Sex Disparities in Resuscitation Quality Following Out of Hospital Cardiac Arrest

Background Women are known to be disadvantaged compared with men in the early links of the Chain of Survival, receiving fewer bystander interventions. We aimed to describe sex‐based disparities in emergency medical service resuscitation quality and processes of care for out‐of‐hospital cardiac arrest. Methods and Results We conducted a retrospective analysis of patients who were nontraumatic with out‐of‐hospital cardiac arrest aged ≥16 years where resuscitation was attempted between March 2019 and June 2023. We investigated 18 routinely captured performance metrics and performed adjusted logistic and quantile regression analyses to assess sex‐based differences in these metrics. During the study period, 10 161 patients with out‐of‐hospital cardiac arrest met the eligibility criteria, of whom 3216 (32%) were women. There were no clinically relevant sex‐based differences observed in regard to external cardiac compressions; however, women were 34% less likely to achieve a systolic blood pressure >100 mm Hg on arrival at the hospital (adjusted odds ratio [AOR], 0.66 [95% CI, 0.47–0.92]). Furthermore, women had a longer time to 12‐lead ECG acquisition after return of spontaneous circulation (median adjusted difference, 1.00 minute [95% CI, 0.38–1.62]) and 33% reduced odds of being transported to a 24‐hour percutaneous coronary intervention‐capable facility (AOR, 0.67 [95% CI, 0.49–0.91]). Resuscitation was also terminated sooner for women compared with men (median adjusted difference, −4.82 minutes [95% CI, −6.77 to −2.87]). Conclusions Although external cardiac compression quality did not vary by sex, significant sex‐based disparities were seen in emergency medical services processes of care following out‐of‐hospital cardiac arrest. Further investigation is required to elucidate the underlying causes of these differences and examine their influence on patient outcomes.

and are more likely to have active resuscitation care withdrawn prematurely. 9Exploring sex differences in the application of care in the prehospital setting is an important step toward improving patient-centered care outcomes at all stages of the Chain of Survival. 1,10mergency medical services (EMS) providers contribute to postresuscitation care through their immediate management before hospital handover, as well as through their decision whether to bypass local hospitals and transport to a tertiary cardiac center.][13] Previously, obtaining objective data about the quality of EMS resuscitation has been limited to retrospectively written patient care records; however, recent technological advances allow increased opportunity to use objective data collected in real time.Although previous studies have found EMS providers are less likely to begin resuscitation on women compared with men, 14 and women are less likely to receive intravenous/intraosseous access and subsequent adrenaline or antiarrhythmic agents, 15 there is no existing evidence related to sex-based differences in external cardiac compression (ECC) quality.The development of CPR feedback pads, which provide both live and recorded audiovisual feedback on the quality of ECC, allows insights into the quality of the fourth link in the Chain of Survival, early advanced life support.
The aim of this study was to examine sex-based disparities in EMS resuscitation quality and processes of care for OHCA.

METHODS Data Availability
The data that support the findings of this study are available upon reasonable request by emailing ziad.nehme@ambulance.vic.gov.au.

Study Design
A retrospective analysis of cases from the VACAR (Victorian Ambulance Cardiac Arrest Registry) was undertaken for all OHCA cases where resuscitation was attempted between March 2019 and June 2023.Patients were excluded if sex was not recorded, their age was <16 years, they had an advance care directive stipulating do not resuscitate, or the OHCA was precipitated by a traumatic event or was EMS witnessed.This project received ethical approval from Monash University Human Research Ethics Committee (project ID: 21046).Implied consent was gained from patients, with the institutional privacy policy available online. 16tting Ambulance Victoria is the single state-wide provider of EMS in the state of Victoria, Australia.The EMS provider services a population of 6.5 million people across 227 000 km 2 .Access to EMS is activated through a single nationwide telephone number (000).Suspected cardiac arrests receive telephone CPR instructions for bystanders 17 and a simultaneously activated 3-tiered response consisting of basic life support first responders (metropolitan Melbourne and regional centers), advanced life support paramedics, and intensive care paramedics. 18Ambulance Victoria has also integrated the alerting of smartphone-activated volunteer responders (GoodSAM) into their emergency dispatch system for eligible cases, which involves responding first aid-trained volunteers to OHCAs to provide

CLINICAL PERSPECTIVE
What Is New?What Are the Clinical Implications?
• Postresuscitation care should be improved for women, although it is not currently known how best to address this sex gap.• Ensuring that women receive an adequate duration of resuscitation before prehospital termination may increase the number of women who survive out-of-hospital cardiac arrest.

Nonstandard Abbreviations and Acronyms
ECC external cardiac compression OHCA out-of-hospital cardiac arrest VACAR Victorian Ambulance Cardiac Arrest Registry initial resuscitation while awaiting EMS, as described elsewhere. 19mbulance Victoria cardiac arrest treatment guidelines are outlined in publicly accessible clinical practice guidelines and are divided into medical and traumatic protocols. 20Paramedics are instructed to use a highperformance CPR approach for medical cardiac arrests, prioritizing rapid defibrillation and minimizing interruptions to ECC. 21Paramedics use feedback pads that record the chest compression depth, rate, recoil velocity, as well as the overall chest compression fraction, single longest pause, pre-and postshock pauses, 12-lead electrocardiography capture, and end-tidal CO 2 capnography.This recording is then uploaded to a central online database.
The EMS employs dedicated resuscitation coordinators who audit paramedic performance of all OHCAs where a resuscitation attempt was provided by EMS to ascertain resuscitation performance metrics using patient care records and the defibrillator recording.

Data Source
Data were sourced from the VACAR, which is a population-based clinical quality registry of all OHCA attended to by EMS in Victoria, as previously described. 22The investigators had full access to the VACAR.Sex is recorded by the treating paramedics, who have a drop-down selection of male, female, or other/unknown.Current data capture therefore does not adequately capture patients whose sex falls outside of the sex binary.
Performance data are divided into the 5 domains of early recognition, quality ECC, defibrillation, advanced interventions, and postresuscitation care, 23 aligning with the Chain of Survival.

Outcomes
A total of 18 metrics across the 5 performance data domains were measured, including a mix of binary, proportional, and time-based performance metrics, as outlined in Table 1.

Statistical Analysis
Statistical analyses were undertaken using Stata Statistical Software 18 (StataCorp, College Station, TX).Baseline characteristics are presented as frequencies and proportions for categorical data and medians and interquartile ranges for continuous data.
Categorical outcomes were assessed with multivariable logistic regression, and continuous variables using quantile regression.All models were adjusted for patient age, arrest cause (presumed cardiac versus other nontraumatic cause), and location of the OHCA (private residence, aged care facility, public location, or other including medical clinics, corrections facilities).Results of the logistic regression analyses are presented as adjusted odds ratio (AOR) and 95% CI.Results from quantile regression are presented as median adjusted difference and 95% CI.Variables with statistically significant differences are visually represented using kernel density plots comparing women and men.Sensitivity analyses were performed, which also included initial presenting rhythm, witnessed status, and presence of bystander CPR on paramedic arrival as confounders.
Missing data were handled using pairwise deletion.

Baseline Characteristics
We included 10 161 OHCA cases, the majority (68%) of which were men (Figure 1).The baseline characteristics for women were less favorable, as shown in Table 2. On average, women were older and were more likely to present in a nonshockable rhythm.
Women were also less likely to have an OHCA witnessed by a bystander, receive bystander CPR, or arrest in a public location.
Resuscitation performance metrics are shown in Table 3, and outcomes with significant sex-based differences are displayed graphically in Figure 2A through 2E.

Early Recognition
The median time from arrival at patient to placement of defibrillation pads was 1 minute for both women and men, with no sex-based difference following adjustment.Likewise, compressions occurred during pad placement for 92% of patients, with women having similar odds to men following adjustment (AOR, 0.99 [95% CI, 0.79-1.25]).Accurate rhythm recognition occurred in 93% of patients, with no difference between women and men (AOR, 1.04 [95% CI, 0.82-1.32]).

Advanced Interventions
The median time from paramedic arrival to insertion of a supraglottic airway device was 3 minutes overall, and first-pass intubation was achieved in 83.7% of the overall cohort.There were no sex-based differences for either outcome (time to supraglottic airway: median adjusted difference, 0.00 minutes [95% CI, −0.05 to 0.05]; first pass intubation success: AOR, 1.09 [95%

Postresuscitation Care
Postresuscitation care was inferior for women within all performance metrics.Women had reduced odds of achieving a systolic blood pressure >100 mm Hg on arrival at the hospital (87.2% versus 91.6%), which remained significant after adjustment (AOR, 0.66 [95% CI, 0.47-0.92]).The median time to 12-lead ECG after return of spontaneous circulation was 6 minutes for women compared with 5 minutes for men (median adjusted difference, 1.00 minute [95% CI, 0.38-1.62]).This translated to a higher proportion of women receiving a 12lead ECG after loading while en route to a hospital (11.5% versus 6.7%; AOR, 0.60 [95% CI, 0.40-0.89]).Similarly, women were less likely to be transported to a 24-hour percutaneous coronary intervention (PCI)capable facility than men (88.5% versus 92.3%), and this remained significant after adjustment (AOR, 0.67 [95% CI, 0.49-0.91]).Lastly, in patients who died on scene, resuscitation was terminated at a median time of 14 minutes for women compared with 24 minutes for men, and this difference was significant after adjustment (median adjusted difference, −4.82 minutes [95% CI, −6.77 to −2.87]).As shown in Figure 2E, there was a bimodal distribution of resuscitation duration for  nonsurvivors, with the major peak representing paramedic decision to withhold resuscitation in the initial stages of the resuscitation attempt and the minor peak representing ceased resuscitations following the recommended minimum duration as per the internal guideline.All results were consistent in our sensitivity analyses (Table S1), with the exception of resuscitation duration for patients who died at scene, in which the adjusted median difference remained significant but reduced to −1.05 (95% CI, −2.09 to 0.00).

DISCUSSION
Our analysis of resuscitation quality metrics among >10 000 patients with OHCA highlights several sexbased disparities in resuscitation care.Compared with men, women experienced longer delays in adrenaline administration and 12-lead ECG acquisition.Women were also less likely to receive perfusion management resulting in a systolic blood pressure >100 mm Hg on arrival to hospital or be transported to a 24-hour PCIcapable facility.In nonsurvivors, women also received shorter resuscitation attempts compared with men.The only metric in which women received superior care to men was postshock pauses.Perman et al investigated public perceptions of why women receive less bystander CPR compared with men following OHCA, and found a common misconception that women are weak and frail and therefore more prone to injury. 24Our study found no sex-based differences in the rate or depth of compressions, suggesting this misconception is likely not perpetuated by EMS providers.Despite this, recoil velocity in our study was slower for women compared with men (median difference, −9.13 mm/s [95% CI, −13.24 to −5.01]).This finding mimics previous research that suggests that, despite similar chest compression depth, structural differences in the elasticity of the female chest wall may be responsible for a slower recoil velocity compared with male patients. 25,26he majority of sex-based disparities in resuscitation quality and processes of care seen in our study relate to the postresuscitation care domain.8][29] Identifying ischemic changes on a 12-lead ECG informs the EMS' decision to transport to a PCI-capable facility, as well as activate code ST-segment-elevation myocardial infarction protocols that have been associated with reduced door-to-balloon time and mortality rates. 30Previous literature in our region reported similar rates of ST-segmentelevation myocardial infarction among men and women who presented postresuscitation for an initially shockable rhythm. 31Despite this, women in our study had a delay to 12-lead ECG acquisition compared with men, suggesting that investigating a cardiac cause of OHCA may be less prioritized for women.Although the median difference was only 1 minute longer for women compared with men, women in our study were more likely (11.5% versus 6.7%) to have their 12-lead ECG taken while en route to a hospital rather than on scene.The acquisition of a 12-lead ECG on scene allows paramedics to appropriately direct the patient to a PCI-capable facility and give adequate notification to the receiving hospital.This finding may have contributed to the sexbased disparity in PCI-facility transports.Furthermore, among patients with a presumed cardiac cause, women were 33% less likely to be transported to a 24-hour PCI-capable facility, even after adjustment for OHCA baseline characteristics.There is current contention on the importance of receiving postresuscitation care in a tertiary cardiac center.Although a recent systematic review suggested that receiving postresuscitation care at a tertiary cardiac facility improves patient outcomes, 13 a prospective, multicenter, randomized clinical trial in London, United Kingdom published shortly after showed no difference in outcomes. 32Locally in the Victorian region, direct transfer to a PCI-capable center was associated with a 40% increase in survival to hospital discharge. 33Inequitable access to tertiary cardiac care centers for men and women following return of spontaneous circulation as a result of EMS decision making may suggest an implicit bias from EMS that women are less likely to require tertiary cardiac care.
Women in our study were less likely to have a systolic blood pressure of at least 100 mm Hg on arrival at the hospital, although we were not able to establish how pharmacologically aggressive patients were managed by paramedics.Patients who remain severely hemodynamically compromised despite aggressive perfusion therapy may be more likely to be diverted to the closest hospital in lieu of further travel for a PCI-capable facility.Some goal-directed therapies within our system are only available after 12-lead ECG acquisition, such as anticoagulants and thrombolysis administration.Women in our study were less likely to have a 12-lead ECG acquired before the transport decision and thus had a reduced opportunity to receive goal-directed therapies that may have stabilized their hemodynamic status to allow for longer transport to a PCI-capable facility.These considerations are already stated in the Clinical Practice Guideline for return of spontaneous circulation management in Victoria 20 ; however, further education on the topic may be necessary to elicit cultural change.
Consistent with previous studies, women in our study were older and more likely to arrest in a residential aged care facility, while being less likely to arrest in public, have a witnessed OHCA, receive bystander CPR, or present in a shockable rhythm compared with men.For patients whose resuscitation was terminated by EMS on scene, women in our study received shorter resuscitation attempts compared with men.However, it is noteworthy that inclusion of the baseline characteristics in our multivariable model reduced the median difference in resuscitation duration to only 1 minute.

Limitations
In this study, we were reliant on the paramedics' assumption of the patient's sex as opposed to the patient's stated sex, which may differ.However, our study focused on paramedic behaviors in response to the patient sex, and the findings are therefore relevant to patients who paramedics perceive as women, even if they self-identify otherwise.Some variables are derived from the paramedic-written patient care records, and inaccuracies may arise from the retrospective recording of the time of paramedic interventions.However, it is expected that any bias would be random across both sexes.It is possible confounders beyond the Utstein variables, such as patient frailty and comorbidities, also influenced paramedic decision making, and these were not controlled for in our study.

CONCLUSIONS
Our study indicates that, although women received comparable initial resuscitation efforts such as time to pad placement, first defibrillation, and ECC quality to men, women received adrenaline later than men and shorter resuscitation attempts for nonsurvivors.Furthermore, 12-lead ECGs were obtained more slowly for women, women were less likely to be transported to a tertiary cardiac center, and upon arrival at the hospital they had poorer perfusion status.Considering these disparities, future research should prioritize investigating the underlying causes of these discrepancies, including potential biases in decision making during resuscitation attempts.Furthermore, EMS clinicians should receive targeted education to raise awareness and address implicit biases that may contribute to these disparities, ultimately working toward a more equitable and effective resuscitation care system.

Figure 1 .
Figure 1.Patient selection for OHCAs occurring between March 2019 and June 2023.CPR indicates cardiopulmonary resuscitation; EMS, emergency medical services; OHCA, out-ofhospital cardiac arrest; and ROSC, return of spontaneous circulation.

Figure 2 .
Figure 2. Kernel density estimation plots for unadjusted outcomes with significant sex-based differences, by sex, between March 2019 and June 2023.A, The mean manual release velocity during compressions.B, Time between arrival of fourth personnel at patient and administration of initial bolus of adrenaline.C, Mean length of time between defibrillation and hands off chest postdefibrillation.D, Time between sustained ROSC (≥10 minutes) and time of first 12-lead ECG after ROSC.E, Duration of ECC delivered by paramedics where resuscitation is terminated prehospitally.ECC indicates external cardiac compressions; and ROSC, return of spontaneous circulation.

Table 1 .
Resuscitation Performance Metric Definitions as Per the Victorian Ambulance Cardiac Arrest Registry

Table 2 .
Characteristics of Patients With OHCA With an Attempted Resuscitation Between March 2019 and June 2023 CPR indicates cardiopulmonary resuscitation; IQR, interquartile range; OHCA, out-of-hospital cardiac arrest; and ROSC, return of spontaneous circulation.

Table 3 .
Resuscitation Performance and Outcomes of Patients With Out-of-Hospital Cardiac Arrest With an Attempted Resuscitation Between March 2019 and June 2023