34,36], sex [37,38], number [36,42], bystander-patient relationship [39], and socio-economic status [40] among countries and may explain the heterogeneity in the pandemic effects among countries.
Moreover, the pandemic may alter the location of arrest [39,41], time of delay [39,43], and season [44] of OHCA. A nationwide or regional “lockdown” or “stay home” order might alter not only the location of OHCA but also the bystander-patient relationship for daytime OHCA [39]. The presence of trained family members might improve the quality of bystander CPR for daytime OHCA at home but might result in reduced quality at public locations owing to the decreased number of rescuers involved in BLS. The effect of the pandemic on the quality of bystander CPR and the overall outcomes of OHCA might differ between winter and other seasons since the OHCA of older adults occurs more frequently in the winter [44].
DA-CPR increases bystander CPR frequency and improves OHCA outcomes [38,39,45,46]. The educational approach enhances the ability to accurately detect cardiac arrest and increases the rate of bystander CPR. The quality of DA-CPR might have decreased during the pandemic.as the pandemic disrupted the system of care even in countries without an increased incidence of OHCA [47].
However, data regarding the effect of the pandemic on the quality of bystander CPR or the factors associated with its quality are lacking. Longterm observation of bystander and EMS response to OHCA is required because the COVID pandemic has persisted for nearly 2 years. The longest observation period among previous studies was only 6 months [29]. More detailed changes in the factors associated with early, high-quality bystander CPR are required to develop a new plan for quality improvement.
Strategy for effective bystander CPR in the COVID-19 era (Fig. 4)
Assurance of the safety of the victim, bystanders, and EMS personnel is increasingly important in the COVID-19 era as emphasized by recent guidelines for BLS [32]. Many countries have modified the BLS guidelines in many countries after considering the following risks involved in CPR procedures (https://www.youtube.com/watch?v=UfllQqYs-4A): 1) possible aerosol generation by chest compressions, positive-pressure ventilation, and advanced airway management; and 2) the close distance between the providers and the victim. The importance of personal protective equipment use was reconfirmed. Trained bystanders protected by a face mask should initiate chest compressions as early as possible for presumed cardiac arrest without getting close to the victim's face, as the risk of infection is very low. Healthcare providers should always apply personal protective equipment because their chances of infection are relatively high. In addition, the use of a mechanical device is recommended for chest compressions and a video laryngoscope for tracheal intubation.
Assurance of the safety of the victim, bystanders, and EMS personnel is increasingly important in the COVID-19 era as emphasized by recent guidelines for BLS [32]. Many countries have modified the BLS guidelines in many countries after considering the following risks involved in CPR procedures (https://www.youtube.com/watch?v=UfllQqYs-4A): 1) possible aerosol generation by chest compressions, positive-pressure ventilation, and advanced airway management; and 2) the close distance between the providers and the victim. The importance of personal protective equipment use was reconfirmed. Trained bystanders protected by a face mask should initiate chest compressions as early as possible for presumed cardiac arrest without getting close to the victim's face, as the risk of infection is very low. Healthcare providers should always apply personal protective equipment because their chances of infection are relatively high. In addition, the use of a mechanical device is recommended for chest compressions and a video laryngoscope for tracheal intubation.
Calling emergency services does not necessarily mean delaying CPR because mobile phones have become ubiquitous for telecommunication [32,45]. A well-trained bystander who is initially a sole rescuer at the scene is permitted to initiate chest compressions before placing an emergency call while shouting for help and activating the speaker function of his mobile phone to place an emergency call (CPR-first action [48]), particularly in cases of presumed non-cardiac etiology and non-older adult patients. However, untrained bystanders who are unable to determine what to do should immediately place an emergency call (119 call) using a mobile phone (Call-first action) [45]. Emergency calls beside OHCA victims via mobile phones help dispatchers obtain more accurate and timely information. Dispatchers should aim to detect cardiac arrest through well-organized key questions to bystanders. Dispatchers should start DA-CPR (instruction on chest compressions) as early as possible when the victim is suspected to be in cardiac arrest. DA-CPR with encouragement and quality feedback to bystanders should be continued until the EMS ambulance arrives at the scene.
The proportion of citizens who are re-trained every 6 months or less is a potent and potential factor associated with the quality of bystander CPR in patients with OHCA. A recent study conducted by the first author and his collaborators showed a significantly lower rate of bystander CPR in family- and friend/colleague-witnessed OHCA during the impact phase (24 weeks) after the Great East Japan Earthquake with nuclear pollution than that during the control period of the pre- and post-disaster years [49]. However, this recent study showed that the voluntary CPR performance of well-trained bystanders who initiated CPR without DA-CPR was preserved during the impact phase. Furthermore, recruitment of these well-trained citizens to the scene of OHCA via an alert system using a smartphone reportedly augmented the incidence of CPR before EMS arrival at the scene and survival from OHCA [50]. Therefore, the strategy for the COVID-19 era should include a regional system to increase the number of well-trained citizens and recruit them to the scene.
Community-based life-saving systems [32] should play a central role in implementing new strategies in the COVID-19 era. A medical control system was established in all prefectures of Japan after the introduction of ALS by Japanese paramedics. Every medical control system or council should restart continuous quality improvement programs for COVID-19 and future pandemic eras.
Limitations
This systematic review has several limitations. The studies included in the meta-analysis had an overall serious risk of bias. Furthermore, data selection differed among the included studies. The length of the pandemic was short and did not cover the recent pandemic. The strategy presented in this review was based on the results of a study conducted in the Ishikawa Prefecture.
Conclusion
In conclusion, the results of this systematic review showed that the COVID-19 pandemic is likely to decrease the rate of bystander CPR and the incidence of PAD. The medical control system in each prefecture in Japan should play a central role as a community-based life-saving system to maintain and improve the quality of bystander CPR. The medical control strategy in the COVID-19 era should include continuous quality improvement in DA-CPR instruction and the system to recruit well-trained and qualified citizens to the scene of OHCA.
We thank editorial members of the Niigata Journal of Health and Welfare for providing us with the opportunity to submit this review article.
We would like to thank Editage (www.editage.com) for English language editing.
Not applicable.
