Congenital anomalies in portugal, 1997-2001: A spatiotemporal clustering analysis

Congenital Anomalies (CA) have a relevant impact on affected persons (can seriously restrict their capabilities and autonomy), their families, society at large and they constitute a huge challenge to Public Health (PH) in terms of prevention at all levels, namely regarding environmental factors. Numbers are of concern: in 50%-60% of CA, the precise cause is not known; every year, 2% to 3% of births all over the world are affected by one major CA and about 14% of babies are born with a single minor malformation; one CA is diagnosed in approximately 7.5 % of children under 5 years of age.
This study is based on new cases of CA notified to the National Registry of Congenital Anomalies at the National Institute of Health (made relative to the number of births in the same period and in the Portuguese municipality where the mother lived during gestation), between 1997 and 2001. Conclusions must be cautious, as the degree of undernotification is not known (more than 75% of all births are monitored by the National Registry) and it may be heterogeneous in space or time (RENAC, 2007). Previous studies were undertaken to detect CA spatial and temporal patterns, and spatial clusters were found (Nunes et al., 2007). Spatiotemporal clustering is now an important tool for epidemiologists and PH authorities, as it allows to detect and monitor disease patterns timely, in order to intervene on identified causes. Methodologically, the spatial scan statistic (Kulldorff and Nagarwalla, 1995), used to identify spatiotemporal clusters, has been improved by including two new approaches (Nunes, 2007): definition of window sizes in the cluster scanning processes considering empirical mean spatial semivariograms and an independent and posterior validation of identified clusters - based on geostatistical simulations (Goovaerts, 1997). A better understanding of CA distribution is now achieved.