INTRODUCTION: Sleep disturbances are common in older people with cognitive impairment, potentially contributing to negative outcomes. A core outcome set (COS) is required to reduce heterogeneity in clinical trials and promote the development of high-quality evidence to support clinical management. METHODS: A multi-stage mixed methods study was conducted in accordance with The Core Outcome Set Standards for Development. RESULTS: A systematic review identified 287 sleep outcomes from previous clinical trials. Qualitative interviews ensured the COS was informed by what matters most to people with cognitive impairment and their caregivers. A modified Delphi process identified nine outcomes for the COS: total sleep time, sleep onset latency, wakefulness after sleep onset, number of night-time awakenings, sleep efficiency, and measures of sleep quality, daytime sleepiness, cognition, and mood. DISCUSSION: This COS will support researchers to produce more reliable and coherent trial data to guide the management of sleep disturbances in people with neurodegenerative cognitive impairment. HIGHLIGHTS: Evidence is lacking regarding the treatment of sleep disturbances in people with cognitive impairment. Heterogeneity of reported outcomes in clinical trials limits data synthesis. A qualitative analysis established what matters most to people with cognitive impairment and their caregivers when determining treatment effectiveness. A Delphi panel of experts agreed upon a core outcome set. This core outcome set will improve the reliability and comparability of data from future trials.