KCNQ1, KCNH2, and SCN5A are the most common genes responsible for long QT syndrome and Brugada syndrome. However, the genetic variant frequencies of the three genes in different Chinese disease cohorts are largely unknown. In this study, we analyzed the genetic variants of KCNQ1, KCNH2, and SCN5A in patients from seven cohorts (total N = 11945, including patients clinically suspected to have inherited arrhythmia [n = 122], other cardiovascular diseases [n = 1045], epilepsy [n = 4797], or other diseases [n = 5841], and healthy controls [n = 140]) who had undergone genetic testing. All of these variants were identified via genetic testing by two Chinese companies using the Hi-Seq 2000 platform. A total of 1018 variants (minor allele frequency <0.01) were identified, with 186 (18%), 374 (37%), and 458 (46%) variants in the coding exons of KCNQ1, KCNH2, and SCN5A, respectively. Of these variants, 84% had unknown or uncertain clinical significance. The frequency of identified ClinVar pathological/likely pathological variants was higher for KCNQ1 (13/186, 7.0%) than for KCNH2 (6/374, 1.6%) or SCN5A (10/458, 2.2%), and KCNH2 held the highest number and proportion of radical mutations (30/374, 8%). The prevalence of variants was highest in the inherited arrhythmia cohort (35%) and lowest in the healthy controls (<4%), as expected. Noticeably, the variant prevalence was relatively high in the epilepsy cohort (27%). Finally, only 22 of the 82 variants (26%) identified by both companies had consistent interpretations of pathogenicity between the two companies. Our study demonstrated a comprehensive spectrum of variants in KCNQ1, KCNH2, and SCN5A in a large number of Chinese individuals, including inherited arrhythmia, cardiovascular diseases, and epilepsy. The detailed variant frequency data of each cohort could serve as a valuable reference to facilitate further variant classification by others. We also found that the interpretations of pathogenicity differed greatly among the companies.