Genetic variation in the neurofilament heavy chain gene (NEFH) has been convincingly linked to the pathogenesis of multiple neurodegenerative diseases, however, the relationship between NEFH mutations and ALS susceptibility has not been robustly explored. We therefore wanted to determine if genetic variants in NEFH modify ALS risk. We performed fixed and random effects model meta-analysis of published case-control studies reporting NEFH variant frequencies using next-generation sequencing, microarray or PCR-based approaches. Comprehensive screening and rare variant burden analysis of NEFH variation in the Project MinE ALS whole-genome sequencing data set was also conducted. We identified 12 case-control studies that reported NEFH variant frequencies, for a total of 9,496 samples (4,527 ALS cases and 4,969 controls). Fixed effects meta-analysis found that rare (MAF<1%) missense variants in the tail domain of NEFH increase ALS risk (OR 4.56, 95% CI 2.13-9.72, p<0.0001). A total of 591 rare NEFH variants, mostly novel (78.2%), were found in the Project MinE dataset (8,903 samples: 6,469 cases and 2,434 controls). Burden analysis showed ultra-rare (MAF <0.1%) pathogenic missense variants in the tail domain are associated with ALS (OR 1.94, 95% CI 0.86-4.37, Madsen-Browning p=0.039), replicating and confirming the meta-analysis finding. High-frequency rare (MAF 0.1-1%) tail in-frame deletions also confer susceptibility to ALS (OR 1.18, 95% CI 0.67-2.07, SKAT-O p=0.03), which supports previous findings. This study shows that NEFH tail domain variants are a risk factor of ALS and supports the inclusion of missense and in-frame deletion NEFH variants in ALS genetic screening panels.