To investigate developments in perinatal care, all fetal and neonatal deaths among those born after at least 24 weeks of gestation at the University Hospital of Tromso, Norway from 1976 to 1989, were subjected to medical audit. A decrease in total mortality rate was found when based on maturity (>= 24 weeks; 19.9-13.4 parts per thousand; p<0.01), and/or birth weight (>= 500 g; 19.2-13.4%; p<0.05). This was mainly due to a decrease in fetal deaths (14.8-6.6 parts per thousand; p<0.0001). Deaths during labor (5.4-1.1 parts per thousand; p<0.001), and deaths before the onset of labour (9.4-5.5 parts per thousand; p<0.05) declined. The neonatal death rate remained virtually constant (5.2-6.8 parts per thousand). The incidence of conditions affecting the placenta and the umbilical cord, causing asphyxia and intra-uterine growth retardation, declined, from 9.2 to 5.0% (p<0.01), as did that caused by immaturity (2.8-1.3 parts per thousand p<0.05). The rates of death caused by cerebral hemorrhage, respiratory distress syndrome, infections, and malformations did not change. There was no significant proportional change in the causes of death from the first to the last period. The rate of fetal death following suboptimal care declined (2.4-0.4 parts per thousand; p<0.01),while the corresponding neonatal death rate remained unchanged (0.9-1.1 parts per thousand). The proportions of both fetal and neonatal deaths occurring after suboptimal care were low (fetal: 16.2, 8.8, and 5.6%; neonatal: 17.1, 23.5, and 16.2%). These differences did not reach statistical significance. The ratio of neonatal to fetal deaths increased from the first (26.1%) to the last period (50.7%; p<0.01), due mainly to a significant shift from fetal to neonatal death among those weighing 500-999 g (19.6-56.0%; p<0.01). It is concluded that prophylactic efforts to avoid preterm birth and IUGR, and further therapeutic efforts in neonatal intensive care are needed to bring down mortality rates at our hospital in the future.