³Department of Clinical Research Design and Evaluation, Samsung Advanced Institute for Health Sciences and Technology (SAIHST), Sungkyunkwan University, Seoul, Korea

⁴Department of Emergency Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

⁵Department of Pulmonary and Critical Care Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

⁶Department of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

⁷Division of Allergy, Pulmonary and Critical Care Medicine, Department of Internal Medicine, Transplant Research Center, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea

⁸Department of Pulmonary, Allergy and Critical Care Medicine, Hallym University Sacred Heart Hospital, Anyang, Korea

⁹Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea

Corresponding author: Ryoung-Eun Ko Department of Critical Care Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Korea Tel: +82-2-3410-3876, Fax: +82-2-2148-7088 Email: koryoungeun@gmail.com

*These authors contributed equally to this work.

Received 2025 August 30; Revised 2025 October 6; Accepted 2025 October 14.

Abstract

Background

Emergency department (ED) overcrowding poses a global challenge, particularly for critically ill patients requiring intensive care unit (ICU) admission. Although delays in ICU transfer increase mortality in critically ill populations, the optimal timing for septic shock remains uncertain.

Methods

We conducted a target trial emulation using a prospective cohort of 815 septic shock patients from 19 Korean hospitals. Delayed ICU transfer was defined using restricted cubic splines. The primary outcome was in-hospital mortality. Multivariable logistic regression and inverse probability treatment weighting were used to adjust for confounders of age, sex, comorbidities, severity of illness, and mechanical ventilation use. Subgroup analyses were performed to assess the effect across patient characteristics.

Results

The median time of ED-to-ICU transfer was 6.7 hours (interquartile range, 4.7–11.4), and only 7% of patients were transferred within 3 hours. ICU transfer within 3 hours was associated with significantly lower in-hospital mortality (odds ratio, 0.48; 95% CI, 0.24–0.94) compared to later transfers. Mortality risk increased with elapsing time up to 6 hours and then plateaued. The benefit of early ICU transfer was consistent across subgroups but was particularly pronounced in patients requiring extracorporeal membrane oxygenation or continuous renal replacement therapy (P for interaction=0.02).

Conclusions

Early ICU transfer within 3 hours significantly reduces mortality in patients with septic shock, with the greatest benefit observed in those requiring advanced organ support. These findings highlight the need for system-wide strategies to reduce ED boarding time and prioritize timely ICU admission for septic shock management.

Keywords: emergency service; hospital mortality; intensive care units; patient transfer; septic shock

INTRODUCTION

Emergency department (ED) overcrowding has emerged as a global healthcare challenge, particularly impacting critically ill patients requiring intensive care unit (ICU) admission [1,2]. When ICU beds are unavailable [3], critically ill patients are often boarded in the ED, leading to delays in appropriate care and reducing the ED capacity to manage new arrivals. The boarding of critically ill patients in the ED represents a significant burden on healthcare systems worldwide [4]. In particular, it raises concerns regarding the quality of care provided to critically ill patients during the boarding period.

In managing septic shock, timely initiating intensive care is essential to achieving optimal patient outcomes. Evidence suggests that delays in critical care interventions can significantly increase mortality [5,6]. The adverse consequences of delayed ICU admission have been well-documented in general critically ill populations, demonstrating associations with increased mortality and greater resource utilization [7]. However, despite the time-sensitive nature of septic shock management, studies explicitly examining the optimal timing of ICU transfer in this population are limited. Furthermore, there is significant variability in the definition of "delayed admission" across existing studies [8-10], and the ideal ICU transfer timing for septic shock is unclear.

The delivery of critical care within the ED presents significant challenges in managing patients with septic shock. While EDs can provide initial management, they lack the availability of specialized equipment and staffing for sustained intensive management [11]. For instance, nurse-to-patient ratios in EDs are substantially higher than those in ICU settings, potentially compromising the delivery of time-sensitive interventions [12]. Furthermore, ED overcrowding strains available resources and may compromise the quality of care for patients with septic shock.

To address these knowledge gaps, the present study aimed to achieve two primary objectives: to establish an evidence-based threshold for delayed versus non-delayed ICU admission and to evaluate the impact of delayed ICU transfer on mortality and ICU resource utilization in patients with septic shock.

MATERIALS and METHODS

The present study was approved by the institutional review boards of each participating hospital, including Samsung Medical Center (No. 2018-05-108). The requirement for informed consent was waived owing to the observational nature of the study.

Study Design and Population

We performed a target trial emulation using a prospective cohort of the Korean Sepsis Alliance, an ongoing nationwide prospective multicenter study evaluating the clinical characteristics, management, and outcomes of sepsis and septic shock patients (Supplementary Table 1). Patients were enrolled from 19 participating hospitals between September 2019 and December 2023. The detailed protocol for patient enrollment and data collection has been previously described [13].

Our hypothetical target of this trial was to evaluate the efficacy of early ICU transfer on mortality. As there was a significant delay in the entry of ED visits during the coronavirus disease 2019 pandemic, we compared data from the period before and after the start of the pandemic (2019, 2020–2023). In addition, to avoid reverse causation bias arising from the belief that non-critical patients are transferred to the ICU slower or not at all, eligible participants were patients with septic shock who should be transferred early to the ICU. The diagnosis of septic shock was based on the third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) [14].

We excluded patients who were transferred to the ED from the operating room or another ICU (n=43); patients who did not require ICU care, defined as those who were transferred to a general ward (n=552); and patients who were transferred to the ICU more than 72 hours after admission to the ED were excluded from the study (n=10). Thus, the final sample size was 815 patients (Figure 1).

Figure 1.

Flowchart of study patient selection.

Data Collection

The primary endpoint was in-hospital mortality. Based on the Korean Sepsis Alliance registry protocol, the investigators recorded all outcomes at patient discharge and all times during the ED visit, procedure, and ICU visit. The time of the ED triage visit was considered Time 0. We then calculated the time of ED-to-ICU transfer.

Data on demographic characteristics, coexisting conditions, severity of illness, and treatment were collected by trained nurses [13]. In this study, we considered demographic factors of age, sex, and body mass index and clinical characteristics of the Charlson comorbidity index (CCI) and illness severity to calculate the initial Sequential Organ Failure Assessment (SOFA) score at Time 0 [15]. In addition, other study items of interest included the recognition of sepsis by physicians in the ED, the site of infection (e.g., respiratory, abdominal, urinary or skin/soft tissue), the admission source (community-, nursing home–, or hospital-acquired), and the Surviving Sepsis Campaign bundle and its individual component completion at 1 hour [16].

Statistical Analysis

The baseline characteristics of patients are summarized as numbers and proportions for categorical variables and mean with standard deviation or median with interquartile range (IQR) for continuous variables. To minimize systematic differences in the baseline characteristics between the two groups, all covariates were included in the logistic regression model, and we estimated the probability of transfer to the ICU within 3 hours.

To confirm the effect of early transfer to the ICU, we modelled time as a continuous variable using restricted cubic splines to find the optimal cutoff for the time (hour) of ED-to-ICU transfer. Four knots were selected based on model comparison using the Akaike Information Criterion. In determining the optimal location of these knots, Harrell's suggested knot locations recommend using the fifth, 35th, 65th, and 95th percentiles of the continuous variable [17]; thus, we performed cubic splines for the time of ED-to-ICU transfer with knots at the fifth, 35th, 65th, and 95th percentiles of our sample distributions (2.8, 5.2, 8.9, and 27.9 hours, respectively). We then calculated the linearity and non-linearity of the association between the time of ED-to-ICU transfer and in-hospital mortality by confirming that the coefficients associated with the non-linear components equaled 0 [18]. Then, we calculated odds ratio (OR) with 95% CI for in-hospital mortality according to time since ICU transfer using conditional logistic regression models. To account for inter-hospital variability, all models included hospital as a matching variable with fixed effect, thereby adjusting for site-level differences without disclosing hospital-specific results. To control for other potential confounding factors, a literature review was conducted to identify relevant variables. To address multicollinearity issues, we examined the correlation between variables and selected the most comprehensive variables from those with high correlations. The final model was adjusted for age, sex, CCI score, initial SOFA score, and mechanical ventilator use. Then, inverse probability treatment weighting (IPTW) was performed using all eligible participants in the current study as a sensitivity analysis. Differences in baseline covariates between the two groups were evaluated before and after IPTW using an absolute standardized difference with a value >0.2 indicating a significant difference [19].

We also performed a subgroup analysis to confirm that the association between time of ED-to-ICU transfer and in-hospital mortality was consistent across the groups, including those stratified by age (<75 or ≥75), sex, CCI score (<3 or ≥3 points), initial SOFA score (<10 or ≥10 points), site of infection (respiratory no vs. yes), use of mechanical ventilator or continuous renal replacement therapy (CRRT), and the use of extracorporeal membrane oxygenation (ECMO).

All tests were two-sided, and P≤0.05 was considered to indicate statistical significance. All analyses were performed using Stata version 16 (StataCorp LP) and R version 3.6.1 (R Foundation for Statistical Computing).

RESULTS

Among the 815 eligible patients, the median age was 75 years, and 59.6% were male. The median initial SOFA score was 10 points, and the most common site of infection was the respiratory tract, accounting for 44.8% of cases (44.8%). The median initial lactate was 6.55 mmol/L (IQR, 4.19) in patients transferred after ≥3 hours and 7.61 mmol/L (IQR, 5.08) in those transferred within 3 hours (P=0.079). The median time of ED-to-ICU transfer was 6.7 hours (IQR, 4.7–11.4). The median time from ED arrival to vasopressor initiation was significantly shorter in patients transferred to the ICU within 3 hours compared with those transferred later (22.0 [IQR, 10.0–42.0] vs. 36.0 [17.0–54.0] minutes; P=0.004) (Table 1). Supplementary Table 2 presents the ORs for in-hospital mortality based on various patient characteristics. A CCI score ≥3 points was not significantly associated with mortality (OR, 0.91; 95% CI, 0.49–1.66). Meanwhile, a higher initial SOFA score increased the risk of mortality (OR, 1.12; 95% CI, 1.06–1.18). Mechanical ventilator use (OR, 3.10; 95% CI, 2.25–4.29) and ECMO or CRRT use (OR, 2.24; 95% CI, 1.58–3.19) were also both strongly associated with higher mortality rates. In our cohort, 7% of patients were transferred to the ICU within 3 hours. Although these patients were more likely to have more numerous comorbidities, their severity of sepsis, as indicated by the SOFA score or ventilator use, was similar to that of patients with ICU admission after 3 hours (Table 1).

Table 1.

Characteristics of study participants

In the restricted cubic splines model, the risk of mortality increased until 6 hours (P for linearity <0.01) and then seemed to plateau thereafter (P for linearity=0.47) (Figure 2). Among the different scenarios of ICU transfer within 3, 4, 5, and 6 hours, the OR with 95% CI for in-hospital mortality were 0.48 (0.24–0.94), 0.77 (0.48–1.19), 0.91 (0.64–1.30), and 1.00 (0.73–1.38), respectively. The results of the IPTW model were similar; specifically, ICU transfer within 3 hours led to a lower risk of in-hospital death than transfer after 3 hours (OR, 0.53; 95% CI, 0.28–0.98) (Table 2).

Figure 2.

Odds ratio for in-hospital deaths by time of emergency department-to-intensive care unit (ICU) transfer. Adjusted for age, sex, Charlson comorbidity index score, initial Sequential Organ Failure Assessment score, site of infection, hospital, and use of mechanical ventilator.

Table 2.

Risk of in-hospital mortality by time of ED-to-ICU transfer in patients with septic shock

In the analyzed subgroups, patients who were transferred to the ICU within 3 hours exhibited a consistently lower adjusted OR for in-hospital mortality than those who were transferred later. This trend was observed across various patient characteristics, including age, sex, CCI score, initial SOFA score, infection site, and mechanical ventilator use (Figure 3). Among patients who did not receive ECMO or CRRT (n=576), early ICU transfer was not significantly associated with in-hospital mortality (adjusted OR, 0.85; 95% CI, 0.40–1.73). In contrast, among those who received ECMO or CRRT (n=239), early ICU transfer was strongly associated with a lower risk of in-hospital mortality (adjusted OR, 0.17; 95% CI, 0.04–0.56). The interaction between ECMO/CRRT use and early ICU transfer was statistically significant (P for interaction=0.02). In subgroup analyses stratified by 1-hour Surviving Sepsis Campaign bundle completion, the association between early ICU transfer (<3 hours) and in-hospital mortality was consistent across both groups. The adjusted OR was 0.64 (95% CI, 0.11–3.05) among patients without bundle completion and 0.54 (95% CI, 0.26–1.08) among those with bundle completion, with no significant interaction (P for interaction=0.61).

Figure 3.

Subgroup analysis. Adjusted for age, sex, Charlson comorbidity index (CCI) score, initial Sequential Organ Failure Assessment (SOFA) score, infection site, and mechanical ventilator use. OR: odds ratio; ECMO: extracorporeal membrane oxygenation; CRRT: continuous renal replacement therapy; SSC: Surviving Sepsis Campaign; ICU: intensive care unit.

DISCUSSION

Our study demonstrates that early ICU transfer within 3 hours from ED admission is associated with a significant survival benefit in patients with septic shock. Although only 7% of patients were transferred within 3 hours, and these patients had more comorbidities, their severity of illness was comparable to that of those transferred after 3 hours. The survival benefit of early ICU transfer was consistent across various subgroups but was particularly pronounced in patients requiring ECMO or CRRT.

Previous studies into the impact of ED-to-ICU transfer times on sepsis outcomes have produced conflicting evidence. Some smaller single-center studies found no association between transfer timing and mortality. For instance, a study of 172 patients with sepsis showed no difference in 30-day mortality between transfers before and after 6 hours [20], while another study of 206 patients showed no outcome differences between 1-hour and 6-hour cutoffs [21]. In contrast, larger-scale studies have indicated that delays in ICU admission are linked to increased mortality. A retrospective cohort study from Alberta demonstrated that ICU occupancy greater than 90% was associated with increased hospital mortality in sepsis patients, even after adjusting for multiple confounders [9]. Similarly, research conducted in China involving 1,997 sepsis patients found that ED length of stay exceeding 12 hours was independently associated with increased mortality risk (OR, 1.82; 95% CI, 1.28–2.58) compared to stays shorter than 6 hours [8]. Our findings help to reconcile these discrepancies by focusing specifically on patients with septic shock—a subgroup for whom the timing of critical care delivery may be even more crucial than in the broader population of critically ill patients.

To address these conflicting findings, our study implemented several key methodological innovations. One of the significant strengths of our research is the implementation of a target trial emulation design. By explicitly defining eligibility criteria, treatment strategies, and follow-up protocols within an observational dataset, we could approximate the rigor of a randomized controlled trial. This design minimizes selection bias and helps to mitigate confounding, enhancing the causal inference drawn from our findings. Second, by enrolling 815 patients across 19 centers, we ensured sufficient statistical power to detect meaningful differences in outcomes while increasing our results' generalizability within academic hospital settings. Third, restricted cubic splines analysis enabled us to determine an evidence-based optimal transfer window of 3 hours instead of the arbitrary time cutoffs used in prior studies. However, our findings do not support the use of a fixed 3-hour target for ICU transfer. The observed threshold was identified empirically through spline modeling and was not predetermined. It should therefore be viewed as a marker of the need for timely initiation of intensive care rather than as a strict rule.

Septic shock is characterized by a dysregulated host response to infection, resulting in profound circulatory, metabolic, and immunological disturbances [22,23]. The pathophysiology involves inflammatory responses, endothelial dysfunction, and coagulation abnormalities that can precipitate rapid multi-organ failure. While previous research has demonstrated the importance of early ICU transfer in general critically ill populations [7,24], septic shock represents a unique pathophysiological entity characterized by time-sensitive circulatory and metabolic derangements. The strong association between early transfer and reduced mortality suggests that this timing may be even more critical in septic shock than other critical illnesses. The relationship between transfer timing and mortality in our study showed a clear pattern: the risk increased linearly to 6 hours and then plateaued, with the most substantial benefit seen in transfers within 3 hours. This pattern aligns with our understanding of septic shock pathophysiology, where early intervention can prevent the cascade of organ dysfunction. While this association remained consistent across patient subgroups after adjusting for illness severity and comorbidities, the benefit was particularly pronounced in patients requiring ECMO or CRRT, suggesting that timely ICU transfer may be especially critical for patients who eventually require advanced organ support therapies.

Despite these findings, a significant concern is that only 7% of patients were transferred to the ICU within the 3-hour window. This highlights a considerable gap between current evidence and clinical practice [25]. Barriers such as ED overcrowding, ICU bed shortages, and the absence of standardized sepsis protocols likely contribute to this delay [26,27]. Our findings have important implications for sepsis care delivery. They suggest that ED boarding may be particularly detrimental for septic shock patients, even when high-quality ED care is provided [10,20]. This conclusion challenges the findings of some previous studies suggesting that prolonged ED stays do not affect outcomes if appropriate care is delivered. Our results indicate that the location of care (ICU vs. ED) may matter as much as the specific interventions provided.

Despite the methodological strengths of our study, several limitations warrant discussion. Residual confounding remains possible, as not all potential confounders can be measured or adjusted for our analysis. Even with target trial emulation, the observational design does not eliminate the risk of unmeasured bias, but similar findings were observed when we analyzed our data with IPTW as a sensitivity analysis. Additionally, our dataset did not capture detailed information regarding specific interventions administered during the ED boarding period. For example, the capability to initiate ECMO or CRRT in the ED varies across participating hospitals. Although we did not disclose hospital-specific data due to confidentiality agreements, our analyses accounted for hospital effects, and therefore, the observed association was unlikely to be explained solely by differences in site-level availability of advanced organ support. Another limitation is the focus on academic centers, which may limit the generalizability of our findings to community hospitals or settings with different resource constraints. Finally, subsequent studies could explore the cost-effectiveness of strategies to reduce ED-to-ICU transfer delays.

Our study provides the first robust evidence supporting early ICU transfer in septic shock patients, challenging the findings of previous smaller studies. The strong association between ED stay and mortality suggests that ICU transfer should be considered a time-sensitive therapeutic intervention in septic shock management.

KEY MESSAGES

▪ Among patients with septic shock, transferring from the emergency department to the intensive care unit within 3 hours is associated with substantially lower in-hospital mortality, whereas each additional hour of delay up to 6 hours increases risk.

▪ The benefit is most pronounced among those who need advanced organ support. These findings establish emergency department-to-intensive care unit transfer as a time-sensitive intervention.

Notes

CONFLICT OF INTEREST

Woo Hyun Cho is an editorial board member of the journal but was not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflicts of interest relevant to this article were reported.

FUNDING

This work was supported by the Research Program funded by the Korea Disease Control and Prevention Agency (fund code 2019E280500, 2020E280700, 2021-10-026) and supported by Korean Sepsis Alliance (KSA) affiliated with Korean Society of Critical Care Medicine (KSCCM).

ACKNOWLEDGMENTS

None.

AUTHOR CONTRIBUTIONS

Conceptualization: REK, GYS, DK, JHC. Methodology: DK, JC, REK. Formal analysis: DK, JC. Data curation: DGH, WYK, YJL, WHC, SP. Visualization: REK, JHC. Project administration: GYS. Funding acquisition: GYS. Writing - original draft: JHC, DK. Writing - review & editing: REK, GYS. All authors read and agreed to the published version of the manuscript.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found via https://doi.org/10.4266/acc.003575.

Supplementary Table 1.

Speciﬁcation and emulation of a target trial comparing ICU transfer at ≥3 hr and <3 hr on the risk for in-hospital death using observational data

acc-003575-Supplementary-Table-1.pdf

Supplementary Table 2.

Odds ratios for in-hospital death

acc-003575-Supplementary-Table-2.pdf

References

1. Cowan RM, Trzeciak S. Clinical review: Emergency department overcrowding and the potential impact on the critically ill. Crit Care 2005;9:291–5. 10.1186/cc2981. 15987383.

2. Di Somma S, Paladino L, Vaughan L, Lalle I, Magrini L, Magnanti M, et al. Overcrowding in emergency department: an international issue. Intern Emerg Med 2015;10:171–5. 10.1007/s11739-014-1154-8. 25446540.

3. Choi J, Kim SH, Ko RE, Suh GY, Yang JH, Park CM, et al. Simulating the effects of reducing transfer latency from the intensive care unit on intensive care unit bed utilization in a Korean tertiary hospital. Acute Crit Care 2025;40:18–28. 10.4266/acc.002976. 39978945.

4. Mohr NM, Wessman BT, Bassin B, Elie-Turenne MC, Ellender T, Emlet LL, et al. Boarding of critically ill patients in the emergency department. Crit Care Med 2020;48:1180–7. 10.1097/ccm.0000000000004385. 32697489.

5. Li Y, Li H, Zhang D. Timing of norepinephrine initiation in patients with septic shock: a systematic review and meta-analysis. Crit Care 2020;24:488. 10.1186/s13054-020-03204-x. 32762765.

6. Leisman D, Wie B, Doerfler M, Bianculli A, Ward MF, Akerman M, et al. Association of fluid resuscitation initiation within 30 minutes of severe sepsis and septic shock recognition with reduced mortality and length of stay. Ann Emerg Med 2016;68:298–311. 10.1016/j.annemergmed.2016.02.044. 27085369.

7. Kiekkas P, Tzenalis A, Gklava V, Stefanopoulos N, Voyagis G, Aretha D, et al. Delayed admission to the intensive care unit and mortality of critically ill adults: systematic review and meta-analysis. Biomed Res Int 2022;2022:4083494. 10.1155/2022/4083494. 35146022.

8. Zhang Z, Bokhari F, Guo Y, Goyal H. Prolonged length of stay in the emergency department and increased risk of hospital mortality in patients with sepsis requiring ICU admission. Emerg Med J 2019;36:82–7. 10.1136/emermed-2018-208032. 30518642.

9. Yergens DW, Ghali WA, Faris PD, Quan H, Jolley RJ, Doig CJ, et al. Assessing the association between occupancy and outcome in critically ill hospitalized patients with sepsis. BMC Emerg Med 2015;15:31. 10.1186/s12873-015-0049-y. 26481448.

10. Agustin M, Price LL, Andoh-Duku A, LaCamera P. Impact of delayed admission to the intensive care unit from the emergency department upon sepsis outcomes and sepsis protocol compliance. Crit Care Res Pract 2017;2017:9616545. 10.1155/2017/9616545. 28409028.

11. Carlbom DJ, Rubenfeld GD. Barriers to implementing protocol-based sepsis resuscitation in the emergency department--results of a national survey. Crit Care Med 2007;35:2525–32. 10.1097/01.ccm.0000298122.49245.d7. 18075366.

12. Lasater KB, Sloane DM, McHugh MD, Cimiotti JP, Riman KA, Martin B, et al. Evaluation of hospital nurse-to-patient staffing ratios and sepsis bundles on patient outcomes. Am J Infect Control 2021;49:868–73. 10.1016/j.ajic.2020.12.002. 33309843.

13. Park S, Jeon K, Oh DK, Choi EY, Seong GM, Heo J, et al. Normothermia in patients with sepsis who present to emergency departments is associated with low compliance with sepsis bundles and increased in-hospital mortality rate. Crit Care Med 2020;48:1462–70. 10.1097/ccm.0000000000004493. 32931189.

14. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016;315:801–10. 10.1001/jama.2016.0287. 26903338.

15. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707–10. 10.1007/bf01709751. 8844239.

16. Levy MM, Evans LE, Rhodes A. The Surviving Sepsis Campaign bundle: 2018 update. Intensive Care Med 2018;44:925–8. 10.1007/s00134-018-5085-0. 29675566.

17. Harrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis Springer; 2001.

18. Gauthier J, Wu QV, Gooley TA. Cubic splines to model relationships between continuous variables and outcomes: a guide for clinicians. Bone Marrow Transplant 2020;55:675–80. 10.1038/s41409-019-0679-x. 31576022.

19. Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011;10:150–61. 10.1002/pst.433. 20925139.

20. Hirschy R, Sterk E, Dobersztyn R, Rech MA. Time spent in the emergency department and outcomes in patients with severe sepsis and septic shock. Adv Emerg Nurs J 2018;40:94–103. 10.1097/TME.0000000000000188. 29715251.

21. Elay G, Al B. Thirty-day mortality in septic shock patients is directly associated with high disease severity score but not with the length of stay in the emergency department. J Emerg Med 2020;19:172–7.

22. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013;369:840–51. 10.1056/nejmra1208623. 23984731.

23. Park C, Ku NS, Park DW, Park JH, Ha TS, Kim DW, et al. Early management of adult sepsis and septic shock: Korean clinical practice guidelines. Acute Crit Care 2024;39:445–72. 10.4266/acc.2024.00920. 39622601.

24. Groenland CN, Termorshuizen F, Rietdijk WJ, van den Brule J, Dongelmans DA, de Jonge E, et al. Emergency department to ICU time is associated with hospital mortality: a registry analysis of 14,788 patients from six university hospitals in the Netherlands. Crit Care Med 2019;47:1564–71. 10.1097/ccm.0000000000003957. 31393321.

25. Klompas M, Levy MM. What can we learn from differences in hospitals' sepsis care? Am J Respir Crit Care Med 2025;211:546–7. 10.1164/rccm.202501-0106ed. 40019819.

26. Shin TG, Jo IJ, Choi DJ, Kang MJ, Jeon K, Suh GY, et al. The adverse effect of emergency department crowding on compliance with the resuscitation bundle in the management of severe sepsis and septic shock. Crit Care 2013;17:R224. 10.1186/cc13047. 24093643.

27. Gaieski DF, Agarwal AK, Mikkelsen ME, Drumheller B, Cham Sante S, Shofer FS, et al. The impact of ED crowding on early interventions and mortality in patients with severe sepsis. Am J Emerg Med 2017;35:953–60. 10.1016/j.ajem.2017.01.061. 28233644.

Article information Continued

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Variable	Overall (n=815)	Time of ED-to-ICU transfer
Variable	Overall (n=815)	≥3 hr (n=762)	<3 hr (n=53)	P-value	Unweighted SMD	Weighted SMD
Age (yr)	75 (64–82)	76 (64–82)	75 (66–81)	0.403	0.132	0.125
Sex, male	486 (59.6)	454 (59.6)	32 (60.4)	0.999	0.016	0.059
Body mass index (kg/m²)	21.92±4.24	21.91±4.20	22.06±4.81	0.803	0.033	0.044
Charlson comorbidity index (≥3 points)	80 (9.8)	71 (9.3)	9 (17.0)	0.115	0.228	0.189
Initial SOFA score	10.0 (8.0–12.0)	10.0 (8.0–12.0)	10.0 (8.0–13.0)	0.209	0.204	0.051
Initial lactate level (mmol/L)	6.62±4.26	6.55±4.19	7.61±5.08	0.079	0.228	0.111
Site of infection
Lungs	365 (44.8)	343 (45.0)	22 (41.5)	0.724	0.071	0.120
Abdomen	236 (29.0)	222 (29.1)	14 (26.4)	0.791	0.061	0.004
Urinary tract	178 (21.8)	165 (21.7)	13 (24.5)	0.751	0.068	0.048
Others	111 (13.6)	104 (13.6)	7 (13.2)	0.999	0.013	0.063
Type of infection				0.197
Community-acquired	577 (70.8)	537 (70.5)	40 (75.5)		0.130	0.008
Nursing home–acquired	76 (9.3)	70 (9.2)	6 (11.3)		0.066	0.091
Nursing hospital–acquired	106 (13.0)	104 (13.6)	2 (3.8)		0.367	0.319
Hospital-acquired	56 (6.9)	51 (6.7)	5 (9.4)		0.675	0.201
Use of mechanical ventilator	393 (48.2)	367 (48.2)	26 (49.1)	1.000	0.018	0.088
Use of ECMO or CRRT	239 (29.3)	221 (29.0)	18 (34.0)	0.541	0.107	0.071
Year of admission				0.399
2019	111 (13.6)	100 (13.1)	11 (20.8)		0.194	0.194
2021	250 (30.7)	237 (31.1)	13 (24.5)		0.152	0.152
2022	265 (32.5)	249 (32.7)	16 (30.2)		0.072	0.072
2023	189 (23.2)	176 (23.1)	13 (24.5)		0.067	0.067
Time from ED to vasopressor (min)	34.0 (16.0–53.3)	36.0 (17.0–54.0)	22.0 (10.0–42.0)	0.004	0.282	0.201

Time since ICU transfer^a)	Participant	Death, no (%)	Adjusted OR^b) (95% CI)	IPTW OR (95% CI)
<3 hr
<3 hr	53	17 (32.1)	0.48 (0.24–0.94)	0.53 (0.28–0.98)
≥3 hr	762	342 (44.9)	Reference	Reference
<4 hr
<4 hr	129	52 (40.3)	0.77 (0.48–1.19)	0.76 (0.49–1.18)
≥4 hr	686	307 (44.8)	Reference	Reference
<5 hr
<5 hr	244	103 (42.2)	0.91 (0.64–1.30)	0.94 (0.66–1.33)
≥5 hr	571	256 (44.8)	Reference	Reference
<6 hr
<6 hr	338	149 (44.1)	1.00 (0.73–1.38)	0.99 (0.73–1.34)
≥6 hr	477	210 (44.0)	Reference	Reference