Secular trends of amyotrophic lateral sclerosis: The Piemonte and Valle d’Aosta register

Importance: This study reports the long-term epidemiologic trends of amyotrophic lateral sclerosis (ALS) based on a prospective register. Objective: To examine the 20-year epidemiologic trends of ALS in the Piemonte and Valle d'Aosta regions of Italy. Design, Setting, and Participants: The Piemonte and Valle d'Aosta Register for ALS (PARALS) is an epidemiologic prospective register that covers 2 Italian regions (population of 4 476 931 inhabitants according to the 2011 census) from January 1, 1995, through December 31, 2014. Case ascertainment is based on multiple sources (neurologic departments, hospital discharge archives, and mortality records). Incidence rates are age and sex standardized for the Italian population of the 2011 census. Age-period-cohort (APC) analysis was performed using a Poisson regression model. Main Outcomes and Measures: The primary study outcomes were long-term incidence and prevalence rates of ALS using a prospective design and their determinants. Results: During the study period, a total of 2702 patients (mean [SD] age at onset, 65.7 [11.1] years; 1246 [46.1%] female and 1456 [53.9%] male) received a diagnosis of ALS between 1995 and 2014, corresponding to a crude annual incidence rate of 3.03 per 100 000 population (95% CI, 2.85-3.23) and an adjusted incidence rate of 2.78 per 100 000 population (95% CI, 2.57-2.96). The age-adjusted incidence rate increased in the 2 decades of the study (1995-2004: 2.66; 95% CI, 2.50-2.83; 2005-2014: 2.89; 95% CI, 2.71-3.07; P = .04), mostly in women. The adjusted rate ratio of men to women decreased from 1.27:1 (1995-2004) to 1.17:1 (2005-2014). The analysis of deviance for the APC regression models indicated that the drift variable is relevant in explaining the variation of ALS incidence rates over time in the overall population (change in deviance, 4.6553; P = .03) and in women (change in deviance, 3.8821; P = .05) but not in men (change in deviance, 0.77215; P = .38). A total of 479 patients with ALS were alive and had not undergone tracheostomy at the prevalence day (December 31, 2014), corresponding to a crude prevalence rate of 10.54 per 100 000 population (95% CI, 9.64-11.52). Conclusions and Relevance: During the 1995 to 2014 period, the crude and adjusted incidences of ALS increased in Piemonte and Valle d'Aosta, mostly in women. The APC model revealed that the increase of ALS incidence is attributable to a birth cohort effect in women, with a peak in the 1930 cohort. The different increase of ALS incidence in men and women points to an effect of exogenous factors with a differential effect on the 2 sexes, acting on a genetic background.