Survival analysis of geospatial factors in the Irish ALS cohort

Abstract

Variations in environmental risk factors potentially influence incidence and progression in complex multifactorial diseases. Few studies have examined the association of survival in amyotrophic lateral sclerosis (ALS) with environmental geospatial variables. Here we use data from the Irish ALS cohort to perform such an analysis. Geographic data sources were used to generate small area values for four geospatial variables (population density, social deprivation, distance to coast, and distance to ALS multidisciplinary (MDT) clinic) for each ALS case on the Irish ALS register. These were combined with follow-up data and used as covariates in Royston-Parmar regression survival analysis including age of onset, site of onset, diagnostic delay, riluzole prescription and MDT clinic attendance as covariates. One thousand, two hundred and thirty-two patients with median survival of 2.31 years from disease onset were included. After addition of the individual geospatial variables in turn, none of the four variables was found to be associated with survival with a p-value <0.05. The results may reflect the public healthcare system that provides riluzole prescription and access to the MDT to all patients free of charge, and is also congruent with our recent finding that social deprivation is not associated with ALS incidence in Ireland.