Skip to main content

Table 2 Univariate and multivariate† analysis of risk factors for enterococcal bloodstream infection among febrile children admitted to Muhimbili National Hospital (n = 1828)*

From: Genetic relatedness and risk factor analysis of ampicillin-resistant and high-level gentamicin-resistant enterococci causing bloodstream infections in Tanzanian children

 

Univariate anlysis

Multivariate analyses

    

All patients with clinical information (n = 1603)

All patients with known HIV status (n = 842)

Variable

Enterococci

Others

OR (95% CI) P

OR (95% CI) P

OR (95% CI) P

Neonate

28.2% (11/39)

29.3 (524/1789)

0.9 (0.47-1.92) 0.883

0.4 (0.14-0.98) 0.046

g

Prematurity

9.1 (3/33)

3.2 (51/1570)

3.0 (0.88 - 10.08) 0.079

4.4 (1.06-18.1) 0.041

4.7 (1.19-18.8) 0.019

Male sex

61.5 (24/39)

55.4 (992/1789)

1.3 (0.67 - 2.47) 0.450

e

2.9 (1.03-7.91) 0.043

Hospital acquired infection#

48.7 (19/39)

33.2 (594/1789)

1.9 (1.01 - 3.61) 0.046

2.0 (0.96– 4.00) 0.066

1.8 (0.73-4.29) 0.209

Antibiotic use before blood culture

66.7 (26/39)

57.0 (1020/1789)

1.5 (0.77 – 2.95) 0.231

c

b

Cephalosporin use before blood culture

5.1 (2/39)

4.0% (71/1789)

1.3 (0.31 – 5.53) 0.715

a

c

Diagnoses

     

Malnutrition

21.2 (7/33)

15.4 (241/1570)

1.5 (0.64 - 3.46) 0.360

f

3.7 (1.24-11.28) 0.019

HIV infection

13.0 (3/23)

18.8 (173/919)

0.6 (0.19 – 2.20) 0.486

§

0.5 (0.14-2.04) 0.363§

Urinary tract infection

12.1% (4/33)

11.4% (179/1570)

1.1 (0.37- 3.08) 0.898

b

a

Pneumonia

21.2% (7/33)

20.1% (315/1570)

1.1 (0.46- 2.49) 0.871

d

e

Sepsis with unknown focus

45.5% (15/33)

29.4% (462/1570)

2.0 (1.00- 4.00) 0.050

2.9 (1.36-6.20) 0.006

2.6 (1.06-6.55) 0.037

Congenital heart disease

6.1% (2/33)

2.3% (36/1570)

2.7 (0.63- 11.93) 0.177

g

d

Acute watery diarrhoea

9.1 (3/33)

21.8 (343/1570)

0.4 (0.11 - 1.18) 0.091

0.3 (0.09-1.07) 0.065

f

  1. OR = Odds ratio, 95% CI = 95% confidence interval, *There are varying numbers of patients evaluated for each risk factor because, among the 1828 study persons, we only have information on HIV status for n = 942 and sufficient clinical information for n = 1603. #Hospital acquired = blood culture taken >48 hours after admission Multivariate analysis was performed by stepwise backwards logistic regression with threshold p = 0.2 for removal from the model and p = 0.05 for inclusion, Risk factors were removed in a stepwise fashion in the sequence of ascending letters from “a” to “g”, §HIV was removed from the first logistic regression because the many missing values would affect the model, HIV was locked in the second model although number of observations were reduced.