Background: Androgen deprivation therapy (ADT) is the cornerstone of treatment for mCSPC patients. Clinical trials have recently shown that adding docetaxel (DOC) or novel hormonal therapies (NHT) to ADT improves survival in this population. Additionally, long-term (≥3 months) treatment with non-steroidal anti-androgens (AA) has been used to achieve combined androgen blockade. However, real-world evidence on survival outcomes in this setting is limited. Methods: Veterans Health Administration (VHA) claims data was used to compare time to metastatic castration-resistant prostate cancer (mCRPC) and overall survival (OS) in mCSPC men who initiated ADT alone or ADT with AA, DOC, or abiraterone (ABI) from April 1, 2014 to March 31, 2018. Propensity scores were calculated using logistics regression and adjusted for demographics (age, race) and patient characteristics at index (index year to ADT, radiation therapy, chronic corticosteroid use, Charlson Comorbidity Index score, select comorbidities, log(PSA), hemoglobin, alkaline phosphate, site of metastasis). Cox proportional hazard (CPH) models were used to evaluate time to progression to (mCRPC) and OS using stabilized inverse probability of treatment weighting (SIPTW) adjusting for body mass index, site of metastasis, time from metastatic diagnosis to index date, and log(PSA). Sensitivity analyses with CPH were conducted using propensity score matching (PSM). Results: The study included 1395 mCSPC men of which 874 (63%) were ADT only, 338 (24%) were ADT+AA (98% with bicalutamide), 75 (5%) were ADT+ABI, and 108 (8%) were ADT+DOC patients. Due to small sample sizes, adjusted survival analyses were not conducted on ADT+ABI and ADT+DOC cohorts. Post-SIPTW adjustment, patient characteristics were balanced between ADT+AA and ADT only cohorts. After treatment initiation, ADT+AA patients showed similar risk of progression to mCRPC (hazard ratio [HR]: 1.05 (95% CI: 0.87–1.26, P = .622), with a median time to mCRPC equal to 21.74 vs 22.5 months compared with ADT only. Patients treated with ADT+AA had similar risk of death compared to ADT only (HR: 1.22; 95% CI: 0.97–1.54, P = .093), with median survival not reached in both cohorts. Sensitivity analyses with PSM confirmed the results. Conclusions: Our study highlighted that most mCSPC patients in the VHA who initiated treatment between 2014 and 2018 were treated with ADT only despite the emergence of DOC or NHT combination therapies. Patients treated with ADT+AA had similar time to mCRPC and OS compared with ADT alone, but results should be treated with care as we could not adjust for all prognostic factors. Median time to mCRPC and OS in our study for both ADT and ADT+AA was similar to published data from corresponding placebo arms in NHT trials but inferior to outcomes in the treated arms of these trials, indicating significant room for survival improvement in the mCSPC population.