Skip to main content

Table 1 Fixed effects, random effects, Bayesian random effects, and meta-regression estimates of the effectiveness of chlorhexidine bathing

From: Chlorhexidine bathing and health care-associated infections among adult intensive care patients: a systematic review and meta-analysis

 

Summary estimate IRR (95% CI/CrI)

Meta-regression

Outcome

Fixed effects (I)

Random effects (DL)

Bayesian (RE)

Slope (95% CrI)

BSI

0.79 (0.73, 0.86)a

0.79 (0.60, 1.03)

0.78 (0.45, 1.23)

–0.125 (–0.180, –0.071)a

CLABSI

0.64 (0.56, 0.74)a

0.49 (0.35, 0.68)a

0.44 (0.26, 0.75)a

–0.050 (–0.123, 0.0231)

VAP

0.83 (0.71, 0.98)a

0.85 (0.67, 1.07)

0.82 (0.57, 1.25)

–0.026 (–0.055, 0.002)

CAUTI

1.04 (0.91, 1.18)

0.98 (0.79, 1.21)

0.93 (0.45, 1.66)

–0.008 (–0.048, 0.032)

MRSA-C

0.59 (0.53, 0.65)a

0.58 (0.47, 0.71)a

0.59 (0.36, 0.94)a

–0.010 (–0.0154, 0.005)

MRSA-B

0.66 (0.52, 0.84)a

0.66 (0.52, 0.84)a

0.64 (0.43, 0.91)a

–0.155 (–0.364, 0.054)

VRE-C

0.58 (0.47, 0.73)a

0.52 (0.33, 0.82)a

0.53 (0.20, 1.55)

–0.027 (–0.065, 0.010)

VRE-B

0.45 (0.25, 0.82)a

0.63 (0.19, 2.08)

0.53 (0.15, 2.26)

–1.006 (–1.830, –0.182)a

C-diff

1.04 (0.91, 1.18)

0.98 (0.48, 1.80)

0.93 (0.48, 1.80)

0.008 (–0.048, 0.032)

  1. aEstimates with 95% CrI excluding the null (0 for regression, and 1 for IRR)
  2. IRR incidence rate ratio, I inverse variance, DL DerSimonian-Laird, RE random effects, CI confidence interval from Bayesian models, CrI credible interval from Bayesian models, BSI bloodstream infection, CAUTI catheter-associated urinary tract infections, C-diff Clostridium Difficile, CLABSI central line-associated bloodstream infection, MRSA methicillin-resistant Staphylococcus aureus, MRSA-C MRSA colonisation, MRSA-B MRSA-associated BSI, VAP ventilator-associated pneumonia, VRE vancomycin-resistant Enterococcus, VRE-C VRE colonisation, VRE-B VRE-associated BSI