Spatial epidemiology of amyotrophic lateral sclerosis in Piedmont and Aosta Valley, Italy: a population-based cluster analysis

Background and purpose: The analysis of the spatial distribution of cases could give important cues on putative environmental causes of a disease. Our aim was to perform a spatial analysis of an amyotrophic lateral sclerosis (ALS) cohort from the Piedmont and Aosta Valley ALS register (PARALS) over a 20-year period.

Methods: The address at the moment of diagnosis was considered for each ALS case. Municipalities' and census divisions' resident populations during the 1995-2014 period were obtained. A cluster analysis was performed adopting both Moran's index and the Kulldorff spatial scan statistic.

Results: A total of 2702 ALS patients were identified. An address was retrieved for 2671 (99%) patients. Moran's index was -0.01 (P value 0.83), thus revealing no clusters. SaTScan identified no statistically significant clusters. When census divisions were considered, Moran's index was 0.13 (P value 0.45); SaTScan revealed one statistically significant small cluster in the province of Alessandria. Here, 0.0099 cases were expected and three cases were observed (relative risk 304.60; 95% confidence interval 109.83-845.88, P value 0.03).

Discussion: Our study showed a substantial homogeneous distribution of ALS cases in Piedmont and Aosta Valley. The population-based setting and the adoption of proper statistical analyses strengthen the validity of our results. Such a finding further suggests the involvement of multiple environmental and genetic factors in ALS pathogenesis.