Regionalization and improvement in transport systems have had a major impact on the care of LBW infants. However, as birth weight increases, the need for intensive care decreases. To identify subgroups of LBW infants likely to benefit from maternal referral, the 594, 104 infants included in the 1990 California birth/infant death cohort file were analyzed according to the level of neonatal care (1993 survey of the California Association of Neonatologists) at the place of birth: level 1 (well baby nursery), level 2 (mechanical ventilation,MV, available only prior to transport), level 2 plus (availability of MV on a long term basis), and level 3 (long term MV and surgical capabilities). A logistic regression analysis using birth weight and discharge diagnoses was done in the 53,229 ill infants to create a model that best predicted mortality. This model was applied to LBW infants born at three different levels of care and the ratios of risk-adjusted observed versus expected mortality (O/E ratio) were calculated by birth weight(table). Overall, there was an effect of both level of care and birth weight on O/E ratio (p<0.05, χ2). O/E ratios were lower for level 3 vs 1 and 2 (p<0.05) and levels 3 and 2 plus vs 1 and 2(p<0.05). These differences were due to a lower O/E ratio at levels 3 and 2 plus for infants ≤ 1250g (p<0.05). Mortality risk in preterm infants with birth weight ≤ 1250g is lower if birth occurs at a service with a level 2 plus or level 3 nursery.

Table 1
Full size table