Skip to main content

Table 2 Subgroup analysis for meta-analysis of mortality

From: Clinical outcomes and prognostic factors in bloodstream infections due to extended-spectrum β-lactamase-producing Enterobacteriaceae among patients with malignancy: a meta-analysis

Variables

 

No

RR (95% CI)

I2

Pa

Pb

Design

Retrospective cohort

10

1.72 (1.38–2.14)

62.2%

0.005

0.395

 

Prospective cohort

6

2.02 (1.57–2.60)

75.7%

0.001

 

Case–control study

3

3.11(2.47–3.92)

90.2%

 < 0.001

Region

Europe

5

2.79(2.10–3.69)

78.2%

0.001

0.014

 

Asia

12

1.49(1.22–1.82)

27.3%

0.177

 

North America

2

3.47 (2.73–4.41)

93.5%

 < 0.001

Population

Adults and children

7

2.80 (2.29–3.43)

88.3%

 < 0.001

0.303

 

Adults

12

1.80 (1.50–2.15)

28.4%

0.167

Malignancy type

Hematological

7

3.20 (2.54–4.03)

83.0%

 < 0.001

0.355

 

Solid or hematological

12

1.81 (1.54–2.14)

60.9%

0.003

FN

Yes

3

1.48 (0.87–2.53)

0.0%

0.418

0.530

 

No

16

2.21 (1.92–2.54)

80.8%

 < 0.001

ESBL detection methods

Disk diffusion

6

2.21 (1.72–2.83)

64.6%

0.015

0.554

 

MicroScan

4

3.99 (3.10–5.15)

85.9%

 < 0.001

NOS

 < 6

2

0.76 (0.39–1.49)

16.5%

0.274

0.065

 

 ≥ 6

17

2.30 (2.00–2.64)

77.3%

 < 0.001

  1. RR relative risk, CI confidence interval, FN febrile neutropenia, ESBL extended-spectrum β-lactamase; NOS Newcastle–Ottawa scale
  2. ap for heterogeneity within each subgroup
  3. bp for heterogeneity between subgroups with meta-regression analyses