Skip to main content

Table 2 Demographic and clinical characteristics of the study cohort (N=249). Comparator = unanimous method, meaning that the site investigators’ consensus discharge assessment and the individual evaluations by the three external RPD panelists were in complete agreement, regarding the diagnosis of SIRS or sepsis. If the agreement was less than unanimous, then an indeterminate call was made

From: Physician agreement on the diagnosis of sepsis in the intensive care unit: estimation of concordance and analysis of underlying factors in a multicenter cohort

Parameter category Parameter SIRS (n = 106) Sepsis (n = 69) Indeterminate (n = 74) p value1
Demographics Age: median (IQR) 54 (40–65) 60 (47–67) 64 (53–75) 0.002
Sex: female 51 (48%) 30 (44%) 36 (49%) 0.79
White 68 (64%) 41 (59%) 53 (72%) 0.30
Black 30 (28%) 16 (23%) 16 (22%) 0.55
Asian/East Indian 2 (2%) 6 (9%) 2 (3%) 0.06
Hispanic 4 (4%) 5 (7%) 3 (4%) 0.54
Other or unrecorded 2 (2%) 1 (1%) 0 0.51
Blood culture result 1. No blood culture done or blood culture negative 100 (94%) 11 (16%) 66 (89%) < 0.001
2. Blood culture positive 3 (3%) 29 (42%) 4 (5%) < 0.001
3. Gram positive 3 (3%) 13 (19%) 3 (4%) < 0.001
4. Gram negative 0 11 (16%) 1 (1%) < 0.001
5. Mixed Gram pos/neg 0 5 (7%) 0 0.001
6. Fungus 0 0 0 NA
Infection site Respiratory tract (non-lung) 0 3 (4%) 4 (5%) 0.06
Lung (pneumonia) 3 (3%) 15 (22%) 24 (32%) < 0.001
Abdominal 0 10 (14%) 2 (3%) < 0.001
Urinary tract 0 8 (12%) 3 (4%) 0.001
Other site 0 21 (30%) 4 (5%) < 0.001
Not identified 103 (97%) 12 (17%) 37 (50%) < 0.001
Clinical parameters, outcome Days in hospital: median (IQR) 3 (2–5) 8 (5–14)3 6 (4–9) < 0.001
Days in ICU: median (IQR) 2 (1–2) 3 (2–5) 2 (1–4) 0.002
Antibiotics given in ICU 60 (57%) 68 (99%) 68 (92%) < 0.001
APACHE score: median (IQR)2 54 (29–84) 76 (46–95)3 82 (48–103)4 < 0.001
SOFA score: median (IQR) 4 (2–7)6 5 (4–10)5 6 (4–8)7 0.02
Mortality 6 (6%) 9 (13%)3 9 (12%) 0.18
  1. Abbreviations: ANOVA analysis of variance, APACHE Acute Physiology and Chronic Health Evaluation, ICU intensive care unit, IQR inter-quartile range, NC not calculated, neg negative, pos positive, RPD retrospective physician diagnosis, SOFA sequential organ failure assessment
  2. 1For distributions (like age), the p value is derived from ANOVA. For categorical variables such as sex, the p value is derived from a three-sample test for equality of proportions without continuity correction. p values derived from small samples should not be considered definitive.
  3. 2APACHE score, as calculated at different clinical sites (site, version, available to RPD panelists): IMH III yes; LDSH III yes; JHH III no; NH IV no; RUMC II yes; LUMC III no; GMH III no.
  4. 368/69 sepsis patients with data recorded
  5. 473/74 indeterminate patients with data recorded
  6. 555/69 sepsis patients with data recorded
  7. 670/106 SIRS patients with data recorded
  8. 760/74 indeterminate patients with data recorded
Back to article page

Journal of Intensive Care

ISSN: 2052-0492