Effect of changing the stillbirth definition on evaluation of perinatal mortality rates.

The perinatal mortality rate is widely used as a summary statistic for evaluating the effectiveness of perinatal care. Since October, 1992, it has been a legal requirement in England and Wales to register fetal deaths at 24-27 completed weeks of gestation as stillbirths (in addition to those after 28 weeks), thereby altering the definition of perinatal death. In a cohort analysis of all babies born to women resident in Wales during 1993, we assessed whether the revised definition of perinatal mortality rate more appropriately measures effectiveness of care. There were 36,793 births and 313 perinatal deaths (221 stillbirths, 92 early neonatal deaths). At 24-27 weeks' gestation there were 59 (39%) survivors and 93 deaths (52 stillbirths, 36 neonatal deaths [28 early, eight late], and 5 postneonatal deaths). 119 babies had a birthweight below 500 g; one survived and 24 were perinatal deaths. Of the 36 late neonatal deaths all were attributed to perinatally related events. Increased survival of infants at 24-27 weeks' gestation emphasises the importance of including all these infants in the perinatal mortality rate, but it would be a more useful measure of the effectiveness of perinatal care if it excluded babies below 500 g, and included late neonatal deaths.