The influence of hospital volume on disease outcomes of advanced uterine cervical cancer patients: A nationwide cohort study of Taiwan

[[abstract]]Purpose/Objective(s): To investigate the correlation between hospital volume and clinical outcomes in locally advanced cervical cancer patients treated with curative chemoradiation therapy or radiation therapy alone. Materials/Methods: In this population-based retrospective cohort study, total 8,968 patients diagnosed of FIGO stage I to IVuterine cervical cancer from January 2007 to December 2013 were recorded in the Taiwan National Health Insurance Research Databases. This study examined the women who were treatment-naïve before and received curative radiation therapy. The specific hospital volume was defined as the average of one institution’s annual treated cervical cancer patients from January 2007 to December 2013. Univariate and multivariate analysis for prognostic factors including age, FIGO stage, cancer histology, hospital volume, with or without chemotherapy, with or without brachytherapy, pelvic LN status, tumor size, and overall treatment time were analyzed via Cox regression method. The survival analysis was performed via the Kaplan-Meier method. Results: Among the 8968 patients with FIGO stage I to IVuterine cervical cancer, 3607 patients (40%) were treated with chemoradiation therapy or radiation therapy alone as initial treatment. A total of 2751 patients (76%) were treated at 57 hospitals with curative radiation dose. The median annualized number of patients was 6.9 per year. Stratified by tertile analysis, there were 102 patients treated in 20 hospitals with a volume of 33 cases per year. FIGO stage IIB to IV uterine cervical cancer patients comprised sixty-eight percent (n Z 1880) of the cohort. Poor prognostic factors for overall survival in the locally advanced disease subgroup included old age, non-squamous histology type, high grade disease, larger tumor size, advanced FIGO stage, and treatment in low hospital volume institution. Patients received treatment in centers with hospital volume more than 33 cases per year had better overall survival than those treated in institutions with hospital volume of 10-33 cases (P < 0.001) or <=10 cases (P < 0.001) per year. High volume hospital was associated with receipt of brachy-therapy more compared with low volume hospital (86.3% vs. 33.8%, chi-square test P < 0.02). In the multivariable model accounting for clinical and demographic factors, hospital volume remained significant association with survival (P Z 0.04). Conclusion: For the locally advanced cervical cancer, high volume hospital had better treatment outcomes and frequently brachytherapy per-forming may contribute to the positive effect