Low birthweight and mortality: the tendency to repeat low birthweight and its association with early neonatal and infant morbidity and mortality.

Previous studies have demonstrated the tendency to repeat gestational age and birthweight in successive pregnancies and that this tendency is associated with infant survival. Thus, newborn outcome and survival is less favourable if the gestational age and size departs from this maternal tendency. This paper aims to study diseases or conditions that might be associated with this effect. Data were provided through a linkage between three Danish health registries: the Danish Fertility Database, the National Hospital Registry, and the Registry for Preventive Medicine. Such linkage was possible due to the use of unique ID-person numbers. The study included all 8219 second-order low-birthweight (LBW) singleton Danish births, 1980-94, of whom 7811 were liveborn. It was also required that the mother's first delivery took place during that period. The analysis considered 7803 of these births; eight were excluded due to insufficient information. Of the second-order LBW children, 26% had an elder sibling who was also LBW. Early neonatal mortality of a 'non-repeat' LBW birth was 1.3 times higher than 'repeat' LBW births [53.8 vs. 41.2 per 1000; RR 1.31; 95% CI 1.03, 1.65], as was infant mortality [78.4 vs. 60.8 per 1000; RR 1.30; 95% CI 1.06, 1.56]. Also, proportionately more LBW repeat births had Apgar scores of >or=7 after 1 and 5 min. Overall, repeat second-order LBW babies weighed 68 g more than non-repeat LBW babies (P < 0.001). At term, the weight difference was 160 g higher among repeat LBW births (P < 0.001). The mean number of hospitalisations during the first year of life was lower among repeat than non-repeat LBW babies (2.30 vs. 2.46, P < 0.001), while the mean duration of stay was 23.71 vs. 23.97 days (P > 0.05). Newborn immaturity was the most common diagnosis for hospitalisation, and infections the second most common. There were no differences between repeat and non-repeat LBW births in the proportion with each diagnosis. Apart from the differences in birthweight, we were unable to explain the improved survival for repeat compared with non-repeat LBW babies. Except for differences in Apgar scores, we observed no differences in morbidity based on registered hospitalisations during infancy.