OBJECTIVES: To identify predictors of sleep disturbance in persons with cancer, and compare with predictors of hypnotic use in the same population. METHODS: Data were originally collected in a study of cancer symptoms and practice patterns. Subjects were 3,106 adults (aged ≥18 years) receiving treatment for solid tumors (breast, colorectal, lung, or prostate), and their treating clinicians. Baseline demographic, clinical, and treatment data were collected, along with severity ratings (evaluated by both patients and clinicians) for a broad range of cancer symptoms. Clinicians and patients reevaluated symptoms approximately four weeks later. Scores for sleep disturbance severity (from 0=’not present’ to 10=’as bad as you can imagine’) were only available for 2,379 subjects. Linear regression determined predictors for sleep disturbance severity, while logistic regression determined predictors for hypnotic use (‘yes/no’; benzodiazepines [BZDs] or BZD receptor agonists [BzRAs]). RESULTS: Among 2,379 subjects, 62.8% reported some degree of disturbed sleep, and 23.54% reported hypnotic use (Χ[10]=581.49, p<0001). Significant predictors of sleep disturbance were: age (p<0.001), recent employment change (p<0.001), history of depression (p<0.001), cancer type (p=0.002), functional status (p<0.001), psychological distress (p<0.001), prior number of treatment regimens (p<0.004), and type of pain treatment (p<0.001). Significant predictors of hypnotic use were: race (p<0.001), recent employment change (p<0.001), history of depression (p<0.001), clinical stage of cancer (p<0.001), type of pain treatment (p<0.006), and sleep disturbance severity (p<0.001). Notably, odds of hypnotic use were not higher among patients for whom clinicians determined disturbed sleep to be a priority concern (by impact on quality of life). CONCLUSIONS: Predictors and prevalence of hypnotic use are not entirely consistent with predictors and prevalence of sleep disturbance in persons with cancer, suggesting a need to investigate further the determinants of shared decision-making in this context, including possible racial disparities and the influence of prescribers’ attitudes regarding hypnotic use.