This prospective observational cohort study evaluates the effectiveness and safety of two benzodiazepine sedatives-remimazolam and midazolam-used for general anesthesia induction in elderly patients (≥65 years) undergoing elective non-cardiac surgery. Elderly patients have reduced organ reserve and altered pharmacokinetics/pharmacodynamics, increasing their susceptibility to anesthesia-related complications such as intraoperative hypotension, delayed emergence, and postoperative delirium. Remimazolam, an ultra-short-acting benzodiazepine metabolized by organ-independent tissue esterases, is hypothesized to offer greater hemodynamic stability, faster recovery, and a more favorable safety profile than midazolam in this vulnerable population. Investigators prospectively observe and collect data without altering routine clinical care; the choice of sedative is made by the attending anesthesiologist according to standard practice. Propensity score methods will be used to reduce confounding by indication.