Author/Editor		Verdenik, Ivan
Title		Večplastni napovedni model prezgodnjega poroda
Translated title		Multilayer prediction model for preterm delivery
Type		monografija
Place		Ljubljana
Publisher		Medicinska fakulteta
Publication year		2002
Volume		str. 70
Language		slo
Abstract		Background: In obstetrics preterm labour represents one of the highest risk to the developing fetus. Preterm labour is responsible for almost 70% of overall neonatal deaths, and for over 85% of neonatal deaths, if deaths associated with congenital malformations are excluded. Surviving infants are at increased risk of cerebral palsy, hydrocephalus, blindness, deafness, respiratory tract diseases and other complications that require neonatal intensive therapy. Thus the expenses of intensive care for preterm infants are extremely high. The aim of this study was to provide the basis for the earliest possible identification of pregnant women at increased risk of preterm labour as to ensure timely application of preventive measures. Methods. The analysis was based on the data from the National Perinatal Information System regarding the entire Slovene singleton population for the period of eight years (n=148,791). Beside some general information of the mother, the system provides the data on her medical history, course of pregnancy and labour, and the data on the newborn. Complementarily we analysed the results obtained by the study EUROPOP providing the detailed data on social and work conditions in pregnancy (1,700 singleton deliveries) as well as the results of the study on congenital uterine malformations as a risk factor for preterm labour (n = 508). The effect of risk factors was assessed using multiple regression analysis. More models were built to assess the differences in the structure of risk factors between more (up to 32nd week) and less preterm labour (up to 36th week), as well as the differences between primiparas and multiparas. On the basis of a regression model we built a predictive model for the risk for preterm labour; i.e. we built two separate models, one to be applicable at the first antenatal visit, and another to predict the risk taking into account the factors occurring during pregnancy. (Abstract truncated at 2000 characters).
Descriptors		LABOR, PREMATURE PREDICTIVE VALUE OF TESTS HOSPITAL INFORMATION SYSTEMS PERINATOLOGY RISK FACTORS LOGISTIC MODELS BAYES THEOREM DIABETES MELLITUS CONIZATION UTERUS HYPERTENSION SMOKING KIDNEY DISEASES FERTILIZATION IN VITRO