Birth weight and perinatal mortality: a comparison of "optimal" birth weight in seven Western European countries.

Background: Previous studies have suggested that a population's entire birth weight distribution may be shifted towards higher or lower birth weights, and that optimal birth weight may be lower in populations with a lower average birth weight. We evaluated this hypothesis for seven western European countries.

Methods: We obtained data on all singleton births (N = 1,372,092) and extended perinatal deaths (stillbirths plus neonatal deaths; N = 7,900) occurring in Finland, Sweden, Norway, Denmark, Scotland, the Netherlands, and Flanders (Belgium) in 1993-1995. We assessed whether countries differed in the mode of their birth weight distribution and in the birth weight associated with the lowest perinatal mortality, and then correlated the two.

Results: Substantial international differences were found in the mode of the birth weight distribution, which ranged between 3384 gm in Flanders and 3628 gm in Finland. The position of the minimum of the perinatal mortality curve also differed considerably, ranging between 3755 gm in Flanders and 4305 gm in Norway. There was a strong relation between the two: for every 100 gm increase in modal birth weight, optimal birth weight was 170 gm higher (95% confidence interval = 104-236 gm).

Conclusions: Our results confirm those of previous studies that compared two populations. To improve the identification of small babies at high risk of perinatal death, population-specific standards for birth weight should be developed and used.