Background Many maternal characteristics increase the risk for perinatal de
ath. To locate potential sites for intervention, it is important to identif
y these risk factors and examine how much of the excess mortality is explai
ned by infants' low birthweight.
Methods Data on all newborns in Finland born between 1991 and 1993 (N = 199
291, of which 1461 were perinatal deaths) were obtained from the Medical B
irth Register. Logistic regression analysis was used to adjust for backgrou
nd variables, both including and excluding infants' birthweight. The percen
tage reduction in odds ratios after adjustment for infants' birthweight was
used to estimate the contribution of infants' low birthweight to the exces
s mortality.
Results After adjusting confounding factors, increased risk for perinatal d
eath was found for eight maternal characteristics. In the following the inc
reased risk is given as odds ratios and the proportions of the excess morta
lity explained by infants' low birthweight are in parentheses: in-vitro fer
tilization 4.12 (>100%); earlier stillbirth 3.43 (87%); higher maternal age
, from 1.21 to 3.08 (38-99%); maternal diabetes 2.87 (50%); lower socioecon
omic status, from 1.30 to 1.70 (27-44%); smoking during pregnancy 1.45 (>10
0%); single mother 1.44 (50%); first birth 1.36 (75%).
Conclusions Excess mortality due to maternal risk factors occurred mainly t
hrough their tendency to cause low birthweight. However, the excess mortali
ty associated with low socioeconomic status, single motherhood, and diabete
s was mediated by other mechanisms in addition to low birthweight.