Skip to main content

Table 1 LOS estimates reported in the literature

From: Biased and unbiased estimation of the average length of stay in intensive care units in the Covid-19 pandemic

Reference, journal, date of line publication

Summary of important elements relating to the estimate provided

Estimate provided (ALOS, average length of stay; MLOS, median length of stay; IQR, interquartile range)

Wang et al., Jama, Feb 7 2020 [6]

Monocentre study (Wuhan, China). 138 consecutive patients admitted in the hospital from Jan 1 to Jan 28 2020. Last date of follow-up was Feb 3. At this date, of the 36 patients admitted to the ICU, 11 were still in the ICU, 9 had been discharged to home, 10 had been transferred to the general wards, and 6 had died. Censored data are not considered

Hospital MLOS [IQR] = 10 days [7.0–14.0] for those who were discharged at the date of last follow-up (n = 47 out of the 138 patients of the series, including 25 out the 36 ICU patients)

Yang et al., Lancet Respir Med, Feb 21 2020 [7]

Monocentre (Wuhan, China): 52 patients admitted to the ICU between late December 2019 and January 26 2020. At the date of last follow-up, February 3 2020, there were 32 deaths and 20 survivors among which 12 were not discharged. Censored data are not considered

ICU MLOS in patients who died = 7 days

Guan et al., New Engl J Med, Feb 28 2020 [17]

Multicentre study from China. National data of hospitalized patients between Dec 1 and Jan 29. At the date of last follow-up, January 31, 1029 out of the 1099 patients included in the study were still hospitalized (93%), and 154 out of the 173 patients (89%) categorized as severe were still hospitalized. Censored data are not considered

Hospital MLOS [IQR] of severe patients: 13 days [11.5–17]

Cao et al., Intensive Care Med, Mar 2 2020 [18]

Monocentre (Wuhan, China): 18 patients transferred to the ICU out of 102 patients hospitalized between January 3 and February 1, 2020. At the date of last follow-up February 15, the 102 patients were discharged. Estimation based on all stays (n = 18)

ICU MLOS = 11 days for patients who survived (n = 12), ICU MLOS = 4 days for the patients who died (n = 6)

Zhou et al., Lancet, Mar 9 2020 [8]

Two centers (Wuhan, China), 51 ICU patients admitted at hospital between December 29 2019 and January 31 2020, and discharged before Jan 31 out of 813 patients admitted at hospital before January 31. Censored data are not considered

ICU MLOS [IQR] = 8 days [4,5,6,7,8,9,10,11,12]

IHME COVID-19 team Medrxiv, Mar 30 2020 [13]

Modeling study. Uses values reported by Zhou et al. [8]

ICU MLOS = 8 days

Moghadas et al., PNAS Apr 03 2020 [9]

Modeling study. Authors indicate that the ICU ALOS for survivors was based on ventilation data reported by Chen et al. and Wang et al. [6], while ICU ALOS for patients dying in the ICU is derived from Yang et al. [7] (n.b., 7 days was reported as the MLOS, not as the ALOS)

ICU ALOS = 13.25 days for survivors

ICU ALOS = 7 days for patients dying in the ICU

Grasselli et al., Jama, April 6 2020 [5]

ICU network of 72 hospitals in Lombardy Italy. 1591 admitted to the ICU between February 20 and March 18, 2020. At the date of final follow-up, March 25, 920 patients (58%) were still in the ICU, 256 patients had been discharged alive (16%), and 405 (25%) patients had died, 10 patients with missing data. Censored data are not considered

ICU MLOS = 9 days considering all patients,

ICU MLOS = 10 days for those still in the ICU

ICU MLOS = 8 days for those discharged

ICU MLOS = 7 days for those who died

Weissman et al., Ann Intern Med, April 7 2020 [10]

Modeling study. ICU ALOS from Hill's model (https://alhill.shinyapps.io/COVID19seir/) which source is Zhou et al. [8] (n.b., 8 days was reported as the MLOS, not as the ALOS)

ICU ALOS = 8 days