Table 4

Adjusted ORs for outcomes associated with admission to either tertiary or high-volume neonatal care at the hospital of birth using an instrumental variable logistic regression model

OutcomeTertiary neonatal unitHigh-volume neonatal unit†
(1)
≤32+6 weeks
(2)
≤26+6 weeks
(3)
27+0–32+6 weeks
(4)
≤32+6 weeks
(5)
≤26+6 weeks
(6)
27+0–32+6 weeks
Neonatal mortality0.87 (0.66 to 1.15)1.01 (0.63 to 1.61)0.82 (0.58 to 1.14)0.70* (0.53 to 0.92)0.54** (0.33 to 0.87)0.80 (0.56 to 1.13)
Any in-hospital mortality0.85 (0.68 to 1.06)0.95 (0.62 to 1.44)0.84 (0.64 to 1.10)0.68** (0.54 to 0.85)0.51** (0.33 to 0.79)0.80 (0.60 to 1.07)
BPD1.19 (0.95 to 1.49)1.04 (0.66 to 1.64)1.17 (0.91 to 1.51)1.05 (0.85 to 1.29)1.78** (1.12 to 2.81)0.96 (0.75 to 1.22)
Treatment for ROP1.91* (1.16 to 3.14)1.57 (0.83 to 2.96)2.17* (1.06 to 4.47)1.02 (0.60 to 1.73)0.58 (0.29 to 1.15)1.84 (0.83 to 4.05)
Surgery for NEC1.17 (0.72 to 1.90)0.81 (0.40 to 1.66)1.34 (0.76 to 2.38)1.26 (0.76 to 2.07)1.11 (0.54 to 2.28)1.35 (0.75 to 2.43)
PMA at discharge >40+0 weeks0.95 (0.73 to 1.22)0.83 (0.60 to 1.13)0.97 (0.72 to 1.31)0.92 (0.72 to 1.17)1.04 (0.78 to 1.40)0.86 (0.67 to 1.14)
  • Values are ORs (95% CI).

  • Models are adjusted for gestational age, gestational age squared, birthweight z-score, use of antenatal steroids, gender, infant year of birth and deprivation.

  • *p<0.05, **p<0.01, ***p<0.001. †High volume was defined as being in the top quartile of units by number of care days provided to infants born at ≤32+6 weeks gestation.

  • BPD, bronchopulmonary dysplasia; NEC, necrotising enterocolitis; PMA, postmenstrual age; ROP, retinopathy of prematurity.