ObjectiveFriedreich ataxia (FRDA) is an inherited neurodegenerative disease characterized by ataxia and cardiomyopathy. Homozygous GAA trinucleotide repeat expansions in the first intron of FXN occur in 96% of affected individuals and reduce frataxin expression. Remaining individuals are compound heterozygous for a GAA expansion and a FXN point/insertion/deletion mutation. We examined disease‐causing mutations and the impact on frataxin structure/function and clinical outcome in FRDA.MethodsWe compared clinical information from 111 compound heterozygotes and 131 individuals with homozygous expansions. Frataxin mutations were examined using structural modeling, stability analyses and systematic literature review, and categorized into four groups: (1) homozygous expansions, and three compound heterozygote groups; (2) null (no frataxin produced); (3) moderate/strong impact; and (4) minimal impact. Mean age of onset and the presence of cardiomyopathy and diabetes mellitus were compared using regression analyses.ResultsMutations in the hydrophobic core of frataxin affected stability whereas surface residue mutations affected interactions with iron sulfur cluster assembly and heme biosynthetic proteins. The null group of compound heterozygotes had significantly earlier age of onset and increased diabetes mellitus, compared to the homozygous expansion group. There were no significant differences in mean age of onset between homozygotes and the minimal and moderate/strong impact groups.InterpretationIn compound heterozygotes, expression of partially functional mutant frataxin delays age of onset and reduces diabetes mellitus, compared to those with no frataxin expression from the non‐expanded allele. This integrated analysis of categorized frataxin mutations and their correlation with clinical outcome provide a definitive resource for investigating disease pathogenesis in FRDA. Ann Neurol 2016;79:485–495