Abstract: Objective: To describe ICU patients’ self-report assessment of sleep, examine the relationship between patients' self-reported sleep and their reported sleep by the bedside nurse, and describe the strategies suggested by patients to promote sleep. Methods: An exploratory descriptive study was undertaken with communicative adult patients consecutively recruited in 2014–2015. Patients reported sleep using the Richards–Campbell Sleep Questionnaire (score range 0-100 mm; higher score indicates better sleep quality), with nursing assessment of sleep documented across a five level ordinal variable. Patients were asked daily to describe strategies that helped or hindered their sleep. Ethical approval for the study was gained. Descriptive statistical analysis was performed [ median (interquartile range) ]; relationships were tested using Spearman's rank correlation and differences assessed using the Kruskal–Wallis test; P<0.05 was considered significant. Results: Participants (n=151) were recruited [age: 60 (46-71) years; ICU length of stay 4 (2-9) days] with 356 self-reports of sleep. Median perceived sleep quality was 46 (26-65) mm. A moderate relationship existed between patients’ self-assessment and nurses’ assessment of sleep (Spearman's rank correlation coefficient 0.39-0.50; P<0.001). Strategies identified by patients to improve sleep included adequate pain relief and sedative medication, a peaceful and comfortable environment and physical interventions, e.g. clustering care, ear plugs. Conclusion: Patients reported on their sleep a median of 2 (1-3) days during their ICU stay, suggesting that routine use of self-report was feasible. These reports revealed low sleep quality. Patients reported multiple facilitators and barriers for sleep, with environmental and patient comfort factors being most common. Interventions that target these factors to improve patient sleep should be implemented.