The purpose of this study was to investigate the impact of hospital type determined at primary treatment and find possible predictors of survival in a cohort of patients with advanced epithelial ovarian cancer (EOC) who recurred twice and received three lines of treatment during eight-year follow-up. Using the Norwegian Cancer Registry, the authors identified 174 women with FIGO Stage IIIC EOC diagnosed in 2002. First-line treatment consisted of up-front debulking surgery and chemotherapy, received in either a teaching hospital (TH, n = 84) or a non-teaching hospital (NTH, n = 90). After recurrence all patients in Norway are equally consulted at TH. Survival determined for three time intervals (TI): TI-1, from end date of first-line treatment to first recurrence or death, TI-2, from beginning of second-line treatment until second recurrence or death, and TI-3, from beginning of third-line treatment to death or end of follow-up. Extensive surgery carried out in TH followed by at least six cycles of platinol-taxan chemotherapy resulted in longer survival in the TH group during TI-1. Altogether, the majority of those who receive treatment for recurrences were primary better debulked with following platinol-taxane chemotherapy. Survival in TI-2 was influenced by platinol-sensitivity. During TI-3 the majority (96%) had good performance status and their mean age at primary diagnosis at either hospital type was 57 years. Extensive primary surgery at TH, platinol sensitivity, age, and performance status were predictors of survival in this cohort.