In this issue of JAMA Internal Medicine, Mitchell and colleagues1 present their findings from a randomized clinical trial of a video-assisted decision support tool to improve advance care planning in nursing home residents with advanced dementia. They performed a cluster randomized trial of 402 residents with advanced dementia in the setting of 64 Boston-area nursing homes. Proxies of dyads in the intervention arm watched a brief video depicting scenarios for treatment options within 3 care levels—comfort, basic, and intensive. Preferences for care were then elicited and communicated to the patient’s health care team. The investigators examined the relationship of the video intervention to the primary outcome of do-not-hospitalize (DNH) orders, with secondary outcomes including preference for comfort care, documented directives to withhold tube-feeding and intravenous hydration, documented goals-of-care discussions, and burdensome treatments (hospital transfers, tube-feeding, or parenteral therapy). The intervention did not increase DNH directive completion, change care preferences, increase directives to withhold parenteral hydration, or reduce use of burdensome treatments. Residents in the intervention arm were more likely to have directives to withhold tube-feeding.