Changes in Life-Sustaining Treatment Limitations in Swedish ICUs During the COVID-19 Pandemic

doi:10.21203/rs.3.rs-8044900/v1

Download PDF

Research Article

Changes in Life-Sustaining Treatment Limitations in Swedish ICUs During the COVID-19 Pandemic

https://doi.org/10.21203/rs.3.rs-8044900/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The use of limitations of life-sustaining treatment (LLST) in Intensive Care Units (ICUs) varies internationally. The COVID-19 pandemic led to significant changes worldwide, but its impact on ICU LLST remains unclear. This study aimed to assess the prevalence of LLST in Swedish ICUs from 2018 to 2022 and whether the pandemic and COVID-19 influenced their utilization.

Methods

All ICU admissions registered in the Swedish Intensive Care Registry from 2018–2022 were screened. Cases with post-anesthesia care of less than 24 hours and subsequent ICU admissions were excluded. Logistic regression was used to analyze associations with LLST, defined as any limitation of therapy, including withholding or withdrawing therapy. March 16, 2020, marked the start of the pandemic.

Results

In total, 77 735 ICU admissions were analyzed: 39 396 pre-pandemic and 38 338 during the pandemic. Patients admitted during the pandemic had slightly higher odds (OR 1.06, 95% CI 1.03–1.10, p < 0.001) of receiving LLST compared to before the pandemic. COVID-19 patients had an increased odds of receiving a LLST compared to non COVID-19 patients (OR 1.17, 95% CI 1.09–1.25, p < 0.001). ICU capacity utilization was not associated with LLST use in our study.

Conclusions

The findings suggest a shift in LLST use, with increased odds during the pandemic and among COVID-19 patients specifically. This may reflect greater attention to active decision-making regarding life-sustaining therapy during this period.

1.1 Background

The COVID-19 pandemic spread abruptly in 2020, causing a global humanitarian crisis [1]. This posed challenges in healthcare prioritization and emphasized the need for ethical crisis standards to guide clinicians in deciding when standard prioritization approaches can no longer be applied [2].

The diagnostic uncertainty of a new illness, together with resource constraints and workforce shortages, led to debates and recommendations in prioritization within Sweden’s healthcare system. Among these were national guidelines on LLST (limitations of life-sustaining therapies) and ethical frameworks for resource allocation [3]. This prompted extensive debates on increasing the use of LLST such as withholding ventilator treatment, withholding ICU (Intensive Care Unit) care and Do-Not-Attempt Cardiopulmonary Resuscitation (DNACPR) orders [4].

The COVID-19 pandemic challenged the priority of care especially for patients with chronic, life-limiting disease and many countries issued recommendations of advance care planning and decisions about LLST [5]. The first reports during the pandemic indicated that outcomes for COVID-19 treated patients following in-hospital cardiac arrest were poor, suggesting considerations on DNACPR orders for COVID-19 patients [6]. However, to what extent the COVID-19 pandemic affected DNACPR rates and utilization of other LSTT in the ICU remain uncertain.

ICU care targets major organ dysfunction and should only be provided if it benefits the patient and is not futile or against their interests. Determining whether ICU interventions will benefit the patient or are futile is often complex and not always easy to ascertain. LLST—such as DNACPR orders, withholding mechanical ventilation or renal replacement therapy—may be considered when interventions are unlikely to provide meaningful benefit. There is a variability in the use of LLST in the ICU within countries, cultures, hospitals and single sites [7, 8, 9]. The reasons for the variability are partly due to differences between countries, health care systems, hospitals, cultures and individual health care providers [7]. One Swedish study before the pandemic [8] looked at LLST in all Swedish ICUs in the Swedish Intensive Care Registry from 2014–2016 and found a LLST prevalence of 15,4%. Another study looked at LLST in hospitals (not only ICUs) from one Swedish region during the pandemic from 2020–2021 and found a LLST prevalence of 30% [9]. The decisions to LLST are generally perceived difficult for care givers and may raise controversies within care teams, with patients or with next-of-kin [10]. Two previous Swedish studies have shown that factors associated with LLST in the ICU pre-pandemic in general include higher age, female sex, severe comorbidity, longer length of stay (LOS) and higher SAPS3-score [8, 11].

1.2 Objectives

We aimed to investigate differences in LLST before and during the COVID-19 pandemic and whether covariates affecting LLST utilization differed.

2.1 Study design

This is a registry-based cohort study from the Swedish Intensive Care Registry [12]. The Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines were followed [13]. The Swedish Intensive Care Registry was launched 2001 and in 2019, 83 of 84 Swedish ICUs contributed to the registry. The registry contains patient level data, registered daily by physicians and nurses in the Swedish ICUs during the patient stay. Physicians are encouraged to document daily decisions to continue “Full Care” or to withdraw/withhold treatment, aided by reminders in the Electronic Health Record; however, they may choose not to, resulting in a “No Documented Decision.” The information from the registry is confidential and is only to be handed over after ethical review with de-identification of the data [14].

2.2 Participants

All ICU-admissions for adult patients (≥ 18 years of age) in Sweden from January 1st, 2018, to April 14th, 2022, were screened for inclusion. Exclusion criteria were post anesthesia care < 24 hours and subsequent ICU admissions (including index events only), see Fig. 1.

2.3 Data collection and variables

The registry documentation regarding LLST includes data variables that are registered both as withdraw, withhold and what type of LLST. Data completeness regarding LLST varies, as not all hospitals register all variables. To enable correct comparisons, only cases with a verified registration relating to LLST were included (“Full Care or LLST). In Fig. 4, we include patients without documented decisions to illustrate how the frequency of decision making changed over time. Since only a small proportion of hospitals reported what kind of therapy limitation was present (withhold, withdraw care, DNACPR, no renal replacement therapy, no invasive ventilation), we chose not to report that data. For a detailed description of missing data, see Supplement 1.

From the registry, data was extracted for hospital type (community, county, district), patient characteristics, acute and chronic conditions (including COVID-19 status), and SAPS3 (see definitions, 15]. We defined pre–COVID-19 as before March 16, 2020, and the COVID-19 pandemic as March 16, 2020 onward. Within the COVID-19 pandemic we stratified admissions by COVID-19 or non COVID-19 patients. Patients with an active SARS-CoV-2 infection presenting the primary clinical manifestations that required ICU admission were classified as COVID-19 cases.

2.4 Exposure and outcomes

Exposures were COVID-19 pandemic period and COVID-19 vs non COVID-19 patients. A COVID-19 patient was defined as such if COVID-19 was the main reason for ICU admission. The primary outcome was the presence of LLST. Covariates associated with a LLST decision were secondary outcomes.

2.5 Statistical analysis

Baseline characteristics are presented using means, medians and proportions, with appropriate measurements of dispersion. Differences between our groups were analyzed using chi-square test to compare the baseline categorical variables, and students t-test and ANOVA for continuous continuous variables. To visualize potentially complex interactions between covariates, and identify confounders, mediators, and potential collider bias, we used Directed Acyclic Graphs (DAGs) [16]. We utilized the DAG (Supplement 2) to identify covariates for adjustment for the effect of the pandemic on LLST. Multivariable logistic regression analysis was performed to study independent associations with the presence of LLST. Covariates included were; age, sex, SAPS3-core, hospital type, LOS, COVID-19 status, pandemic period or not and ICU occupancy. The most simple model possible including exposure, outcome and confounders was built. To understand the impact of possible mediators, we built additional logistic regression models including COVID-19 status, LOS and ICU occupancy. P-values < 0.05 were considered significant. Standardized mean difference (SMD) was used to quantify differences between groups (SMD = difference between the two groups means divided by a standard deviation). SMD values < 0.1 (10%) were considered negligible. All analysis was performed using R-studio version 4.3.1.

77 735 ICU admissions were included in the final cohort, 39 396 pre- and 38 338 during the COVID-19 pandemic (Supplement 3).

3.1 Baseline characteristics

There was no difference in prevalence of LLST between the pre-pandemic to the pandemic period. ICU-survival was lower in the pandemic period whereas 30-day survival was not significantly different (Table 1).

Table 1

Baseline characteristics during the pre-pandemic and pandemic
Variables	Pre pandemic	Pandemic	p	test	SMD
ICU admissions (n)	39396	38339
Age (mean (SD))	62.39 (18.97)	62.11 (17.98)	00.33		0.015
Sex = F (n, %)	16848 (42.8)	15256 (39.8)	< 0.001		0.06
Hospital type (n, %)			< 0.001		0.033
Community hospital	10033 (25.5)	9327 (24.3)
County hospital	17257 (43.8)	17367 (45.3)
District (University) hospital	12106 (30.7)	11645 (30.4)
*ICU LOS, days (median [IQR])**	1.13 [0.58, 2.75]	1.42 [0.64, 3.91]	< 0.001	nonnorm	0.226
Surgery (n, %)			< 0.001		0.073
Yes - Emergency	5550 (14.1)	4920 (12.8)
Yes - elective	2981 (7.6)	2328 (6.1)
No	30865 (78.3)	31091 (81.1)
SAPS3 Score (mean (SD))	57.07 (16.46)	57.02 (15.56)	0,614		0.004
Body Temp Max (median [IQR])	37.00 [36.20, 37.50]	37.00 [36.30, 37.50]	< 0.001	nonnorm	0.037
Heart Rate Max (mean (SD))	98.93 (27.16)	97.46 (26.76)	< 0.001		0.054
*SBP Min (mean (SD))**	106.43 (34.40)	109.32 (34.04)	< 0.001		0.085
Bilirubin Max (median [IQR])	10.00 [6.00, 17.00]	10.00 [6.00, 17.00]	0,368	nonnorm	0.003
Creatinine Max (median [IQR])	86.00 [65.00, 131.00]	83.00 [62.00, 124.00]	< 0.001	nonnorm	0.017
Leukocytes Max (median [IQR])	11.70 [8.20, 16.10]	11.20 [7.90, 15.70]	< 0.001	nonnorm	0.012
Platelets Min (median [IQR])	221.00 [163.00, 286.00]	225.00 [167.00, 292.00]	< 0.001	nonnorm	0.049
pH-level Min (median [IQR])	7.36 [7.28, 7.42]	7.38 [7.30, 7.44]	< 0.001	nonnorm	0.093
FiO2 (mean (SD))	49.81 (22.19)	55.67 (24.30)	< 0.001		0.252
PaO2 (median [IQR])	11.50 [9.40, 15.00]	11.00 [9.00, 14.20]	< 0.001	nonnorm	0.063
Ventilator treatment = No (n, %)	18382 (55.5)	17120 (51.6)	< 0.001		0.079
FiO2 (mean (SD))	40.68 (21.18)	43.45 (22.95)	< 0.001		0.125
PaO2 (mean (SD))	13.30 (9.43)	12.97 (7.25)	< 0.05		0.04
Thrombocytes (mean (SD))	230.00 (114.74)	237.91 (120.09)	< 0.001		0.067
Bilirubin (mean (SD))	18.00 (35.90)	18.43 (36.66)	0,408		0.012
MAP (mean (SD))	68.09 (19.28)	69.44 (18.52)	< 0.001		0.071
Length, cm (mean (SD))	1.72 (0.11)	1.72 (0.11)	0,85		0.003
Weight, kg (mean (SD))	79.93 (20.40)	81.75 (20.86)	< 0.001		0.088
Non COVID-19 (%)	39384 (100.0)	31168 (81.3)	< 0.001		0.677
ICU Occupancy nr. (mean (SD))	338.43 (29.09)	428.65 (123.85)	< 0.001		1.003
ICU Occupancy > 80% (n, %)	941 (2.4)	17373 (45.3)	< 0.001		1.166
ICU survival (n, %)	34976 (88.8)	33529 (87.5)	< 0.001		0.041
30-day survival (n, %)	29645 (77.2)	28951 (76.9)	0,423		0.006
90-day survival (n, %)	10146 (73.4)	10080 (73.1)	0,374		0.007
LLST (n, %)	9608 (24.4)	9398 (24.5)	0,887		0.001
Note: Baseline table for all patients included in the primary analysis stratified in two groups: Pre Covid19-pandemic: before March 16, 2020. Covid19-pandemic: from March 16, 2020. SMD: the standardized mean difference is the difference between the means for the two groups divided by their standard

There were 31 168 non COVID-19 patients, vs 7 171 COVID-19 patients during the pandemic period. In the COVID-19 group there were more men, the LOS was days longer and the ICU- and 30-day survival rate were lower. There was a minimal difference in LLST use. COVID-19 vs non COVID-19 differed substantially regarding the proportion of patients treated with invasive ventilation (Table 2).

Table 2

Baseline characteristics grouped by COVID-19 status
Variables	Non COVID-19	COVID-19	p	test	SMD
ICU admissions (n)	31168	7171
Age (mean (SD))	62.34 (18.78)	61.10 (13.95)	< 0.001		0.075
Sex = F (n, %)	13090 (42.0)	2166 (30.2)	< 0.001		0.247
Hospital type (n, %)			< 0.001		0.234
Community hospital	8138 (26.1)	1189 (16.6)
County hospital	13799 (44.3)	3568 (49.8)
District (University) hospital	9231 (29.6)	2414 (33.7)
ICU LOS, days (median [IQR])	1.07 [0.55, 2.56]	6.57 [2.49, 14.04]	< 0.001	nonnorm	0.916
Surgery (n, %)			< 0.001		0.636
Yes - Emergency	4758 (15.3)	162 (2.3)
Yes - elective	2303 (7.4)	25 (0.3)
No	24107 (77.3)	6984 (97.4)
SAPS3 Score (mean (SD))	57.12 (16.38)	56.54 (11.25)	0,004		0.041
Body Temp Max (median [IQR])	36.80 [36.20, 37.30]	37.50 [36.90, 38.20]	< 0.001	nonnorm	0.63
Heart Rate Max (mean (SD))	98.33 (27.38)	93.71 (23.57)	< 0.001		0.181
SBP Min (mean (SD))	106.94 (34.61)	119.62 (29.32)	< 0.001		0.395
Bilirubin Max (median [IQR])	11.00 [6.50, 18.00]	9.00 [6.00, 12.00]	< 0.001	nonnorm	0.273
Creatinine Max (median [IQR])	86.00 [64.00, 133.00]	71.00 [57.00, 95.00]	< 0.001	nonnorm	0.271
Leukocytes Max (median [IQR])	11.80 [8.30, 16.40]	9.20 [6.70, 12.50]	< 0.001	nonnorm	0.194
Platelets Min (median [IQR])	221.00 [163.00, 287.00]	240.00 [182.00, 313.00]	< 0.001	nonnorm	0.204
pH-level Min (median [IQR])	7.36 [7.28, 7.42]	7.44 [7.38, 7.48]	< 0.001	nonnorm	0.681
FiO2 (mean (SD))	49.74 (23.20)	71.39 (19.74)	< 0.001		1.002
PaO2 (median [IQR])	11.80 [9.60, 15.10]	9.00 [7.80, 10.80]	< 0.001	nonnorm	0.49
Ventilator treatment = No (n, %)	15042 (57.0)	2078 (30.6)	< 0.001		0.552
FiO2 (mean (SD))	39.27 (21.01)	64.91 (20.39)	< 0.001		1.239
PaO2 (mean (SD))	13.60 (7.61)	9.78 (3.68)	< 0.001		0.641
Thrombocytes (mean (SD))	232.68 (121.25)	259.71 (112.58)	< 0.001		0.231
Bilirubin (mean (SD))	19.99 (37.96)	11.92 (29.81)	< 0.001		0.236
MAP (mean (SD))	68.41 (18.77)	73.63 (16.81)	< 0.001		0.293
ICU Occupancy nr. (mean (SD))	408.96 (116.07)	514.21 (120.22)	< 0.001		0.891
ICU Occupancy > 80% (n, %)	11839 (38.0)	5534 (77.2)	< 0.001		0.864
30-day survival (n, %)	23644 (77.3)	5307 (75.3)	< 0.001		0.047
90-day survival (n, %)	22386 (73.6)	4938 (70.1)	< 0.001		0.062
LLST (n, %)	7693 (24.7)	1675 (23.4)	0,019		0.031
Note: Baseline table for all patients during the COVID-19 pandemic stratified in two groups: Non COVID-19: Tested negative for SARS-CoV-2 virus COVID-19: Tested positive for SARS-CoV-2 virus Definition of COVID-19 pandemic: March 16, 2020, to April 14, 2022. SMD: the standardized mean difference is the difference between the means for the two groups divided by their standard deviation (SD). Values below 0.1 (10%) are considered inconsequential (i.e., no difference between the groups). LLST: limitations of life-sustaining therapies ICU, intensive care unit; SAPS3, Simplified Acute Physiology Score 3

3.2 Main results

In the multivariable model (Fig. 1), patients during the pandemic period had slightly higher odds ratio (OR 1.06, 95% CI 1.03–1.10, p < 0.001) of receiving a LLST compared to before the pandemic. Higher age and sex were associated with an increased risk of a LLST both before and during the pandemic (Figs. 1 and 2). In contrast, University-/district, compared to community hospitals in Sweden, were associated with lower odds ratios of a LLST both before and during the pandemic (OR 0.76, 95% CI 0.72–0.80, p < 0,01 vs OR 0.79, 95% CI 0.73–0.85, p < 0.01) (Figs. 1 and 2). To investigate potential mediation of the effect of the pandemic on LLST, we performed regression analyses with/ without COVID-19 status, ICU occupancy, SAPS-III score, and length of stay (Supplement 4–9). The analyses did not suggest significant mediation of either factor (Supplement 4–9).

COVID-19 patients received LLST a higher degree compared to non COVID-19 patients (OR 1.17, 95% CI 1.09–1.25, p < 0.001) (Fig. 2).

4.1 Key results

We observed that the pandemic period was independently associated with an increased likelihood of receiving LLST, and that this association was more pronounced among patients with COVID-19. The effects of key covariates on LLST also varied depending on COVID-19 status and the timing of the pandemic. (Supplement 4–9).

4.2 Increased likelihood of LLST in the COVID-19 period

The observed increase in LLST during the COVID-19 pandemic, although modest, suggests a shift in clinical practice and decision-making. The change over time, however, was surprisingly small. The reasons for this merit further discussion.

Before the pandemic, a larger proportion of patients had no documented LLST decision (“No Documented Decision”; Fig. 3). Because these cases were excluded from the analysis, LLST use before the pandemic may have been overestimated. During the pandemic, the number of patients with “No Documented Decisions” decreased, meaning that fewer such cases were excluded. Consequently, the observed increase in LLST use during the pandemic may appear smaller than it truly was, since part of the pre-pandemic LLST rate may reflect missing documentation rather than actual treatment limitation decisions. This interpretation aligns with findings by Jönson et al. (2022) [17], who reported similar survival among patients with “Full Care” and those with “No Documented Decision”, suggesting that the latter group may have been more comparable to “Full Care” than to LLST cases.

Another reason why the increase in LLST was only marginal in the ICU during the pandemic, may be a higher threshold for ICU admission due to actual or perceived bed shortages. As shown in Table 1, mean ICU occupancy was substantially higher during the COVID-19 pandemic, and higher LLST rates outside the ICU [18, 19] indicate that more patients were deemed “Not for ICU” during this period. Our study did not capture LLSTs for patients on general wards and could not adjust for this. In our multivariable analysis, ICU occupancy was not independently associated with LLST presence (Supplement 4 and 9).

Sotoodeh et al. (2025) studied 20,261 hospitalized Swedish patients and found that higher occupancy increased the likelihood of patients being ‘Not for ICU’ [20]. Since our analysis only included admitted patients, ICU occupancy likely influenced ICU admission decisions more than LLST use among those admitted. Limited ICU resources may have necessitated prioritizing patients with more severe acute conditions, raising ethical considerations. However, the mean SAPS3 score was similar before and during the pandemic (p = 0.64), suggesting that illness severity among admitted patients remained largely unchanged despite these pressures

4.3 COVID-19 status and the pandemic

During the COVID-19 pandemic period, we found a significantly increased risk of LLST among COVID-19 compared to non-COVID-19 patients (OR 1.17, 95%, CI 1.07–1.25). This somewhat contradicts findings from an Australian study [21], where frail patients with, compared to without, COVID-19 had the same frequency of LLST. As shown in our study, SAPS3 scores—used to predict hospital mortality at ICU admission—did not differ significantly before or during the pandemic, and only minimally between COVID-19 and non-COVID-19 patients (Tables 1 and 2).

Multiple studies demonstrated that the pandemic, and more specifically, COVID-19 status, had a significant but divergent impact on time to surgery and surgical outcomes [22, 23] and delayed or avoided acute care [24, 25, 26].

The mechanisms underlying the increased likelihood of LLST among COVID-19 patients remain unclear. Syrous et al. [27] examined end-of-life decision-making in critically ill elderly patients during the pandemic and found that LLST decisions in COVID-19 patients were influenced by different factors compared with non-COVID-19 patients. Similarly, the European multicenter COVID-ICU study [28] identified age, frailty, and early severity of respiratory failure as key determinants of LLST, with notable variation across centers. LLST was, as expected, strongly associated with higher mortality, highlighting a fundamental challenge for intensivists: whether LLST contributes to mortality, reflects the severity of illness, or represents a combination of both.

Although the overall rise in LLST was modest during the COVID-19 pandemic, the qualitative change in documentation practices—from 18.5% of patients lacking a recorded decision in 2018 to 11.4% in 2022, a 39% relative decrease (see Fig. 3)—may reflect a heightened ethical awareness, which would make it a sign of meaningful evolution. Future research could explore whether these changes have persisted beyond the acute phase of the pandemic and how they have impacted patient outcomes, staff experiences, and ethical climate in healthcare settings.

4.2 Strengths and Limitations

This is a nationwide registry study with strengths including high completeness and limited selection bias. Limitations include accuracy variations due to human error at registration and measurement bias due to variability in data completeness collections across different sites. There are limitations to the observational design with possible unmeasured confounders and residual confounding. Although we used DAGs and attempted to analyse mediation, such analyses rely on multiple assumptions that are difficult to meet, why the mediation analyses should be interpreted with caution.

The results of this study suggest a shift in use of LLST during the COVID-19 pandemic as evidenced by the increased use of LLST during the pandemic and among COVID-19 patients. It is possible that the increased emphasis on active decision making during the pandemic was the driver of this change.

6.1 Ethics approval and consent to participate

This retrospective registry-based study used pseudonymized data from the Swedish Intensive Care Registry (Svenska Intensivvårdsregistret, SIR; Dnr 202224). Ethical approval was obtained from the Swedish Ethical Review Authority (Etikprövningsmyndigheten), approval numbers Dnr 2022-02760-01 and Dnr 2023-02923-02. The requirement for informed consent was waived because the study was based on existing registry data that were pseudonymized prior to access. All data handling complied with the General Data Protection Regulation (GDPR) and institutional data protection policies. Clinical trial registration: Not applicable.

6.2 Consent for publication

Not applicable.

6.5 Funding

This study received funding from the Department of Anaesthesiology and Intensive Care at Sahlgrenska University Hospital, Mölndal, Sweden.

Author Contribution

Lisa Wiltz drafted the manuscript with substantial input from Fredrik Hessulf, Kasper Glerup Lauridsen, Adam Piasecki, and Joar Björk. Lisa Wiltz and Fredrik Hessulf performed the main statistical analyses. Kasper Glerup Lauridsen and Tobias Siöland provided statistical support and guidance. All authors contributed to interpretation, critically revised the text, and approved the final version.

Acknowledgement

We thank the Swedish Intensive Care Registry (SIR) for providing data and the clinical teams for their contributions.

Data Availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request, subject to ethical and legal restrictions.

World Health Organization [WHO] (2025, May 7.) “Corona virus disease (COVID-19)”. https://www.who.int/emergencies/diseases/novel-coronavirus-2019
Kramer, Daniel B., Bernard Lo, and Neal W. Dickert. 2020. “CPR in the Covid-19 Era, An Ethical Framework.” New England Journal of Medicine 383 (2): e6. https://doi.org/10.1056/nejmp2010758.
Socialstyrelsen. (2020, april). Nationella principer för prioritering av rutinsjukvård under covid-19-pandemin: Kunskapsstöd för att utveckla regionala och lokala riktlinjer (Dnr: 13865/2020). https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/dokument-webb/ovrigt/nationella-principer-for-prioritering-av-rutinsjukvard-covid19.pdf
Ludvigsson, Jonas F. 2022. “How Sweden Approached the COVID-19 Pandemic: Summary and Commentary on the National Commission Inquiry.” Acta Paediatrica 112 (1): 19–33. https://doi.org/10.1111/apa.16535.
Curtis, J. Randall, Erin K. Kross, and Renee D. Stapleton. 2020. “The Importance of Addressing Advance Care Planning and Decisions About Do-Not-Resuscitate Orders During Novel Coronavirus 2019 (COVID-19).” JAMA, March. https://doi.org/10.1001/jama.2020.4894.
Mark Taubert, John Idris Baker, Anna Hudson, Elin Harding, ‘Do Not Attempt CPR’: how the pandemic changed perceptions and practice, Medicine, Volume 52, Issue 7, 2024, Pages 426-428, ISSN 1357-3039, https://doi.org/10.1016/j.mpmed.2024.04.008.
Mark, N. M., S. G. Rayner, N. J. Lee, and J. R. Curtis. 2015. “Global Variability in Withholding and Withdrawal of Life-Sustaining Treatment in the Intensive Care Unit: A Systematic Review.” Intensive Care Medicine 41 (9): 1572–85. https://doi.org/10.1007/s00134-015-3810-5.
Block, Linda, Max Petzold, Alma Nordenskjöld Syrous, Birgitta Lindqvist, Helena Odenstedt Hergès, and Silvana Naredi. 2019. “Age, SAPS 3 and Female Sex Are Associated with Decisions to Withdraw or Withhold Intensive Care.” Acta Anaesthesiologica Scandinavica 63 (9): 1210–15. https://doi.org/10.1111/aas.13411.
Öhrling G, Taxbro K. Inga skillnader i andel beslut att avstå livsuppehållande covidvård [Limitations of life-sustaining treatment in hospitalized patients with COVID-19: A retrospective study in a Swedish healthcare region]. Lakartidningen. 2022 Nov 30;119:22110. Swedish. PMID: 36448934.
K. Beck, A. Vincent, H. Cam, C. Becker, S. Gross, N. Loretz, et al. 2022. Medical futility regarding cardiopulmonary resuscitation in in-hospital cardiac arrests of adult patients: A systematic review and Meta-analysis. Resuscitation 2022 Vol. 172 Pages 181-193. DOI: 10.1016/j.resuscitation.2021.11.041
Jönsson, Nino, Niklas Pettersson, Peter Asplund, Anders Bremer, Stefan Lehtipalo, and Fredrik Hessulf. 2023. “Factors Associated with Treatment Limitations in Two Swedish Intensive Care Units: Prevalence and Patient Involvement.” Acta Anaesthesiologica Scandinavica 67 (3): 339–46. https://doi.org/10.1111/aas.14185.
Svenska Intensivvårdsregistret. (n.d.). The Swedish Intensive Care Registry (SIR). Retrieved September 25, 2025, from https://www.icuregswe.org/en/
Elm, Erik von, Douglas G. Altman, Matthias Egger, Stuart J. Pocock, Peter C. Gøtzsche, and Jan P. Vandenbroucke. 2014. “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies.” International Journal of Surgery 12 (12): 1495–99. https://doi.org/10.1016/j.ijsu.2014.07.013.
Svenska Intensivvårdsregistret. (n.d.). R & D applications. Retrieved September 25, 2025, from https://www.icuregswe.org/en/research/r--d-applications/
Moreno, Rui P., Philipp G. H. Metnitz, Eduardo Almeida, Barbara Jordan, Peter Bauer, Ricardo Abizanda Campos, Gaetano Iapichino, David Edbrooke, Maurizia Capuzzo, and Jean-Roger Le Gall. 2005. “SAPS 3From Evaluation of the Patient to Evaluation of the Intensive Care Unit. Part 2: Development of a Prognostic Model for Hospital Mortality at ICU Admission.” Intensive Care Medicine 31 (10): 1345–55. https://doi.org/10.1007/s00134-005-2763-5.
Katikireddi SV, et al. Evidence synthesis for constructing directed acyclic graphs (ESC-DAGs): a novel and systematic method for building directed acyclic graphs. Int J Epidemiol. 2019;49(1):322–329. doi:10.1093/ije/dyz150.
Jönsson N, Pettersson N, Asplund P, Bremer A, Lehtipalo S, Hessulf F. Actors associated with treatment limitations in two Swedish intensive care units: prevalence and patient involvement. Acta Anaesthesiol Scand. 2022 Dec 19;67(2):224–232. doi:10.1111/aas.14185
Piscitello GM, Parker WF. Do-Not-Resuscitate Orders by COVID-19 Status Throughout the First Year of the COVID-19 Pandemic. Chest. 2024 Mar;165(3):601-609. doi: 10.1016/j.chest.2023.09.024. Epub 2023 Sep 29. PMID: 37778695; PMCID: PMC10925541.
Connellan D, Diffley K, McCabe J, Cotter A, McGinty T, Sheehan G, Ryan K, Cullen W, Lambert JS, Callaly EL, Kyne L. Documentation of Do-Not-Attempt-Cardiopulmonary-Resuscitation orders amid the COVID-19 pandemic. Age Ageing. 2021 Jun 28;50(4):1048-1051. doi: 10.1093/ageing/afab075. PMID: 33909020; PMCID: PMC8135469.
Sotoodeh A, Hedberg P, Mårtensson J, Nauclér P. Association of hospital and intensive care unit occupancy and non-admission to the intensive care unit decisions: a retrospective cohort study. Intensive Care Med. 2025 Mar;51(3):621-623. doi: 10.1007/s00134-025-07790-8. Epub 2025 Jan 20. Erratum in: Intensive Care Med. 2025 Mar;51(3):660. doi: 10.1007/s00134-025-07817-0. PMID: 39831999; PMCID: PMC12018513.
Subramaniam A, Tiruvoipati R, Pilcher D, Bailey M. Treatment limitations and clinical outcomes in critically ill frail patients with and without COVID-19 pneumonitis. J Am Geriatr Soc. 2023 Jan;71(1):145-156. doi: 10.1111/jgs.18044. Epub 2022 Sep 24. PMID: 36151970; PMCID: PMC9539196.
Codner JA, Archer RH, Lynde GC, Sharma J. Timing is everything: surgical outcomes for SARS-CoV-2 positive patients. World J Surg. 2023;47(2):437-444. doi: 10.1007/s00268-022-06814-4.
﷟HYPERLINK "https://pubmed.ncbi.nlm.nih.gov/36316514/?utm_source=chatgpt.com"
Kim HJ, Ahn E, Oh EJ, Bang SR. Perioperative Coronavirus Disease 2019 Infection and Its Impact on Postoperative Outcomes: Pulmonary Complications and Mortality Based on Korean National Health Insurance Data. J Pers Med. 2025;15(4):157. doi: 10.3390/jpm15040157.
Smith M, Vaughan Sarrazin M, Wang X, Nordby P, Yu M, DeLonay AJ, Jaffery J. Risk from delayed or missed care and non-COVID-19 outcomes for older patients with chronic conditions during the pandemic. J Am Geriatr Soc. 2022;70(5):1314-1324. doi:10.1111/jgs.17722.
﷟HYPERLINK "https://www.researchgate.net/publication/358853265_Risk_from_delayed_or_missed_care_and_non-COVID-19_outcomes_for_older_patients_with_chronic_conditions_during_the_pandemic?utm_source=chatgpt.com"
Tripathy S, Vijayaraghavan BKT, Panigrahi MK, Shetty AP, Haniffa R, Mishra RC, Beane A. Collateral Impact of the COVID-19 Pandemic on Acute Care of Non-COVID Patients: An Internet-based Survey of Critical Care and Emergency Personnel. Indian J Crit Care Med. 2021;25(4):374-381. doi:10.5005/jp-journals-10071-23782.
﷟HYPERLINK "https://pubmed.ncbi.nlm.nih.gov/34045802/?utm_source=chatgpt.com"
Frey A, Tilstra AM, Verhagen MD. Inequalities in healthcare use during the COVID-19 pandemic. Nat Commun. 2024;15(1):1894. doi:10.1038/s41467-024-45720-2.
Syrous AN, Gudnadottir G, Oras J, Ferguson T, Lilja D, Odenstedt Herges H, Larsson E, Block L. End-of-life decision-making in critically ill old patients with and without coronavirus disease 2019. Acta Anaesthesiol Scand. 2024 Jan;68(1):63-70. doi: 10.1111/aas.14326. Epub 2023 Sep 5. PMID: 37670491.
Giabicani, M., Le Terrier, C., Poncet, A. et al. Limitation of life-sustaining therapies in critically ill patients with COVID-19: a descriptive epidemiological investigation from the COVID-ICU study. Crit Care 27, 103 (2023). https://doi.org/10.1186/s13054-023-04349-1.

No competing interests reported.

WiltzSupplementChangesinLifeSustainingTreatmentLimitationsinSwedishICUsDuringtheCOVID19Pandemic251030.docx.pdf

Download PDF

Reviewers agreed at journal
29 Dec, 2025
Reviewers invited by journal
12 Dec, 2025
Editor invited by journal
14 Nov, 2025
Editor assigned by journal
12 Nov, 2025
Submission checks completed at journal
12 Nov, 2025
First submitted to journal
06 Nov, 2025

You are reading this latest preprint version

Changes in Life-Sustaining Treatment Limitations in Swedish ICUs During the COVID-19 Pandemic

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

1 INTRODUCTION

1.1 Background

1.2 Objectives

2 METHODS

2.1 Study design

2.2 Participants

2.3 Data collection and variables

2.4 Exposure and outcomes

2.5 Statistical analysis

3 RESULTS

3.1 Baseline characteristics

3.2 Main results

4 DISCUSSION

4.1 Key results

4.2 Increased likelihood of LLST in the COVID-19 period

4.3 COVID-19 status and the pandemic

4.2 Strengths and Limitations

5 CONCLUSION

Declarations

6.5 Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1