Multivariable regression showing the relative risk of NEC associated with each factor
| Factor | Level | Sub-HR (95% CI) | P value |
| Epoch | Routine vs pre-probiotics | 0.44* (0.23 to 0.85) | 0.014 |
| Gestational age at birth | <25 weeks | 1.00 (ref) | |
| 25–26 weeks | 0.34** (0.16 to 0.70) | 0.003 | |
| 27–30 weeks | 0.09*** (0.03 to 0.26) | <0.001 | |
| ≥31 weeks | – | – | |
| Milk type† | Mother's milk | 1.00 (ref) | |
| No enteral feed | 0.15** (0.04 to 0.52) | 0.003 | |
| Donor milk | – | – | |
| Formula feed | 0.85 (0.29 to 2.52) | 0.775 | |
| Mixed | 0.82 (0.41 to 1.64) | 0.576 | |
| Birth weight | <1000 g | 1.00 (ref) | |
| 1000–1499 g | 0.71 (0.26 to 1.99) | 0.520 | |
| ≥1500 g | 1.12 (0.22 to 5.84) | 0.891 | |
| Sex | Male (vs female) | 1.05 (0.59 to 1.86) | 0.872 |
| Antenatal steroids | At least one dose (vs never) | 1.10 (0.45 to 2.68) | 0.840 |
| NSAID | None | 1.00 (ref) | |
| Indometacin | 1.21 (0.47 to 3.08) | 0.694 | |
| Ibuprofen | 1.20 (0.41 to 3.56) | 0.738 | |
| PROM | No | 1.00 (ref) | |
| Yes | 0.53 (0.22 to 1.29) | 0.164 | |
| Not recorded | 0.78 (0.33 to 1.88) | 0.583 | |
| Mode of birth | Caesarean (vs vaginal) | 1.82 (0.92 to 3.60) | 0.086 |
There were no cases of NEC among those fed using donor milk, or those born after 31 weeks in either epoch and so these groups are excluded from regression analysis.
*P<0.05, **p<0.01, ***p<0.001,
†Mode of milk feeding between first feed and full feeds or NEC/death if earlier.
NEC, necrotising enterocolitis; NSAID, non-steroidal anti-inflammatory drug; PROM, premature rupture of membranes; Ref., reference group; Sub-HR, sub-hazard ratio.