Background <p>Delirium is a common, serious, and potentially preventable neuropsychiatric syndrome among hospitalized older adults. It is characterized by acute disturbance in attention and awareness with additional cognitive disturbance and a fluctuating course. Dual orexin receptor antagonists (DORAs), including suvorexant, lemborexant, and daridorexant, promote sleep by antagonizing orexin-mediated wakefulness rather than by direct gamma-aminobutyric acid-ergic sedation. Whether DORAs prevent delirium in hospitalized older adults remains uncertain, and prior syntheses have not fully resolved the applicability of this evidence to older inpatients.</p> Methods <p>We will conduct a randomized trial systematic review and meta-analysis, if feasible, evaluating DORAs for delirium prevention in hospitalized adults aged 65&#xa0;years or older. We will search MEDLINE via PubMed, Embase via Ovid, the Cochrane Central Register of Controlled Trials via the Cochrane Library, Web of Science Core Collection, ClinicalTrials.gov, and the WHO International Clinical Trials Registry Platform from inception to the planned final search date, without language or date restrictions. Eligible interventions will include suvorexant, lemborexant, or daridorexant administered before delirium onset. Eligible comparators will include placebo, usual care, no prophylaxis, or clinically interpretable active comparators. The primary outcome will be incident delirium during hospitalization. Secondary outcomes will include delirium severity, duration, subtype, time to onset, sleep-related outcomes, length of stay, mortality, discharge destination, treatment discontinuation, and adverse events. Two reviewers will independently screen records, extract data, assess risk of bias using RoB 2, and judge applicability for the primary quantitative synthesis. Meta-analysis will be performed only when at least two studies are clinically and methodologically sufficiently similar. We will use random-effects models, prespecified sparse-data methods, exploratory trial sequential analysis only if feasible, and GRADE to assess certainty of evidence.</p> Discussion <p>This protocol addresses a focused clinical question while explicitly separating formal review inclusion from eligibility for the primary quantitative synthesis. This distinction is intended to prevent clinically indirect evidence, including mixed-age data, post-delirium treatment, unbalanced combination regimens, or unvalidated delirium ascertainment, from being interpreted as practice-changing evidence for prevention in hospitalized older adults.</p> Systematic review registration <p>PROSPERO CRD420251087526.</p>