Objective: The aim of this study was to determine whether the distance of residence from a Gynecological Oncology Service (GOS) was associated with a better survival from ovarian cancer.
Methods: We linked cancer registry records to hospital records for 3749 women with ovarian cancer diagnosed between 2000 and 2008 in New South Wales, Australia. Access to a GOS was measured in kilometers from a woman's geocoded address to the geocoded address of the closest public GOS hospital. Flexible parametric survival, Cox proportional hazards, and logistic regression models were fitted to examine whether better access to a GOS was associated with a better survival and whether extensive surgery was received for ovarian cancer after adjustment for patient, tumor, and treatment factors.
Results: Hazard of death from ovarian cancer was greater in women who were treated in a public general hospital than in women treated in a GOS hospital (hazards ratio, 0.77; 95% confidence interval [CI], 0.64-0.95), and greater in those who did not have extensive surgery than in those who did (hazards ratio, 0.47; 95% CI, 0.38-0.58). The further women with ovarian cancer lived from a public GOS hospital, the more likely they were to be treated in a public general hospital. Women were 19 times more likely (odds ratio, 19.40; 95% CI, 13.92-27.04) to be treated only in a general hospital when they lived 187 km or more from a public GOS hospital than women who lived within 5 km of one.
Conclusions: Distance of residence from GOS hospitals in Australia is an important determinant of access to GOS hospitals. Treatment in a public or private GOS hospital and having surgery were the strongest predictors of survival from epithelial ovarian cancer. Research is required into the barriers to referral of patients with ovarian cancer for care in GOS hospitals; low population density limits options for supply of GOS in rural areas.