Skip to main content

Table 2 Association between 21 variables and sepsisa among 551b ambulance patients with suspected infection

From: The predictive value of variables measurable in the ambulance and the development of the Predict Sepsis screening tools: a prospective cohort study

Variable

Category

Crude

Univariable, unadjusted

Multivariable, adjusted

n = 551

P-value

n = 551

Adjusted for all factors that were significant in the univariable analysis,n = 484

nc

% sepsis

OR

95% CI

P-value

OR

95% CI

Keywords related to medical history

 Fever or suspected fever

yesd

417

45.1

0.005

1.8

1.2–2.7

0.98

1.0

0.6–1.7

no

134

31.3

Ref

Ref

Ref

Ref

 Pain

yesd

277

43.0

0.53

1.1

0.8–1.6

no

273

40.3

Ref

Ref

Ref

Ref

 Acute altered mental status

yesd

344

49.4

< 0.001

2.4

1.7–3.5

0.03

1.8

1.1–2.9

no

207

29.0

Ref

Ref

Ref

Ref

 Weakness of the legs

yesd

442

44.6

0.007

1.9

1.2–2.9

0.61

1.2

0.6–2.1

no

109

30.3

Ref

Ref

Ref

Ref

 Breathing difficulties

yesd

299

48.2

0.001

1.8

1.3–2.6

0.24

1.3

0.8–2.2

no

250

34.0

Ref

Ref

Ref

Ref

 Loss of energy

yesd

501

43.1

0.04

2.0

1.0–3.7

0.79

1.1

0.5–2.6

no

50

28.0

Ref

Ref

Ref

Ref

 Gastrointestinal symptoms

yesd

214

52.8

< 0.001

2.1

1.5–2.3.0

0.006

1.9

1.2–2.9

no

336

34.8

Ref

Ref

Ref

Ref

 Risk factors for sepsis

yesd

256

47.7

0.008

1.6

1.1–2.2

0.46

1.2

0.8–1.9

no

293

36.5

Ref

Ref

Ref

Ref

Vital signs

 Respiratory rate > 24 breaths/min

yes

189

56.1

< 0.001

2.4

1.7–3.5

0.53

1.2

0.7–2.0

no

361

34.3

Ref

Ref

Ref

Ref

 Oxygen saturation < 94%

yes

232

55.6

< 0.001

2.8

1.9–3.9

0.08

1.6

1.0–2.5

no

319

31.2

Ref

Ref

Ref

Ref

 Heart rate > 110 beats/min

yes

106

58.1

< 0.001

2.3

1.5–3.5

0.09

1.6

0.9–2.8

no

446

37.8

Ref

Ref

Ref

Ref

 Systolic blood pressure ≤ 100 mmHg

yes

54

68.5

< 0.001

3.4

1.9–6.3

0.001

3.6

1.7–7.6

no

496

38.7

Ref

Ref

Ref

Ref

 Level of consciousness, GCS <15

yes

115

67.0

< 0.001

3.8

2.4–5.8

< 0.001

3.5

2.0–6.2

no

416

35.0

Ref

Ref

Ref

Ref

Temperature, °C

      

< 0.001

  

  ≤ 38.0

yes

235

27.0

Ref

Ref

Ref

Ref

 38.1–38.5

yes

91

41.8

0.01

1.9

1.2–3.2

0.02

2.2

1.1–4.2

  > 38.5

yes

223

56.1

< 0.001

3.4

2.3–5.1

< 0.001

3.3

2.0–5.6

POC-tests

 P-Glucose > 6.5 mmol/L

yes

410

42.7

0.10

1.5

0.9–2.3

no

105

33.7

Ref

Ref

Ref

Ref

P-Lactate, mmol/L

      

0.08

 

  ≤ 2.0

yes

340

36.1

Ref

Ref

Ref

Ref

 

 2.1–4.0

yes

160

46.3

0.03

1.5

1.0–2.2

0.35

1.3

0.8–2.1

  > 4.0

yes

38

76.3

< 0.001

5.7

2.6–12.4

0.03

2.8

1.1–7.3

P-suPAR, ng/mL

      

0.31

  

  < 4.0

yes

184

30.4

Ref

Ref

Ref

Ref

 4.0–7.99

yes

263

41.4

0.02

1.6

1.1–2.4

0.41

1.2

0.8–2.0

  ≥ 8.0

yes

93

64.5

< 0.001

4.2

2.4–7.1

0.13

1.7

0.9–3.4

 P-HBP ≥15.0 ng/mL

yes

235

52.8

< 0.001

2.3

1.6–3.2

0.37

1.2

0.8–2.0

no

290

33.1

Ref

Ref

Ref

Ref

Demographic variables

 Age ≥ 65 years

yes

474

41.6

0.83

0.9

0.6–1.5

no

77

42.9

Ref

Ref

Ref

Ref

 Gender, male

yes

331

43.8

0.23

1.2

0.9–1.8

no

220

38.6

Ref

Ref

Ref

Ref

 Charlson comorbidity score ≥ 5 points

yes

86

55.8

0.004

2.0

1.2–3.1

0.13

1.6

0.9–3.1

no

465

39.1

Ref

Ref

Ref

Ref

  1. OR Odds Ratio, CI Confidence Interval, Ref Reference, GCS Glasgow Coma Scale, suPAR soluble urokinase Plasminogen Activating Receptor, HBP Heparin Binding Protein
  2. aSepsis is defined in accordance with Sepsis-3 as a) infection + ≥2 SOFA criteria, or b) infection + vasopressor need and lactate > 2 (septic shock)
  3. bOf total 553 patients with infection according to clinical judgment by ambulance personnel, 551 patients had the required documentation to determine whether the patient had sepsis or not. These 551 patients were included in the regression analysis
  4. cof patients with documentation of the variable
  5. dPatients that were not able to answer yes or no were included in the yes-category in the logistic regression and classification tree analyses (11–30 patients/keyword), based on similarity in prevalence of sepsis and overlapping CIs in these groups
  6. Significant P-values are bolded