Aim: The aim of this paper is to determine the risk factors of perinatal mortality in Lithuania during 1997-1998.
Methods: The focus of the study is the 1997-1998 Medical Birth Register Database (N = 75,178) covering social-demographic factors, the obstetric anamnesis of previous pregnancies, as well as mothers health behaviour and pathology of the current pregnancy, delivery and obstetric aid. The multivariate analysis assessing odds ratios of risk factors for perinatal mortality using logistic regression has been carried out and related population attributable risk fractions were calculated.
Results: A statistically significant (p < 0.01) model predicting perinatal mortality has been obtained in which 19 out of 76 analysed factors have been distinguished to be statistically significant. The septic state of the mother during the delivery (OR = 13.3; CI = 8.9-19.9) and haemorrhage due to a premature placental separation (OR = 11.6; CI = 7.9-16.9) were associated with substantial increased risks of the foetus. Factors from other groups (perinatal pathology during previous pregnancies and social, behavioural or environmental factors) had somewhat lower, though significant, risk on perinatal mortality. The highest population attributable risk fraction (PARF) was found for the breech birth and foetal growth retardation (PARF = 13 percent and PARF = 9.7 percent, respectively). It should be pointed out that the impact of marital status on perinatal mortality was very high (PARF = 6.4 percent), even higher than that of some clinical factors, though its odds ratio was among the lowest (OR = 1.5; CI = 1.2-1.7).
Conclusion: During 1997-1998, the highest risk for perinatal mortality in Lithuania was associated with obstetric aid during delivery and pathology of current pregnancy. These groups of risk factors had also the highest population attributable risk fraction.