Table 5 Logistic regression analysis with random effect after propensity score matching.

From: The Influence of the Variation in Sepsis Rate between Neonatal Intensive Care Units on Neonatal Outcomes in Very-Low-Birth-Weight Infants

 

Univariate logistic regression

Multivariate logistic regression adjusted for gestational age and groups according to the mean annual number of very-low-birth-weight infants

IS group versus LS group

HS group versus IS group

HS group versus LS group

IS group versus LS group

HS group versus IS group

HS group versus LS group

Death

1.3 (0.95–1.86)

1.5 (1.11–1.99)

2.0 (1.44–2.71)

1.1 (0.79–1.66)

2.0 (1.39–2.76)

2.2 (1.57–3.19)

Moderate to severe BPD

1.1 (0.90–1.45)

1.5 (1.20–1.91)

1.7 (1.37–2.18)

1.2 (0.90–1.51)

1.7 (1.34–2.24)

2.0 (1.57–2.61)

Severe BPD

1.2 (0.89–1.63)

1.7 (1.29–2.27)

2.1 (1.55–2.75)

1.2 (0.87–1.66)

2.0 (1.45–2.69)

2.4 (1.74–3.24)

Moderate to severe BPD or death

1.2 (0.96–1.47)

1.6 (1.30–1.96)

1.9 (1.54–2.34)

1.2 (0.93–1.51)

1.9 (1.50–2.42)

2.3 (1.78–2.87)

Severe BPD or death

1.2 (0.98–1.59)

1.7 (1.36–2.14)

2.1 (1.69–2.68)

1.2 (0.92–1.59)

2.1 (1.65–2.80)

2.6 (1.99–3.40)

PVL

1.2 (0.84–1.86)

1.6 (1.07–2.27)

1.9 (1.34–2.82)

1.3 (0.85–1.89)

1.5 (1.06–2.26)

2.0 (1.34–2.86)

Survival without major morbidities*

1.1 (0.89–1.34)

0.7 (0.54–0.83)

0.7 (0.59–0.91)

1.1 (0.90–1.38)

0.7 (0.53–0.83)

0.7 (0.59–0.92)

  1. *Major morbidities include moderate to severe BPD, NEC ≥ stage 2, IVH ≥ grade 3, PVL, ROP requiring surgery or VEGF.
  2. HS; high sepsis, IS; intermediate sepsis, LS; low sepsis, BPD; bronchopulmonary dysplasia, PVL; periventricular leukomalacia.
Back to article page