Trends in ICU mortality and underlying risk over three decades among mechanically ventilated patients. A group level analysis of cohorts from infection prevention studies

Hurley, James C.

doi:10.1186/s13613-023-01159-0

Annals of Intensive Care

Table 1 Characteristics of studies

From: Trends in ICU mortality and underlying risk over three decades among mechanically ventilated patients. A group level analysis of cohorts from infection prevention studies

Characteristics	Observational	Non- decontamination	Decontamination
Study characteristics
Listing	Additional file 1: Table S1	Additional file 1: Table S2	Additional file 1: Table S3
Number of studies^a	63	90	57
MV for > 48 h for < 90%^b	9	6	8
PPAP for control groups^c	0	0	5
Trauma ICUs^d	5	12	10
Late mortality census^e	27	22	11
North American ICU	27	36	8
Study publication year (range)	1988–2021	1987–2021	1988–2021
Group characteristics
Numbers of patients per study group; median (IQR)^f	296 103–759	69 41–140	75 32–117
Mean patient LOS per study group; mean 95% CI^g	12.2 10.8– 13.4	11.7 10.3–13.2	14.0 12.1–16.0
Mean patient age per study group; mean 95% CI^h,i	58.2 56.4–60.1	55 52.8–56.4	52.2 49.6—55.3
Mean patient APACHE II per study group; mean 95% CI^j	18.9 18.3–19.9	19.6 18.8–20.4	18.7 17.5–19.9
ICU Mortality incidence per 100 patients (mean 95% CI, n)^k
Observational	22.0 19.5–24.8 (41)
Concurrent control groups		22.4 20.5–24.4 (84)	27.0^l^,m,n,o 22.9–31.5 (34)
Intervention groups		22.6 20.9–24.5 (80)	23.6^p,q,r 21.1–26.4 (56)

^aNote, 39 studies had more than one observational, control or intervention group and four studies provided both concurrent control and non-concurrent control groups. Hence, the number of groups does not equal the number of studies. Any non-concurrent control groups are included as observational groups in the analysis, except where otherwise stated
^bStudies for which less than 90% of patients were reported to receive > 48 h of MV
^cPPAP was used within five control and 36 intervention groups
^dTrauma ICU arbitrarily defined as an ICU with more than 50% of admissions for trauma
^eLate mortality is either hospital or beyond day 21 mortality census versus ICU mortality census
^fData are median and inter-quartile range (IQR)
^gLength of stay was similar for the three categories of groups; (p = 0.16; one-way ANOVA)
^hGroup mean age was not available for 12 studies
ⁱThe group mean age for the groups from non-decontamination and decontamination studies were similar but differed from the groups from the observational studies (p = 0.002; one-way ANOVA)
^jThe group mean APACHE II score for the groups from non- decontamination, decontamination and the observational studies were similar (p = 0.40; one-way ANOVA)
^kBartlett’s test for equality of variances, (Chi-square, df 4; 37.14) p = 0.001
^lICU Mortality incidence-non-concurrent control groups of eight studies as in Table S3; 23.2; 18.5–28.7 (9)
^mICU Mortality incidence-including on treatment data for control groups of eight studies as in Table S4; 26.5; 22.5–31.0 (34)
ⁿICU Mortality incidence-control groups of studies of studies of antiseptic interventions; 25.7; 18.1–35.1 (9)
^oICU Mortality incidence-control groups of studies of topical antibiotic interventions; 27.5; 22.8–32.8 (24)
^pICU Mortality incidence–including on treatment data for intervention groups eight studies as in Table S4; 23.3; 20.8–25.9 (56)
^qICU Mortality incidence-antiseptic intervention groups; 25.5; 21.0–30.7 (18)
^rICU Mortality incidence-topical antibiotic intervention groups; 22.7; 19.8–26.0 (37)

Back to article page