THE SARDINIAN EXPERIENCE OF THE LOWEST ITALIAN INFANT MORTALITY AT THE TURN OF THE TWENTIETH CENTURY. TRUE OR FALSE EMPIRICAL EVIDENCE?

According to the Italian statistical official sources, in the decades after national unification (1861), Sardinia was the region with the lowest neonatal and infant mortality rate in the country and a stillbirth rate significantly lower than the national average. The principal aim of this article is to reconsider the evolution of infant mortality in the island taking into account the general process of improving registration coverage of those vital events (stillbirths and early neonatal deaths) whose recording was crucial for the correct measurement of infant mortality. The official statistics of the newly formed Kingdom of Italy were affected by imperfections and biases, particularly in the decades under examination. In the various regions the recording procedures were so differentiated that they could even distort the geographical as well as the temporal trend in stillbirth and neonatal mortality rates. In Italy, as well as in many other countries, the topic of stillbirth identification and recording procedures has received up to now little attention. A detailed analysis of a large and reach data set reconstructed for the north western Sardinian town of Alghero, at the individual nominative level, derived from civil and parish records of birth, death and marriage, allows us to show that official data – at least for that community –, underestimate infant mortality, particularly early neonatal mortality. This result calls into question the Sardinian record of the lowest Italian infant mortality. Further controls in other communities are necessary to verify a hypothesis so subverting. In the final section of this article we carry out a multivariate analysis on the determinants of early life mortality in Alghero. The results confirm the overwhelmingly predominance of bio-demographic variables, while the socio-economic effect appears to have been modest if not absent.