This study compared the survival and the risk of heart failure (HF), chronic obstructive pulmonary disease (COPD), diabetes mellitus (DM), hypoglycemia, and renal failure (RF) hospitalizations in geriatric patients exposed to carvedilol or metoprolol. Data sources were Danish administrative registers. Patients aged ≥65 and having HF, COPD, and DM were followed for 1 year from the first β-blocker prescription redemption. Patients' characteristics were used to 1:1 propensity score match carvedilol and metoprolol users. A Cox regression model was used to compute the hazard ratio (HR) of study outcomes. For statistically significant associations, a conditional inference tree was used to assess predictors most associated with the outcome. In total, 1,424 patients were included. No statistically significant differences were observed for survival (HR 0.86; 95% confidence interval [CI] 0.67 to 1.11, p = 0.240) between carvedilol/metoprolol users. The same applied to COPD (HR 0.88; 95% CI 0.75 to 1.05, p = 0.177), DM (HR 0.95; 95% CI 0.82 to 1.10, p = 0.485), hypoglycemia (HR 0.88; 95% CI 0.47 to 1.67, p = 0.707), and RF (HR 1.25; 95% CI 0.93 to 1.69, p = 0.142) hospitalizations. Carvedilol users had a 38% higher hazard then metoprolol users of HF hospitalization during the follow-up period (HR 1.38; 95% CI 1.19 to 1.60, p <0.001). Artificial intelligence identified carvedilol exposure as the most important predictor for HF hospitalization. In conclusion, we found an increased risk of HF hospitalization for carvedilol users with this triad of diseases but no statistically significant differences in survival or risk of COPD, DM, hypoglycemia, and RF hospitalizations.