Genetic diagnosis has become increasingly important to guide clinical decision-making for patients with dilated cardiomyopathy (DCM). Pathogenic or likely pathogenic (P/LP) missense variants in the gene RBM20 cause a highly penetrant arrhythmogenic DCM, but the role of RBM20 truncating variants (RBM20tvs) is unclear. To assess the contribution of RBM20 variants to arrhythmogenic DCM. In this cohort study, participants in the genome-first UK Biobank (UKB) and All of Us populations were evaluated to assess the etiologic fraction, natural history and penetrance of RBM20 variants. Retrospective data were collected from an international cohort of patients with DCM and RBM20 variants identified at centers of excellence for genetic heart disease and compared based on time to event. Study dates are not disclosed because the institutional review board did not authorize the sharing of this information. RBM20 variants were compared to known P/LP variants and variants of uncertain significance in RBM20 as well as titin truncating variants (TTNtvs). Major ventricular arrhythmias, end-stage heart failure, and heart failure hospitalization as measured by medical record review (retrospective cohort) and diagnostic codes (UKB). Two main cohorts were studied for this project. In UK Biobank, a cohort of participants with RBM20tvs, RBM20 synonymous variants, and TTNtvs was studied. Of these 4249 participants, 1869 (44%) were male. The mean (SD) age at enrollment was 56 (8.2) years. In the RBM20 registry, of 179 patients, 105 (58.6%) were male, and the mean (SD) age at enrollment was 43.8 (19.1) years. A validation cohort from the All of Us biobank was also used. This consisted of 7002 participants, 4342 of whom (62.0%) were male, and the mean (SD) age was 52.7 (16.7) years. The etiologic fraction of RBM20 variants in arrhythmogenic DCM was 0.53 (95% CI, 0.32-0.67; P < .001). In genome-first biobanks, lifetime incidence of cardiomyopathy, heart failure, or major ventricular arrhythmia diagnosis was lower in participants with RBM20 variants than in those with TTNtvs (hazard ratio, 0.55; 95% CI, 0.36-0.84; P < .001). Patients with RBM20tvs and DCM presented to referral centers later in life than those with P/LP RBM20 and DCM (mean [SD], 53 [10] vs 34 [18] years; P < .001) and were less likely to have a family history of sudden cardiac arrest (2 of 10 [20%] vs 11 of 17 [65%]; P = .046) or cardiomyopathy (2 of 10 [20%] vs 14 of 18 [78%]; P < .001). There was no significant difference in age- and sex-adjusted incident major heart failure or arrhythmia events between patients with RBM20tv and DCM or those with P/LP RBM20 and DCM, though sex-adjusted lifetime hazard was reduced in those with RBM20tv and DCM (hazard ratio, 0.13; 95% CI, 0.03-0.56; P = .01). This study found that RBM20 variants contributed to arrhythmogenic DCM phenotypes but conferred reduced lifetime disease penetrance compared to TTNtvs and milder disease severity alone than P/LP RBM20 variants. Their potential for additive interactions with other damaging variants should be considered in patients with DCM and their families.