Objective: To determine if clustering methods can use a holistic assessment of health-related quality-of-life after bladder cancer diagnosis to predict survival outcomes independent of clinical characteristics. In the United States, an estimated 81,180 cases of bladder cancer will be diagnosed in 2022. We aim to help address the knowledge gap concerning the impact of patient functional status on outcomes., Materials and Methods: This is a cross-sectional, retrospective cohort study of patients in the End Results-Medicare Health Outcomes Survey Registry. Age and 36-Item Short Form Survey (SF-36) responses were used as K-means inputs to identify homogenous clusters of older patients with bladder cancer. We analyzed the association between the identified clusters, patient and disease characteristics, and outcomes. We used Cox proportional hazard regression to compare overall survival., Results: We identified 5 homogenous clusters that exhibited differences in patient characteristics and survival. There was no significant difference in cancer stage or surgery type among the clusters. The Cox proportional hazard regression demonstrated significant associations of cluster with gender, age, education, marital status, smoking status, type of surgery, and cancer stage on overall survival. Cluster independently predicted overall survival., Conclusion: Using unsupervised machine learning, we identified clusters of patients with bladder cancer who had similar mental and physical function scores. Cluster grouping suggests that patients' mental and physical function may not be based on disease or treatment. There are significant survival differences between all clusters, demonstrating that a holistic assessment of patient-reported health-related quality-of-life has the potential to predict survival and possible modifiable risk factors in older patients with bladder cancer., (Copyright © 2023 Elsevier Inc. All rights reserved.)