Objectives: To analyze secular changes in the rates of death and of major impaired outcome in surviving outborn infants who weighted < or = 800 gm at birth and were admitted in 1980 to 1989, with adjustment for changes in risk factors and severity of illness around the time of birth; and to identify changes in these factors that might explain changes in outcomes.
Design: Retrospective cohort study with follow-up to a minimum of 18 months of postterm age. After preliminary screening, multivariate models of association between risk/severity of illness factors and outcomes were constructed, validated, and used to adjust outcomes (death and major impairment to 18 to 24 months of age).
Setting: Regional neonatal intensive care unit for referral of "outborn" infants.
Patients: Two hundred eighty-seven consecutively admitted infants who weighted < or = 800 gm at birth (97% follow-up).
Results: The death rate during the 1980s did not fall significantly (p adjusted for risk factors = 0.115). The major impairment rate fell (odds ratio, 0.24 (95% confidence interval, 0.10, 0.60); p = 0.002, adjusted for delivery route and respiratory failure measures), mainly because of a reduced rate of blindness, not attributable to cryotherapy. The risk factors that improved and were possibly related to the reduced impairment rate were blood pH and glucose concentration, and serum sodium concentration in the first 48 hours of life.
Conclusions: Despite an increasing selection for referral of less mature and more severely ill outborn babies near the "limit of viability," and despite more aggressive care, the rate of major impairment fell significantly during the 1980s. This trend was enhanced by adjustment for severity of illness. The fall was attributable to a reduced rate of blindness, and was associated with evidence of improved control of physiologic balance after birth.