[The validity of the certification of natimortality: the experience of a local study].

Although several study confirmed the inaccuracy of information in death certificates, very few investigations were performed to assess the validity of stillbirth certificate reporting. The aim of the present study was to evaluate the extent to which the clinical diagnosis of stillbirth cause reported in certificates can be considered sensitive and specific. Eighty-five consecutive stillbirths were independently classified by two neonatologists with certificate information, using the new Wigglesworth classification, and the reliability of the classification process was evaluated. Afterwards we compared the classes made with certificate causes and the ones made with clinical-autoptic data (the "gold standard") and we estimated sensitivity and specificity of certificate cause of stillbirth. The reliability of the classification process was very high. Values of sensitivity were considered unacceptable, mainly in the field of congenital malformations, intra-partum asphyxia and other specific causes. By contrast, specificity was found to be satisfactory. The accuracy of clinical stillbirth diagnosis is not sufficient and good description needs the contribute of post-mortem necropsy.