BackgroundAsthma is a heterogeneous disease with multiple phenotypes that are useful in precision medicine. As the population ages, the elderly asthma (EA, aged ≥ 65 years) population is growing, and EA is now a major health problem worldwide. ObjectiveTo characterize EA and identify its phenotypes. MethodsIn adult patients with asthma (aged ≥ 18 years) who had been diagnosed with having asthma at least 1 year before study enrollment, 1925 were included in the NHOM-Asthma (registered in UMIN-CTR; UMIN000027776), and the data were used for this study, JFGE-Asthma (registered in UMIN-CTR; UMIN000036912). Data from EA and non-EA (NEA) groups were compared, and Ward's minimum-variance hierarchical clustering method and principal component analysis were performed. ResultsEA was characterized by older asthma onset, longer asthma duration and smoking history, more comorbidities, lower pulmonary function, less atopic, lower adherence, and more hospital admissions because of asthma. In contrast, the number of eosinophils, total immunoglobulin E level, oral corticosteroid use, and asthma control questionnaire scores were equivalent between EA and NEA. There were 3 distinct phenotypes in EA, which are as follows: EA1: youngest, late onset, short duration, mild; EA2: early onset, long duration, atopic, low lung function, moderate; and EA3: oldest, eosinophilic, overweight, low lung function, most severe. The classification factors of the EA phenotypes included the age of onset and asthma control questionnaire—6. Similarities were observed between EA and NEA phenotypes after principal component analysis. ConclusionThe EA in Japan may be unique because of the population's high longevity. Characterization of EA phenotypes from the present cohort indicated the need for distinct precision medicine for EA. Trial RegistrationJFGE-Asthma registered in UMIN-CTR (https://www.umin.ac.jp/ctr/); UMIN000036912.