Hospital organisation, clinical characteristics, and unadjusted outcomes for 2612 very low birthweight (VLBW) or preterm infants admitted to Scottish and Australian neonatal intensive care units in 1993 and 1994
| Scottish tertiary centres | Scottish non-tertiary centres | Australian tertiary centres | p Value | |
| Number of centres | 5 | 3 | 6 | |
| Total VLBW/preterm infants admitted | 827 | 156 | 1629 | 0.00011-151 |
| Number of intensive care cots1-150 | 10 (10, 20) | 4 (4, 6) | 20 (13, 20) | 0.00011-151 |
| Annual volume of VLBW/preterm infants per centre1-150 | 104 (65, 121) | 30 (30, 31) | 297 (255, 422) | 0.00011-151 |
| Gestation (weeks)1-150 | 29 (27, 31) | 29 (27, 31) | 29 (27, 30) | 0.0161-151 |
| Birth weight (g)1-150 | 1180 (900, 1400) | 1171 (880, 1388) | 1166 (908, 1375) | 0.6651-151 |
| Five minute Apgar score1-150 | 9 (7, 9) | 9 (7, 9) | 8 (7, 9) | 0.00011-151 |
| Worst base excess in first 12 hours (mmol/l)1-150 | −5 (−8.45, −2) | −5 (−8.85, −2.6) | −4.05 (−6.7, −1.9) | 0.00011-151 |
| Maximum Fio 2in first 12 hours1-150 | 0.5 (0.25, 0.77) | 0.5 (0.3, 0.8) | 0.45 (0.26, 0.75) | 0.4491-151 |
| Minimum Fio 2 in first 12 hours1-150 | 0.21 (0.21, 0.38) | 0.25 (0.21, 0.4) | 0.25 (0.21, 0.4) | 0.00031-151 |
| CRIB score1-150 | 2 (1,6) (n = 761) | 3 (1,7) (n = 136) | 2 (1,6) (n = 1476) | 0.4101-151 |
| Hospital deaths (%) | 172/775 (22%)1-152 | 33/148 (22%)1-152 | 216/1628 (13%) | < 0.00011-154 |
| Major brain damage in infants who had ultrasound scanning (%) | 87/734 (12%) | 12/151 (8%) | 118/1480 (8%) | < 0.00011-154 |
↵1-150 Median (interquartile range).
↵1-151 From Kruskal-Wallis test for three independent samples.
↵1-152 If all missing cases in Scotland had survived, the hospital mortality for Scottish neonatal intensive care units would be 20.9%. If the missing case in Australia had died, the hospital mortality rate would be 13.3%. The difference between these rates remains statistically significant.
↵1-154 From χ2 test for 2 × 2 table comparing Scottish and Australian infants.