Skip to main content
. 2021 Jan 15;9(4):407–418. doi: 10.1016/S2213-2600(20)30560-9

Table 2.

Intensive care admission, need for respiratory support, ICU mortality, and in-hospital mortality among patients with a defined hospital outcome

Brazil (n=232 036) North (n=13 496) Northeast (n=45 238) Central-West (n=17 012) Southeast (n=131 556) South (n=24 734)
ICU admission and mortality
ICU admission (n=205 493) 79 687 (39%) 3786 (32%) 14 867 (43%) 6682 (42%) 45 224 (38%) 9128 (38%)
ICU mortality* 23 780/43 582 (55%) 2037/2569 (79%) 4834/7357 (66%) 1753/3447 (51%) 11 058/22 472 (49%) 4098/7737 (53%)
Respiratory support (n=196 248)
None 54 314 (28%) 3047 (28%) 8177 (25%) 4076 (27%) 32 756 (29%) 6258 (27%)
Yes, non-invasive 96 729 (49%) 4743 (43%) 14 485 (44%) 7561 (49%) 58 444 (51%) 11 496 (50%)
Yes, invasive 45 205 (23%) 3155 (29%) 10 322 (31%) 3667 (24%) 22 648 (20%) 5413 (23%)
Place of non-invasive respiratory support(n=91 816)
In ICU 27 236 (30%) 695 (15%) 3899 (29%) 2359 (32%) 16 930 (31%) 3353 (30%)
Outside ICU 64 580 (70%) 3889 (85%) 9675 (71%) 4904 (68%) 38 138 (69%) 7974 (70%)
Place of invasive respiratory support(n=44 055)
In ICU 38 079 (86%) 2577 (83%) 8415 (84%) 2970 (83%) 19 160 (87%) 4957 (92%)
Outside ICU 5976 (14%) 516 (17%) 1545 (16%) 629 (17%) 2855 (13%) 431 (8%)
Hospital mortality
In-hospital mortality 87 515 (38%) 6727 (50%) 21 858 (48%) 5964 (35%) 45 269 (34%) 7697 (31%)
Median duration of stay, days (IQR)
ICU (n=43 680) 7 (3–15) 6 (3–12) 7 (3–13) 7 (3–13) 7 (3–14) 9 (4–17)
Hospital (n=218 281) 8 (4–14) 7 (4–14) 8 (4–16) 8 (4–14) 8 (4–14) 8 (4–15)

Data are n (%) and represent the available data for each variable. ICU=intensive care unit.

*

ICU mortality was derived for patients with date of ICU discharge equal to the date of hospital death, so it was available for patients without missing values on both dates (n=43 582).

The sum of non-invasive and invasive respiratory support when stratified by place (in ICU and outside ICU) does not match the total respiratory support type because of missing values on the variable ICU admission.