Highlights
- •Women had a better chance of surviving an IHCA compared to men.
- •No signs of treatment disparities or discrimination related to sex were identified.
- •Using a comprehensive set of control variables is important.
Abstract
Background
Studies investigating sex disparities related to treatment and outcome of in-hospital cardiac arrest (IHCA) have produced divergent findings and have typically been unable to adjust for outstanding confounding variables.
Objectives
The aim was to examine sex differences in treatment and survival following IHCA, using a comprehensive set of control variables including e.g., age, comorbidity, and patient-level socioeconomic status.
Methods
This retrospective study was based on data from the Swedish Register of Cardiopulmonary Resuscitation and Statistics Sweden. In the primary analyses, logistic regression models and ordinary least square regressions were estimated.
Results
The study included 24,217 patients and the majority (70.4%) were men. In the unadjusted analyses, women had a lower chance of survival after cardiopulmonary resuscitation (CPR) attempt, at hospital discharge (with good neurological function) and at 30 days (p<0.01). In the adjusted regression models, female sex was associated with a higher chance of survival after the CPR attempt (B = 1.09, p<0.01) and at 30-days (B = 1.09, p<0.05). In contrast, there was no significant association between sex and survival to discharge with good neurological outcome. Except for treatment duration (B=-0.07, p<0.01), no significant associations between sex and treatment were identified.
Conclusions
No signs of treatment disparities or discrimination related to sex were identified. However, women had a better chance of surviving IHCA compared to men. The finding that women went from having a survival disadvantage (unadjusted analysis) to a survival advantage (adjusted analysis) attests to the importance of including a comprehensive set of control variables, when examining sex differences.
Keywords
List of abbreviations:
CA (Cardiac arrest), CPC (Cerebral performance category), CPR (Cardiopulmonary resuscitation), DNAR (Do-not-attempt-resuscitation), IHCA (In-hospital cardiac arrest), OHCA (Out-of-hospital cardiac arrest), PCI (Percutaneous coronary intervention), SES (Socioeconomic status), SRCR (The Swedish register of cardiopulmonary resuscitation)Introduction
Sociodemographic group disparities have attracted increasing attention in numerous societal domains, health care being no exception. Regarding sudden cardiac arrest (CA), research has clearly shown that racial minorities and individuals with lower socioeconomic status (SES), such as uneducated and poor people, receive inferior treatment and are less likely to survive a sudden cardiac arrest both outside (of-hospital cardiac arrest=OHCA) and inside (in-hospital cardiac arrest=IHCA) the hospital.
1
, 2
, 3
, - Agerstrom J.
- Carlsson M.
- Bremer A.
- et al.
Treatment and survival following in-hospital cardiac arrest: does patient ethnicity matter?.
Eur J Cardiovasc Nurs. 2021; https://doi.org/10.1093/eurjcn/zvab079
4
, 5
, - Stankovic N.
- Hoybye M.
- Lind P.C.
- Holmberg M.
- Andersen L.W.
Socioeconomic status and in-hospital cardiac arrest: a systematic review.
Resusc Plus. 2020; 3100016https://doi.org/10.1016/j.resplu.2020.100016
6
, 7
Regarding sex disparities, individual studies have produced rather divergent findings,8
, 9
, 10
although recent meta-analyses11
,12
suggest that men show a survival advantage after OHCA compared to women, particularly when it comes to survival to discharge and to 30 days. Various empirically supported explanations have been provided for this disparity. The most robust findings seem to be that women are more likely to be older10
,13
,14
and to suffer an unwitnessed OHCA in their homes rather than in a public location.10
,13
Moreover, women are less likely to present with a shockable initial heart rhythm (ventricular fibrillation/tachycardia), and to receive advanced post-arrest therapeutic measures, such as percutaneous coronary intervention (PCI).10
,13
,14
Much less research has been conducted on sex differences in conjunction with IHCA. A recent meta-analysis based on the results from only seven studies finds a trend (P = 0.052) towards higher mortality rates for women.
11
More research on sex differences following IHCA is clearly needed, particularly when it comes to treatment since extant research has focused more on survival. Probing for sex differences in treatment is important as they could potentially reflect treatment bias. Presumably, female patients could receive delayed treatment because medical staff may be quicker at recognizing CAs when the patient is male due to prevalent cognitive associations whereby cardiac problems are more strongly associated with men than women.15
, - Jerkeman M.
- Sultanian P.
- Lundgren P.
- et al.
Trends in survival after cardiac arrest: a Swedish nationwide study over 30 years.
Eur Heart J. 2022; https://doi.org/10.1093/eurheartj/ehac414
16
, - Cader F.A.
- Banerjee S.
- Gulati M.
Sex Differences in Acute Coronary Syndromes: a Global Perspective.
J Cardiovasc Dev Dis. 2022; 9https://doi.org/10.3390/jcdd9080239
17
Additionally, it should be noted that none of the aforementioned meta-analyses- Daugherty S.L.
- Blair I.V.
- Havranek E.P.
- et al.
Implicit Gender Bias and the Use of Cardiovascular Tests Among Cardiologists.
J Am Heart Assoc. 2017; 6https://doi.org/10.1161/JAHA.117.006872
11
,12
on sex differences in survival have adjusted for potential confounders such as comorbidity and cause of the CA, which is often symptomatic of individual studies too. Moreover, sociodemographic factors that correlate with sex, e.g., lower income among women, have seldom been accounted for.13
For example, we are not aware of any study that have adjusted for patient SES despite research showing that lower SES predicts lower survival after OHCA1
and IHCA,2
including less in-hospital prophylactic treatment (heart rhythm monitoring) and greater CPR delay immediately after the onset of the CA.2
There is a need for studies investigating associations between sex and CA outcome, adjusting for a more comprehensive set of potential confounders, including SES and ethnicity.The aim of the current research was to examine sex differences in treatment and survival following IHCA. In doing so, it examines how sex combine with age to influence treatment and survival, while also controlling for numerous clinical and contextual factors. The reason for studying potential age-dependent sex differences is that some OHCA studies suggest that increasing age has a more detrimental impact on survival for women than for men.
1
,14
Whether this is also the case for IHCA is not known. Possibly, age may be less important for sex differences in relation to IHCA because some factors that put women at a disadvantage (e.g., greater likelihood of having an unwitnessed arrest at home) are removed from the equation. Essentially, then, IHCA constitutes a more conservative test of sex differences than OHCA.Methods
Design
This retrospective register study is based on data from the Swedish Register of Cardiopulmonary Resuscitation (SRCR) and Statistics Sweden. It was approved by the Regional Ethical Review Board in Linköping, Sweden (No. 2017/293–31).
The Swedish Register of Cardiopulmonary Resuscitation (SRCR)
The SRCR is a national quality registry aimed at describing and improving resuscitation practices in Sweden, using a predefined, Utstein-style reporting framework. Data from the CA event are most often reported by the nurse in charge of the patient´s care. Follow-up data are typically reported by a resuscitation coordinator at each hospital. All registrations are performed using a registry manual and prompts within the digital protocol. In order to validate data, random inspections are performed by the Registry Committee.
The SRCR includes all CAs in Sweden where resuscitative treatments (e.g., chest compression and/or defibrillation) have been initiated regardless of place (IHCA or OHCA). The current study is based on data from the “IHCA-part” of the registry. As of 2019, all (100%) 74 Swedish emergency hospitals report data to the registry, including sociodemographic variables (e.g., sex and age), comorbidity, variables related to CA and treatment (e.g., place of CA, monitoring, CPR delay, defibrillation, PCI), and outcome (e.g., survival and neurological function). In addition, the treatment satisfaction among hospital personnel involved in providing resuscitative efforts are included.
Statistics Sweden
Patient-level data on SES were obtained from the Statistics Sweden's longitudinal integrated database for health insurance and labor market studies (LISA). LISA holds annual registers from 1990 and includes all individuals being 16 years of age and older in Sweden. The database integrates existing data from the labor market with the educational and social sectors. SES captures the social standing or class of an individual or social group.
18
In the current study, education and income were used as SES proxies. In terms of international comparisons, Sweden is considered to be at the forefront of equality.19
However, in recent decades OECD data reveal that inequality in, for example, income has increased at a faster rate compared to other countries.Oxfam. Development Finance International and Oxfam Research Report: The Commitment to Reducing Inequality Index [Internet]. Oxford: Oxfam GB; 2017. https://d1tn3vj7xz9fdh.cloudfront.net/s3fs-public/file_attachments/rr-commitment-reduce-inequality-index-170717-en.pdf (14th of April 2020).
20
Hence, SES appears to be an important construct also in Sweden. Importantly, as noted above, SES matters for the current study context as previous Swedish research has shown that patient-level SES is clearly associated with treatment and survival following IHCA.OECD. divided we stand: why inequality keeps rising, income inequality data update. Sweden (Januari. 2015), 2015. https://www.oecd.org/sweden/OECD-Income-Inequality-Sweden.pdf (14th of April 2020).
2
Participants
All patients, ≥40 years suffering an IHCA from January 1, 2005 to August 20, 2018 (extraction date) were included in the present study. The age criterion was used for two main reasons: 1) most younger patients have not completed their education or reached their highest income level, implicating that our SES proxies are not applicable, and 2) the prevalence of CA is low in younger patients, and they might constitute a selective group with better unobserved health prior to their CA.
Study variables
Outcome variables: CPR delay indicates the time elapsed from finding the patient to CPR is initiated (0=less than one minute, 1=one minute or longer); CPR duration (minutes); Survival after CPR (0=dead, 1=alive); Treatment satisfaction is reported by personnel involved in resuscitative treatment (0=unsatisfactory, 1=satisfactory)
1
; Survival to hospital discharge with good neurological function (0=Cerebral Performance Category (CPC)In the SRCR, treatment satisfaction is a binary measure (satisfied or not) based on an overall assessment of team performance according to international resuscitation guidelines and Swedish educational material. Data on treatment satisfaction should most often be based on consensus in the team and are reported by the nurse in charge of the patient´s care. All registrations are performed using a registry manual and prompts within the digital protocol.
21
3–5, indicating poor neurological function or death 1=CPC≤2, indicating good neurological function with no to moderate deficits) 2
; 30-day survival (0=dead, 1=alive).The full CPC categories are: CPC 1=Good cerebral performance. No or minor cerebral deficits. CPC 2=Moderate cerebral disability. Able to independently perform daily activities. CPC 3=Major cerebral disability. Unable to independently perform daily activities. CPC 4=Coma or vegetative state. CPC 5=Brain death/deceased. In this study we used CPC as a dichotomized variable in conjunction with survival at hospital discharge.
Predictor variables: Sex (0=male, 1=female), Age (0 = 40–65, 1=>65 years of age) and Age x Sex (interaction term for Sex and Age).
Control variables: Ethnicity (African, Asian, Eastern Europe, Middle Eastern, Nordic, South American, Southern Europe, Western Europe or “Other”); Education (0=high school or below, 1=college/university education). Income reflects the relative standing of the patient in the income distribution using a percentile score. Since a large proportion of the participants are retired, both annual earned income and retirement pension were used for this variable depending on whether the patient were retired or not. Hospital (which hospital); Year (year of CA); Comorbidity (previous history of heart failure, myocardial infarction, stroke, respiratory insufficiency, diabetes, cancer, and metastatic cancer); initial Heart rhythm (first documented rhythm during CA, i.e., ventricular fibrillation, ventricular tachycardia, pulseless electrical activity, or asystole); Heart rhythm monitoring (if the patient was monitored or not at the time of CA) and CA Etiology (e.g., myocardial infarction or respiratory insufficiency).
Statistical analysis
To investigate sex differences in CPR delay, Survival after CPR, Treatment satisfaction, Survival to discharge with good neurological outcome, and 30-day survival (0/1 variables), we estimated logistic regression models. To investigate sex differences in CPR duration (a continuous variable), ordinary least squares regressions were estimated.
All regressions included the control variables listed above as well as fixed effects (dummy variables) for hospital and year. The fixed effects were included to address the fact that hospital characteristics and time trends could be correlated with sex and at the same time affect outcomes, which would result in inconsistent estimates due to omitted variable bias. The regressions were estimated using the statistical software Stata 17 (StataCorp LLC, College Station, TX, USA).
Results
Patient characteristics
In total, 24,217 patients were included in the study. A majority were men (n = 14,930) and the mean (SD) age was 73.6 (11.6) years. Most of the patients (59.1%) received CPR in <1 min from CA, while the treatment duration in average was 16.2 min (SD=14.8). Half of the patients (51.6%) initially survived their CA. In 71.8% of the cases the personnel involved was satisfied with the treatment they provided. Almost one fourth of patients (23.1%) survived to hospital discharge (with good neurological outcome), while 29.4% survived to 30 days post CA. More information is presented in Table 1.
Table 1Descriptive statistics (unadjusted) for the full sample and by sex.
| All | Women | Men | ||
|---|---|---|---|---|
| (n = 24,217) | (n = 9287) | (n = 14,930) | ||
| Age, mean (SD) | 73.6 (11.6) | 75.3 (11.7) | 72.6 (11.4) | |
| SES education, n (%) | ||||
| High SES | 3760 (15.5) | 1227 (13.2) | 2533 (17.0) | |
| Low SES | 19,959 (82.4) | 7801 (84.0) | 12,158 (81.4) | |
| Missing | 498 (2.1) | 259 (2.8) | 239 (1.6) | |
| SES income, mean (SD) | 52.3 (24.6) | 39.2 (22.3) | 60.4 (22.3) | |
| Missing, n (%) | 5446 (22.5) | 2103 (22.6) | 3343 (22.4) | |
| Ethnic background, n (%) | ||||
| Nordic | 22,266 (91.9) | 8545 (92.0) | 13,721 (91.9) | |
| Africa | 110 (0.5) | 38 (0.41) | 72 (0.48) | |
| Asia | 146 (0.6) | 59 (0.64) | 87 0.58) | |
| Eastern Europe | 393 (1.6) | 164 (1.77) | 229 (1.53) | |
| Middle East | 437 (1.8) | 145 (1.56) | 292 (1.96) | |
| South Europe | 469 (1.9) | 175 (1.88) | 294 (1.97) | |
| Western Europe | 338 (1.4) | 140 (1.51) | 198 (1.33) | |
| Other | 58 (0.2) | 21 (0.23) | 37 (0.25) | |
| Comorbidity index (0–7), mean (SD) | 1.38 (1.19) | 1.27 (1.15) | 1.45 (1.21) | |
| Initial heart rhythm, n (%) | ||||
| Ventricular fibrillation | 3938 (16.3) | 1180 (12.7) | 2758 (18.5) | |
| Ventricular tachycardia | 1565 (6.5) | 487 (5.2) | 1078 (7.2) | |
| Pulseless electrical activity | 4785 (19.8) | 1880 (20.2) | 2905 (19.5) | |
| Asystole | 7788 (32.2) | 3255 (35.1) | 4533 (30.4) | |
| Missing | 6141 (25.4) | 2485 (26.8) | 3656 (24.5) | |
| Cardiac etiology, n (%) | ||||
| Yes | 11,514 (47.6) | 4090 (44.0) | 7424 (49.7) | |
| No | 2281 (9.4) | 975 (10.5) | 1306 (8.8) | |
| Missing | 10,422 (43.0) | 4222 (45.5) | 6200 (41.5) | |
| Monitored, n (%) | ||||
| Yes | 12,502 (51.6) | 4604 (49.6) | 7898 (52.9) | |
| No | 11,360 (46.9) | 4559 (49.1) | 6801 (45.6) | |
| Missing | 355 (1.5) | 124 (1.3) | 231 (1.6) | |
| CPR delay, n (%) | ||||
| Yes | 6118 (25.3) | 2405 (25.9) | 3713 (24.9) | |
| No | 14,303 (59.1) | 5460 (58.8) | 8843 (59.2) | |
| Missing | 3796 (15.7) | 1422 (15.3) | 2374 (15.9) | |
| CPR duration in minutes, mean (SD) | 16.2 (14.8) | 15.7 (14.3) | 16.5 (15.1) | |
| Missing, n (%) | 14,561 (60.1) | 5603 (60.3) | 8958 (60.0) | |
| Survival after CPR, n (%) | ||||
| Yes | 12,503 (51.6) | 4654 (50.11) | 7849 (52.57) | |
| No | 11,714 (48.4) | 4633 (49.89) | 7081 (47.43) | |
| Treatment satisfaction, n (%) | ||||
| Yes | 17,378 (71.8) | 6630 (71.4) | 10,748 (72.0) | |
| No | 6839 (28.2) | 2657 (28.6) | 4182 (28.0) | |
| Survival to discharge with good neurological outcome, n (%) | ||||
| Yes | 5597 (23.1) | 1892 (20.4) | 3658 (24.5) | |
| No | 16,839 (69.5) | 6893 (74.2) | 10,178 (68.2) | |
| Missing | 1781 (7.4) | 502 (5.4) | 1094 (7.3) | |
| 30-day survival, n (%) | ||||
| Yes | 7130 (29.4) | 2527 (27.2) | 4603 (30.8) | |
| No | 17,087 (70.6) | 6760 (72.8) | 10,327 (69.2) | |
Notes: CPR duration was longer than 90 min in 0.9% of the cases. These cases were recoded as missing, since such high numbers are unrealistic. The reason for the high fraction of missing values for CPR duration is mainly that this variable did not exist in the register until 2013.
Unadjusted analyses
Unadjusted analyses in the form of Spearman's rank correlation coefficients showed that women were older (p<0.01) and received a statistically significant shorter CPR duration compared to men (p<0.05). In addition, the personnel were less satisfied with their treatment of women (p<0.05). Women also had a lower chance of survival after CPR attempt, at hospital discharge (with good neurological function) and at 30 days (p<0.01).
Patients 65+ had a longer CPR delay (p<0.01) and a shorter CPR duration (p<0.05). Similar to women, the chance of survival for patients 65+ were lower regardless of timepoint (p<0.01). All bivariate correlations for outcome and predictor variables are presented in Table 2.
Table 2Bivariate correlations among the studied variables (Spearman's rank correlation coefficients). n = 9319–24,217.
| CPR delay (0/1) | CPR duration (ln) | Survival after CPR (0/1) | Treatment satisfaction (0/1) | Survival to discharge with good neurological outcome (0/1) | 30-day survival (0/1) | Female (0/1) | Age 65+ (0/1) | |
|---|---|---|---|---|---|---|---|---|
| CPR delay (0/1) | 1 | 0.0585** | −0.0542** | −0.0599** | −0.0789** | −0.0769** | 0.0107 | 0.0305** |
| 9319 | 20,331 | 16,590 | 18,967 | 20,421 | 20,421 | 20,421 | ||
| CPR duration (ln) | – | 1 | −0.5020** | −0.0446** | −0.4924** | −0.5122** | −0.0224* | 0.0879* |
| 9382 | 7625 | 8685 | 9444 | 9444 | 9444 | |||
| Survival after CPR (0/1) | – | – | 1 | 0.0879** | 0.5778** | 0.6191** | −0.0242** | −0.1092** |
| 18,969 | 22,304 | 24,039 | 24,039 | 24,039 | ||||
| Treatment satisfaction (0/1) | – | – | – | 1 | 0.0741** | 0.0857** | −0.0154* | −0.0071 |
| 17,630 | 19,057 | 19,057 | 19,057 | |||||
| Surv. w. good neur. outcome (0/1) | – | – | – | – | 1 | 0.9209** | −0.0437** | −0.1471** |
| 22,436 | 22,436 | 22,436 | ||||||
| 30-day survival (0/1) | – | – | – | – | – | 1 | −0.0386** | −0.1626** |
| 24,217 | 24,217 | |||||||
| Female (0/1) | – | – | – | – | – | – | 1 | 0.0621** |
| 24,217 | ||||||||
| Age 65+ (0/1) | – | – | – | – | – | – | – | 1 |
Notes: ** significant at the 1% level, * significant at the 5% level.
Primary analyses: adjusted analyses of associations between sex and outcome variables
The full results of the regression analyses, investigating associations between sex and outcome variables, are presented in Table 3. All regression models included fixed effects for hospital and year, and were adjusted for age, SES (income and education), ethnicity, comorbidity, initial heart rhythm, heart rhythm monitoring and etiology.
Table 3Association between sex and outcome variables (treatment and survival).
| CPR delay (0/1) odds ratios (se) | CPR duration (ln) B (se) | Survival after CPR (0/1) odds ratios (se) | Treatment satisfaction (0/1) odds ratios (se) | Survival to discharge with good neurological outcome (0/1) odds ratios (se) | 30-day survival (0/1) odds ratios (se) | |
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Female (0/1) | 0.9863 | −0.0709** | 1.0929** | 0.9241 | 1.0821 | 1.0913* |
| (0.0365) | (0.0227) | (0.0361) | (0.0560) | (0.0464) | (0.0420) | |
| N | 20,407 | 9444 | 24,030 | 18,953 | 22,155 | 24,030 |
Notes: Each cell/estimate is from a separate regression. All regressions used all observations of the full study population for which the dependent variable is non-missing. All regressions included fixed effects for hospital and year and controlled for age, SES (income and education), ethnicity, comorbidity, heart rhythm, etiology, and heart rhythm monitoring. ** significant at the 1% level, * significant at the 5% level.
Treatment outcomes
No statistically significant associations between sex and delay to CPR or treatment satisfaction were identified. However, being woman was significantly associated with a shorter duration of resuscitative efforts compared to men (B=−0.07, p<0.01).
Patient outcomes
Female sex was associated with a higher chance of survival after the CPR attempt (B = 1.09, p<0.01) and at 30-days post arrest (B = 1.09, p<0.05). In contrast, there was no significant association between sex and survival to discharge with good neurological outcome. In a sensitivity analysis, where age was included as a continuous variable, similar associations were displayed (Appendix, Table A1).
Interaction effects
In addition, to examine how sex combine with age to influence treatment and survival, interaction terms were included. In Table 4, the female dummy was replaced by two interaction variables: Female x Age 65+ and Female x Age less than 65 (the regressions also contain the 65+ dummy itself). Hence, the estimated coefficients show the female association with the outcomes separately for the two age groups. The difference in the sex coefficient between the two age groups is not significant for any of the outcomes (see the p-values in the table which are from F-tests of equal female coefficients). However, when age was included as a continuous variable, significant interaction effects for treatment duration and survival at 30 days emerged such that older women, relative to men, receive shorter CPR duration and are more likely to survive (Appendix, Table A2). Note that these findings are consistent with those in Table 4, where the coefficients show the same tendency although they are not significant in this different model specification. Taken together, we conclude that there is some suggestive evidence of an interaction between sex and age for these two outcomes.
Table 4Association between sex, age and outcome variables (treatment and survival).
| CPR delay (0/1) odds ratios (se) | CPR duration (ln) B (se) | Survival after CPR (0/1) odds ratios (se) | Treatment satisfaction (0/1) odds ratios (se) | Survival to discharge with good neurological outcome (0/1) odds ratios (se) | 30-day survival (0/1) odds ratios (se) | ||
|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | ||
| Female x Age 65+ (0/1) | 0.9685 | −0.0781** | 1.0830* | 0.9321 | 1.1040* | 1.1101* | |
| (0.0407) | (0.0248) | (0.0407) | (0.0640) | (0.0554) | (0.0500) | ||
| Female x Age less than 65 (0/1) | 1.0463 | −0.0453 | 1.1249 | 0.8992 | 1.0294 | 1.0453 | |
| (0.0770) | (0.0515) | (0.0734) | (0.1056) | (0.7926) | (0.0733) | ||
| p-value (F-test of equal female coeff.) | .355 | .563 | .609 | .787 | .437 | .462 | |
| N | 20,407 | 9444 | 24,030 | 18,953 | 22,155 | 24,030 |
Discussion
This nationwide population study, based on merged data from two separate quality registries, demonstrates significant sex differences in patient treatment and outcome when suffering IHCA. In the primary analyses, women were more likely to receive shorter treatment duration and had a higher likelihood of survival (immediately after resuscitation and at 30 days) compared to men. Whether sex differences related to survival exist have been debated for decades. Our results contradict the recently published meta-analyses on potential sex differences,
11
,12
indicating a survival advantage for men. This might be an effect of the comprehensive set of control variables in our regression models. In contrast to the previously mentioned meta-analyses, our “dual-registry design” enabled us control for additional individual-level sociodemographic factors i.e., SES and ethnicity. These factors are recently reported to be associated with both IHCA treatment and survival.2
,3
In addition, only a few IHCA registry studies have been able to control for comorbidity when examining sex differences. A previous IHCA study from SRCR, found a 10% survival advantage (at 30 days) for women after controlling for comorbidity, but not SES and ethnicity.- Agerstrom J.
- Carlsson M.
- Bremer A.
- et al.
Treatment and survival following in-hospital cardiac arrest: does patient ethnicity matter?.
Eur J Cardiovasc Nurs. 2021; https://doi.org/10.1093/eurjcn/zvab079
8
Another study from the US, by Teodorescu et al., examining sex differences in OHCA where the initial rhythm was ventricular fibrillation or pulseless electrical activity, found a survival advantage for women after controlling for a similar set of control variables like ours, including SES, race and comorbidity.10
The importance of using an extended set of control variables should also be seen in the light of the results in our unadjusted bivariate analyses, where being male correlated significantly with a higher chance of good outcome; survival after CA, survival at discharge with good neurological outcome, and survival at 30 days. The reasons of the survival advantage for women in our study are not clear. An American study by Topjian et al., suggested a protective effect of estrogen resulting in a better chance of survival merely among younger women.22
Consequently, examining potential interaction effects with age may be important. We find suggestive evidence of that older women, relative to men, receive shorter CPR duration and are more likely to survive 30 days after a CA. Research also including patients suffering IHCA and not receiving any treatment, e.g., those with a Do-Not-Attempt-Resuscitation (DNAR) decision, have reported a higher proportion of male patients receiving CPR.23
,24
If there are sex differences related to such decisions, women receiving CPR may constitute a more selective (healthier) group with a higher chance of survival. In our sample, women had a lower comorbidity index compared to men. Although, we have controlled for comorbidity, there may be other important aspects of health not included in our comorbidity index. The potential relation between sex and DNAR decisions should be further examined in order to identify any disparities.While no sex differences in CPR delay or treatment satisfaction were identified, women received shorter CPR duration compared to men. This is most likely not due to any treatment disparity. Since women are more likely to be successfully resuscitated, they may not need prolonged treatment to the same extent as men. This might explain the somewhat (mean difference = 0.8 min) shorter treatment duration.
Limitations
This study has some limitations. The SRCR merely includes those receiving CPR and/or defibrillation. Further studies including those not receiving treatment should be performed to be able to identify sex disparities related to the decision-making process prior to suffering an IHCA. In addition, compared with our other measurements (e.g., survival or CPR delay) the treatment satisfaction measure could be less reliable since the hospital staff may use and interpret the reporting instructions differently. Also, the retrospective design of does not allow us to draw any casual conclusions. To be able to gain more knowledge of potential sex differences and treatment disparities, additional study designs, e.g., qualitative or experimental, are warranted. For example, experimental studies that hold potential confounders constant are needed to pinpoint the mechanism(s) underlying sex differences. A rare example of such a study is Kramer et al.’s patient simulation study. It showed that participants were more reluctant to remove the clothing when giving CPR to simulated OHCA female patients, but also that the hand placement was better (more centered between the breasts) when the simulated patient was female.
25
This suggests that socio-cultural norms may be one factor that could contribute to sex differences in CA treatment.Conclusions
In this nationwide registry study, no signs of treatment disparities or discrimination related to sex were identified. However, women had a better chance of surviving IHCA compared to men. The finding that women went from having a survival disadvantage in the unadjusted analysis to a survival advantage in the adjusted analysis attests to the importance of including a comprehensive set of control variables, including SES and comorbidity, when examining sex differences in IHCA outcome.
Declaration of Competing Interest
None declared.
Acknowledgements
This research was supported by the Swedish Research Council for Health, Working Life and Welfare (Forte) Grant no. 2018–00256 to J.A.
Appendix
Tables A1 and A2
Table A1Association between sex and outcome variables (treatment and survival). Age as continuous variable.
| CPR delay (0/1) odds ratios (se) | CPR duration (ln) B (se) | Survival after CPR (0/1) odds ratios (se) | Treatment satisfaction (0/1) odds ratios (se) | Survival to discharge with good neurological outcome (0/1) odds ratios (se) | 30-day survival (0/1) odds ratios (se) | |
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Female (0/1) | 0.9874 | −0.0640** | 1.1069** | 0.9218 | 1.0993 | 1.1130* |
| (0.0366) | (0.0227) | (0.0363) | (0.0559) | (0.0471) | (0.0429) | |
| N | 20,407 | 9444 | 24,030 | 18,953 | 22,155 | 24,030 |
Table A2Association between sex, age and outcome variables (treatment and survival). Age as continuous variable.
| CPR delay (0/1) odds ratios (se) | CPR duration (ln) B (se) | Survival after CPR (0/1) odds ratios (se) | Treatment satisfaction (0/1) odds ratios (se) | Survival to discharge with good neurological outcome (0/1) odds ratios (se) | 30-day survival (0/1) odds ratios (se) | |
|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | |
| Female x Age | 0.9984 | −0.0052** | 0.9979 | 1.0056 | 1.0059 | 1.0071* |
| (0.0029) | (0.0019) | (0.0028) | (0.0048) | (0.0033) | (0.0030) | |
| Female | 1.0015 | −0.0185 | 1.1271** | 0.8792 | 1.0577 | 1.0612 |
| (0.0448) | (0.0300) | (0.0451) | (0.0636) | (0.0508) | 0.0463 | |
| Age | 1.0030 | −0.0013 | 0.9814** | 0.9989 | 0.9678** | 0.9630** |
| (0.0020) | (0.0013) | (0.0017) | 0.0034 | (0.0021) | (0.0020) | |
| N | 20,407 | 9444 | 24,030 | 18,953 | 22,155 | 24,030 |
Notes: The regressions are similar to those in Table 4. However, they enter age as a continuous variable instead of a dummy variable and include the Female x Age interaction instead of the two dummy interaction variables in Table 4. See the notes below Table 3. ** significant at the 1% level, * significant at the 5% level.
References
- Inequalities in income and education are associated with survival differences after out-of-hospital cardiac arrest: nationwide observational study.Circulation. 2021; 144: 1915-1925https://doi.org/10.1161/CIRCULATIONAHA.121.056012
- Discriminatory cardiac arrest care? Patients with low socioeconomic status receive delayed cardiopulmonary resuscitation and are less likely to survive an in-hospital cardiac arrest.Eur Heart J. 2021; 42: 861-869https://doi.org/10.1093/eurheartj/ehaa954
- Treatment and survival following in-hospital cardiac arrest: does patient ethnicity matter?.Eur J Cardiovasc Nurs. 2021; https://doi.org/10.1093/eurjcn/zvab079
- Kragholm K., et al. Socioeconomic differences in coronary procedures and survival after out-of-hospital cardiac arrest: a nationwide Danish study.Resuscitation. 2020; 153: 10-19https://doi.org/10.1016/j.resuscitation.2020.05.022
- Socioeconomic status and in-hospital cardiac arrest: a systematic review.Resusc Plus. 2020; 3100016https://doi.org/10.1016/j.resplu.2020.100016
- A disparity in outcomes of out-of-hospital cardiac arrest by community socioeconomic status: a ten-year observational study.Resuscitation. 2018; 126: 130-136https://doi.org/10.1016/j.resuscitation.2018.02.025
- Systematic review and meta-analysis of out-of-hospital cardiac arrest and race or ethnicity: black US populations fare worse.Eur J Prev Cardiol. 2014; 21: 619-638https://doi.org/10.1177/2047487312451815
- Characteristics and outcome among 14,933 adult cases of in-hospital cardiac arrest: a nationwide study with the emphasis on gender and age.Am J Emerg Med. 2017; 35: 1839-1844https://doi.org/10.1016/j.ajem.2017.06.012
- The effect of sex and age on return of spontaneous circulation and survival to hospital discharge in patients with out of hospital cardiac arrest: a retrospective analysis of a Canadian population.Resusc Plus. 2021; 5: 100084https://doi.org/10.1016/j.resplu.2021.100084
- Survival advantage from ventricular fibrillation and pulseless electrical activity in women compared to men: the Oregon Sudden Unexpected Death Study.J Interv Card Electrophysiol. 2012; 34: 219-225https://doi.org/10.1007/s10840-012-9669-2
- Association between sex and mortality in adults with in-hospital and out-of-hospital cardiac arrest: a systematic review and meta-analysis.Resuscitation. 2020; 155: 119-124https://doi.org/10.1016/j.resuscitation.2020.07.031
- Sex differences in survival after out-of-hospital cardiac arrest: a meta-analysis.Crit Care. 2020; 24: 613https://doi.org/10.1186/s13054-020-03331-5
- Women have lower chances than men to be resuscitated and survive out-of-hospital cardiac arrest.Eur Heart J. 2019; 40: 3824-3834https://doi.org/10.1093/eurheartj/ehz297
- Differential survival for men and women from out-of-hospital cardiac arrest varies by age: results from the OPALS study.Acad Emerg Med. 2014; 21: 1503-1511https://doi.org/10.1111/acem.12540
- Trends in survival after cardiac arrest: a Swedish nationwide study over 30 years.Eur Heart J. 2022; https://doi.org/10.1093/eurheartj/ehac414
- Sex Differences in Acute Coronary Syndromes: a Global Perspective.J Cardiovasc Dev Dis. 2022; 9https://doi.org/10.3390/jcdd9080239
- Implicit Gender Bias and the Use of Cardiovascular Tests Among Cardiologists.J Am Heart Assoc. 2017; 6https://doi.org/10.1161/JAHA.117.006872
- APA Task Force On Socioeconomic Status. Report of the APA Task Force On Socioeconomic Status.American Psychological Association, Washington, DC2007
Oxfam. Development Finance International and Oxfam Research Report: The Commitment to Reducing Inequality Index [Internet]. Oxford: Oxfam GB; 2017. https://d1tn3vj7xz9fdh.cloudfront.net/s3fs-public/file_attachments/rr-commitment-reduce-inequality-index-170717-en.pdf (14th of April 2020).
OECD. divided we stand: why inequality keeps rising, income inequality data update. Sweden (Januari. 2015), 2015. https://www.oecd.org/sweden/OECD-Income-Inequality-Sweden.pdf (14th of April 2020).
- Assessment of outcome after severe brain damage.Lancet. 1975; 1: 480-484https://doi.org/10.1016/S0140-6736(75)92830-5
- Women of child-bearing age have better inhospital cardiac arrest survival outcomes than do equal-aged men.Crit Care Med. 2010; 38: 1254-1260https://doi.org/10.1097/CCM.0b013e3181d8ca43
- Is there a difference between women and men in characteristics and outcome after in hospital cardiac arrest?.Resuscitation. 2001; 49: 15-23https://doi.org/10.1016/s0300-9572(00)00342-7
- Is there a difference in survival between men and women suffering in-hospital cardiac arrest?.Heart Lung. 2014; 43: 510-515https://doi.org/10.1016/j.hrtlng.2014.05.012
- Does the sex of a simulated patient affect CPR?.Resuscitation. 2015; 86: 82-87https://doi.org/10.1016/j.resuscitation.2014.10.016
Article info
Publication history
Published online: December 24, 2022
Accepted:
December 9,
2022
Received in revised form:
December 7,
2022
Received:
October 2,
2022
Identification
Copyright
© 2022 The Authors. Published by Elsevier Inc.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
