Confounders other than comorbidity explain survival differences in Danish and Swedish ovarian cancer patients–a comparative cohort study

Abstract

Objective: Danish ovarian cancer (OC) patients have previously been found to have worse prognosis than Swedish patients, and comorbidity has been suggested as a possible explanation for this survival difference. We aimed to investigate the prognostic impact of comorbidity in surgically treated OC patients in Denmark and Sweden. Methods: This comparative cohort study was based on data from 3118 surgically treated OC patients diagnosed in 2012–2015. The Swedish subcohort (n = 1472) was identified through the Swedish National Quality Register of Gynecological Surgery, whereas the Danish subcohort (n = 1646) originated from the Danish Gynecological Cancer Database. The clinical databases have high coverage and similar variables included. Comorbidity was classified according to the Ovarian Cancer Comorbidity Index and overall survival was the primary outcome. Data were analyzed using Kaplan Meier and Cox regression analyses. Multiple imputation was used to handle missing data. Results: We found comparable frequencies of the following comorbidities: Hypertension, diabetes and ‘Any comorbidity’. Arteriosclerotic cardiac disease and chronic pulmonary disease were more common among Swedish patients. Univariable survival analysis revealed a significant better prognosis for Swedish than for Danish patients (HR 0.84 [95% CI 0.74–0.95], p