The estimated prevalence of insomnia in cancer patients varies between 20% and 50%, which is substantially higher than the general population. To date, little is known about the risk factors for insomnia in patients with cancer. This study examines the prevalence and predictors of insomnia in a population-based sample of women with ovarian cancer. Participants were 772 women participating in the Australian Ovarian Cancer Study - Quality of Life Study. Insomnia was assessed using the Insomnia Severity Index (ISI). Demographic, disease and treatment variables, and psychosocial variables, including anxiety and depression, support care needs and social support and coping, were investigated as potential predictors of insomnia. Twenty-seven percent of women reported sub-clinical symptoms of insomnia (ISI score 8-14) and 17% reported clinically significant insomnia (ISI score 15-28). Three variables were significant predictors of clinically significant insomnia: young age (<50 years: Odds Ratio (OR)=2.36; Confidence Interval (CI) 1.06-5.26; 50-59 years: OR=2.73; CI 1.33-5.64) relative to 70+ years; higher unmet needs in the physical/daily living domain (OR=1.02; CI 1.01-1.03) and elevated anxiety (sub-clinical anxiety (Hospital Anxiety and Depression Scale (HADS) score 8-10): OR=1.83; CI 1.04-3.24; clinical anxiety (HADS score 11-21): OR=2.03; CI 1.08-3.85). In contrast to predictors of primary insomnia, women with cancer aged <60 years were more likely to report clinical levels of insomnia than women of 70+ years. Consistent with primary insomnia, elevated anxiety predicted insomnia in women with ovarian cancer. Given that both anxiety and insomnia are relatively common, and the relationship may potentially be bi-directional, the efficacy of interventions targeting insomnia and anxiety, rather than insomnia alone, is worthy of consideration.